Laboratory Techniques in

Thrombosis - a Manual
Laboratory Techniques in
Thrombosis - a Manual
Second revised edition of the ECAT
Assay Procedures

Edited by
J. Jespersen
Department of Clinical Biochemistry,
South Jutland University,
Ribe County Hospital, Esbjerg, Denmark

R. M. Bertina
Haemostasis and Thrombosis Research Centre,
Leiden University Medical Centre,
Leiden, The Netherlands

F. Haverkate
ECAT Foundation, Oegstgeest,
The Netherlands

__ SPRINGER-SCIENCE+BUSINESS MEDIA, B.V.

Library of Congress Cataloging-in-Publication Data is available.

ISBN 978-0-7923-6472-6 ISBN 978-94-011-4722-4 (eBook)

DOI 10.1007/978-94-011-4722-4

Printed on acid-free paper

All Rights Reserved

© 1999 Springer Science+Business Media Dordrecht
Originally published by Kluwer Academic Publishers in 1999
Softcover reprint ofthe hardcover 2nd edition 1999
No part of this publication may be reproduced or utilized in any form or by any means, electronic,
mechanicaI, inc1uding photocopying, recording or by any information storage and retrieval system,
without written permission from the copyright owner.
Contents

Preface to the Second Edition ix

Preface to the First Edition Xl

List of Contributors xiii

Nomenclature of haemostasis factors xvii

Introduction to laboratory assays in haemostasis and

thrombosis
1. Gram and 1. Jespersen

2 Good medical laboratory services - guidelines 9

J.-c Libeer
3 Blood collection and sample preparation: pre-analytical
variation 21
I D. Walker

4 Quality assessment of haemostatic assays and external

quality assessment schemes 29
T. A. L. Woods, S. Kitchen and F E. Preston

5 Activated partial thromboplastin time (APTT) 37

L. Paller

6 Prothrombin time (PT) 45

L. Paller

7 Endogenous thrombin potential 63

H. C Hemker and S. Beguin

v
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
8 Fibrinogen 79
M P. M. de Maat, G. D. 0. Lowe and F. Haverkate

9 Activated factor VII 89

J. H. Morrissey

10 Factor VII activity and antigen 99

G. Mariani, G. Liberti, T. D'Angelo and L. Lo Coco

11 Factor VIII clotting activity 107

P. M. Mannucci and A. Tripodi

12 Von Willebrand factor 115

P. M Mannucci and R Coppola

13 Antithrombin activity and antigen 121

J. Conard

14 Protein C activity and antigen 129

R. M Bertina

15 Protein S antigen 141

R M Bertina

16 Protein S activity 153

E. M Faioni

17 Activated protein C (APC) resistance 163

A. Tripodi

18 Tissue factor pathway inhibitor (TFPI) 171

P. M. Sandset

19 Lupus anticoagulant 183

D. A. Triplett

20 Heparin cofactor II 189

S. J. Bauman and F. C. Church

21 Fibrinopeptide A (FPA) 199

A. Haeberli

22 Thrombin-antithrombin (TAT) complexes 209

J. Harenberg

23 Prothrombin fragment Fl +2 217

A. Haeberli

vi
 CONTENTS
24 Tissue-type plasminogen activator (t-PA) activity 223
C. Kluft, P. Meijer, E. Ersdal and S. Rosen

25 Tissue-type plasminogen activator antigen (t-PA Ag) 231

M C. Alessi and L Juhan- Vague

26 Plasminogen activator inhibitor-l (PAl-I) antigen 239

P. J. Declerck

27 Plasminogen activity 247

P. J. Gaffney

28 Plasmin inhibitor activity (previously <l2-antiplasmin) 257

C. Kluft and P. Meijer

29 Plasmin-<l2-antiplasmin complexes (plasmin-plasmin

inhibitor complexes) 265
E. Hattey, M. Haumer, M R. Griffiths, V. Carroll
and B. R. Binder

30 Soluble fibrin and degradation products of fibrinogen

(FgDP), fibrin (FbDP; D-dimer) and total of FgDP and
FbDP(TDP) 275
W. Nieuwenhuizen and R. Bos

31 Venous occlusion test in fibrinolysis assays 285

J. Jespersen

32 List of manufacturers 293

Index 305

vii
Preface to the Second Edition

The first edition of this manual appeared in 1992 and was entitled ECAT Assay
Procedures. It was the result of a unique cooperation between experts brought
together by the European Concerted Action on Thrombosis and Disabilities
(ECAT). The Concerted Action was at that time under the auspices of the
Commission of the European Union. The second edition, like the first edition,
deals with diagnostic tests within the field of thrombosis. However, the second
edition has a broader scope because it is no longer limited by the frontiers of
ECAT. Experts allover the world, in and outside ECAT, have contributed to
this edition. The editors are very grateful for their contributions.
The need for a new edition is obvious. Since 1992 new assays have been
introduced for research, diagnosis, and therapy of thrombosis; for other assays
improvements have been suggested, while a few others became redundant. The
editors waived the radioimmunoassays of ~-thrombog1obulin and platelet factor
4 due to the fact that the kits required for these assays are rarely, or no longer,
available. Also the PAI-1 activity assay was waived as it is liable to many
inconsistencies and to large variations.
A list of names and addresses of manufacturers marketing the kits and
reagents has been compiled, together with a list of the recommended
nomenclature of quantities in thrombosis and haemostasis, in order to facilitate
the use of the updated version. These lists have been carefully compiled by
Johannes J. Sidelmann, PhD, Department of Clinical Biochemistry in Esbjerg,
Denmark.
The editors hope that the updated second edition will add to the greatly
needed harmonization and standardization. Both are mandatory for diagnosis,
prognosis, monitoring of disease, and for the improvement of the accuracy of
tests within the field of thrombosis.

Esbjerg, Leiden 1998

JOI'gen Jespersen
Rogier Bertina
Frits Haverkate

ix
Preface to the First Edition

This book offers a description of current and recently developed laboratory

assays in the field of haemostasis and thrombosis. It is the result of a unique
cooperation between experts from more than 60 institutes in 12 European
countries, brought together by the ECAT (European Concerted Action on
Thrombosis and Disabilities) under the auspices of the Commission of the
European Communities in Brussels, Belgium.
The ECAT, which was initiated in 1981, designed and performed three
prospective clinical studies to establish haemostatic factors as risk indicators
of thrombosis. Included were patients with angina pectoris at risk from
myocardial infarction, patients undergoing angioplasty at risk from re-stenosis,
and patients receiving hip replacement at risk from deep venous thrombosis.
Assay procedures were chosen, training courses for technicians held, and
essential reagents were supplied from a central source. A quality control
assessment scheme served to compare assay results both within and between
laboratories. In the angina pectoris study, centres determined most of the assays
locally; in the other two studies assays were performed centrally. The need for
further quality assessment in Europe led to a separate activity coordinated
by Dr. J. F. Davidson in Glasgow, including coagulation inhibitors and
plasminogen as risk factors for familial venous thrombosis. The Editors
hope the EeAT Assay Procedures book will contribute to further
harmonization of haemostasis assays, and ultimately to their standardization.
The need for this is evident from the failure of the ECAT quality controls
to demonstrate an obvious improvement in the agreement in haemostasis
assay results between laboratories.
This book has been preceded by two editions produced for internal use only.
Each assay was assessed by an expert who functioned in the ECAT as
representative of a reference laboratory. An assay committee coordinated the
activities. Each chapter in this book is written by the assay expert. Most of the
chapters give a detailed description of the assay recommended for use in EACT
and briefly mention alternatives. For this reason we have included a list of firms
marketing kits and reagents at the back of the book. The Editors will be happy
to receive critical comments, as well as suggestions for reagents and assays.
They can be included in the next updated version which will be prepared as
soon as the need arises.
xi
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
The Editors are indebted to the Commission of the European Communities,
to all ECAT participants who directly or indirectly were of great value, and in
particular to Professor F. Duckert, Basel, who did most of the pioneering
work.

Esbjerg, Leiden, 1991 J0rgen Jespersen

Rogier Bertina
Frits Haverake

xii
List of Contributors

M. CH. ALESSI P.O. Box 2215

Haematology Laboratory 2301 CE Leiden
Faculty of Medicine Timone The Netherlands
27, Boulevard Jean Moulin
F-13385 Marseille Cedex 5 V. CARROLL
France Department of Vascular Biology and
Thrombosis Research
S.J. BAUMAN University of Vienna Medical Faculty
Division of Hematology/Oncology Schwarzspanierstrasse 17
University of North Carolina School A-1090 Vienna
of Medicine Austria
932 Mary Ellen Jones Building
Campus Box #7035 F. C. CHURCH
Chapel Hill, NC 27599-7035 Division of Hematology/Oncology
USA University of North Carolina School
of Medicine
S. BEGUIN 932 Mary Ellen Jones Building
Department of Biochemistry Campus Box #7035
Cardiovascular Research Institute Chapel Hill, NC 27599-7035
University of Maastricht USA
P.O. Box 616 J. CONARD
6200 MD Maastricht Service d-Hematologie Biologique
The Netherlands Hospital HOtel-Dieu
1, Place du Parvis Notre-Dame
R. M. BERTINA F-75181 Paris Cedex
Hemostasis and Thrombosis Research France
Centre, C2-R
Leiden University Medical Centre R.COPPOLA
P.O. Box 9600 Hemophilia and Thrombosis Center
2300 RC Leiden IRCSS
Maggiore Hospital
B. R. BINDER Via Pace 9
Department of Vascular Biology and 1-20122 Milan
Thrombosis Research Italy
University of Vienna Medical Faculty
Schwarzspanierstrasse 17 T. D'ANGELO
A-1090 Vienna Hematology and Bone Marrow
Austria Transplantation Unit
University of Palermo Medical School
R.BOS Via del Vespro 129
Gaubius Laboratory, TNO-PG 1-90127 Palermo
Zernikedreef 9 Italy

xiii
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
P. DECLERCK Schwarzspanierstrasse 17
Laboratory for Pharmaceutical Biology and A-1090 Vienna
Phytopharmacology Austria
E. van Evenstraat 4
B-30oo Leuven M.HAUMER
Belgium Department of Vascular Biology
and Thrombosis Research
E.ERSDAL University of Vienna
Chromogenix AB Vienna, Austria
Taljegardsgatan 3
Molndahl, S-431 53 Sweden F. HAVERKATE
ECAT Foundation
E. M. FAIONI Geversstraat 19 B
Hemophilia and Thrombosis Centre 2341 GA Oegstgeest
Angelo Bianchi Bonomi The Netherlands
Via Pace 9
1-20122 Milan, Italy H.C.HEMKER
Department of Biochemistry
P. J. GAFFNEY Cardiovascular Research Institute
Division of Haematology University of Maastricht
National Institute for Biological P.O. Box 616
Standards and Control 6200 MD Maastricht
Blanche Lane, South Mimms The Netherlands
Potters Bar, Hertfordshire EN6 30G
UK J. JESPERSEN
Department of Clinical Biochemistry
J. GRAM Ribe County Hospital in Esbjerg
Department of Clinical Chemistry 0stergade 80
Ribe County Hospital DK-6700 Esbjerg
0stergade 80 Denmark
DK-6700 Esbjerg
Denmark I. JUHAN-VAGUE
Haematology Laboratory
M. R. GRIFFITHS Faculty of Medicine Timone
Department of Vascular Biology 27, Boulevard Jean Moulin
and Thrombosis Research F-13385 Marseille Cedex 5
University of Vienna France
Vienna, Austria
S. KITCHEN
A. HAEBERLI Department of Haematology
Department of Internal Medicine Royal Hallamshire Hospital
University Hospital Glossop Road
Freiburgerstrasse Sheffield S10 2JF
CH-3010 Bern UK
Switzerland
C.KLUFT
J.HARENBERG Gaubius Laboratory, TNO-PG
I. Medical Clinic P.O. Box 2215
Faculty of Clinical Medicine 2301 CE Leiden
University Hospital The Netherlands
Theodor-Kutzer-Ufer
D-68167 Mannheim J.-C. LlBEER
Germany Department of Clinical Biology
Scientific Institution of Public Health
E.HATTEY Louis Pasteur
Department of Vascular Biology Juliette Wytsmanstraat 14
and Thrombosis Research B-1 050 Brussels
University of Vienna Medical Faculty Belgium

xiv
 LIST OF CONTRIBUTORS

G. LIBERTI P.O. Box 2215

Hematology and Bone Marrow 2301 CE Leiden
Transplantation Unit The Netherlands
University of Palermo Medical School
Via del Vespro 129 L. POLLER
1-90127 Palermo European Concerted Action
Italy on Anticoagulation
The University of Manchester
L. LO COCO Stopford Building
Hematology and Bone Marrow Oxford Road
Transplantation Unit Manchester M13 9PT
University of Palermo Medical School UK
Via del Vespro 129
1-90127 Palermo F. E. PRESTON
Italy Department of Haematology
Royal Hallamshire Hospital
G. D. O. LOWE Glossop Road
University Department of Medicine Sheffield S10 2JF
Royal Infirmary
UK
10 Alexandra Parade
Glasgow G31 2ER, UK
S. ROSEN
M. P. M. DE MAAT Chromogenix AB
Gaubius Laboratory, TNO-PG Taljegardsgatan 3
Zernikedreef 9 M6lndahl, S-431 53 Sweden
P.O. Box 2215
2301 CE Leiden P. M. SANDSET
The Netherlands Hematological Research Laboratory
Ullev&1 University Hospital
P. M. MANNUCCI N-0407 Oslo
Hemophilia and Thrombosis Center Norway
IRCSS, Maggiore Hospital
Via Pace 9 D. A. TRIPLETT
1-20122 Milan, Italy Midwest Hemostasis and Thrombosis
Laboratories
G. MARIANI Ball Memorial Hospital
Hematology and Bone Marrow 2401 W. University Ave
Transplantation Unit Muncie, IN 47303-3499
University of Palermo Medical School USA
Via del Vespro 129
1-90127 Palermo, Italy A. TRIPODI
Hemophilia and Thrombosis Genter
P. MEIJER IRCGS Maggiore Hospital
Gaubius Laboratory, TNO-PG Via Pace 9
Zernikedreef 9 1-20122 Milan
P.O. Box 2215 Italy
2301 CE Leiden
The Netherlands I. D. WALKER
Department of Haematology
J. H. MORRISSEY Glasgow Royal Infirmary
Cardiovascular Biology Research Glasgow G4 OSF
Programme UK
Oklahoma Medical Research Foundation
825 N.E. 13th Street T. A. L. WOODS
Oklahoma City, OK 73104, USA Department of Haematology
Royal Hallamshire Hospital
W. NIEUWENHUIZEN Glossop Road
Gaubius Laboratory, TNO-PG Sheffield S10 2JF
Zernikedreef 9 UK

xv
Nomenclature of haemostasis factors

Chapter Title Nomenclature according to A Code value (B)

5 Activated partial Plasma-Coagulation, QU60476

thromboplastin time surface-induced

6 Prothrombin time Plasma-Coagulation, tissue QU60478

factor-induced

7 Thrombin generation

8 Fibrinogen Plasma-Fibrinogen CAS9001-32-5

9 Activated factor VII Plasma-Coagulation factor EC3.4.21.2l

VII, activated

10 Factor VII clotting activity Plasma-Coagulation factor CAS9001-25-6

VII (coag.; procedure)

Factor VII antigen Plasma-Coagulation factor CAS9001-25--6

VII (imm.; procedure)

11 Factor VIII clotting activity Plasma-Coagulation factor CAS9001-27-8

VIII (coag.; procedure)

12 Von Willebrand factor Plasma-Von Willebrand MSH94D01484l

factor

13 Antithrombin activity Plasma-Antithrombin CAS9000-94 -6

(enz.; procedure)

Antithrombin antigen Plasma-Antithrombin CAS9000-94-6

(imm.; procedure)

14 Protein C activity Plasma-Protein C QU60410

(enz.; procedure)

Protein C antigen Plasma-Protein C QU60410

(imm.; procedure)

15 Protein S antigen Plasma-Protein S MSH94DO 17293

(enz.; procedure)

16 Protein S activity Plasma-Protein S MSH94DO 17293

(coag.; procedure)

17 Activated protein C resistance

18 Tissue factor pathway Plasma-Tissue-factor- MSH94C05l928

inhibitor pathway coagulation
inhibitor

19 Lupusanticoagulan~ Plasma-Lupus MSH94D016682

anticoagulant

20 Heparin cofactor II Plasma-Heparin CAS8l604-65-l

cofactor II

21 Fibrinopeptide A Plasma-Fibrinopeptide A MSH94DOO5344

xvii
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

Nomenclature of haemostasis factors - Cont.

Chapter Title Nomenclature according to A Code value (B)

22 Thrombin-antithrombin Plasma-Thrombin- MSH94C046193

complexes Antithrombin complex
23 Prothrombin fragment F 1+2 Plasma-Prothrombin CAS72270-84--9 +
fragment 1+2 CAS78768-79-3
24 Tissue plasminogen activator Plasma-Plasminogen QU60496
(t-PA) activity activator, tissue type
(enz.; procedure)
25 Tissue plasminogen activator Plasma-Plasminogen QU60496
(t-PA) antigen activator, tissue type
(imm.; procedure)

26 Plasminogen activator Plasma-Plasminogen MSH94D017395

inhibitor-l (PAl-I) antigen activator inhibitor 1
(imm.; procedure)

27 Plasminogen activity Plasma-Plasminogen CAS9001-91-6

(enz.; procedure)
28 Plasmin inhibitor activity Plasma-Plasmin inhibitor QU60494
(enz.; procedure)
29 Plasmin-plasmin inhibitor Plasma-Plasmin-Plasmin MSH94C037742
complexes inhibitor complex
30 Fibrin degradation products Plasma-Fibrin fragments QU60486
(FbDP)
Fibrinogen degradation Plasma-Fibrinogen fragments QU60489
products (FgDP)
Total degradation products Plasma-Fibrin + Fibrinogen QU60468 +
(TOP) fragments QU60489
D-Dimer Plasma-Fibrin D-dimer MSH94C036309
Soluble fibrin monomer Plasma-Fibrin soluble QU60488

A: Scientific and Standardization Committee Communications. Nomenclature of Quantities and

Units in Thrombosis and Haemostasis (Recommendation 1993). A Collaborative Project of
the Scientific and Standardization Committee of the International Society on Thrombosis and
Haemostasis (lSTH/SSC) and the Commission/Committee on Qualities and Units (in Clinical
Chemistry) of the International Union of Pure and Applied Chemistry-International Federation
of Clinical Chemistry (lUPAC-IFCClCQU(Cc)). Thromb Haemostas 1994; 71: 375-94.
B: The code values given by the International Federation of Clinical Chemistry and Laboratory
Medicine (IFCC) and the International Union of Pure and Applied Chemistry (IUPAC) are
obtained from http://inet.uni-c.dklhome/ifcc_iupac3npu/

xviii
1
Introduction to laboratory assays in
haemostasis and thrombosis
J. GRAM and J. JESPERSEN

INTRODUCTION

The aim of this Assay Manual is to provide the routine laboratory with
information on the quantitation of a number of key variables within the
haemostatic system. Only via harmonization, or if possible standardization,
establishment of quality assessment schemes, and performing 'good laboratory
practice' can the routine laboratory provide 'good medical service'. Besides the
specific description of the methods of assay, these important aspects are dealt
with in depth in the chapters at the beginning of the book, including an extensive
description of sampling procedure.
Here we shall concentrate on a short description of the haemostatic system,
the bridging between research and service, refinement of methodology, new
demands of the clinician and the laboratory, accuracy of measurement, etc.

THE HAEMOSTATIC SYSTEM

The mechanism of haemostasis protects the integrity of the vascular system

so that, in case of injury, the tissues are healed and their function restored.
The haemostatic process depends on complex interactions between the vessel
wall, the platelets and the process of coagulation and fibrinolysis.
Several mechanisms, both local and humoral, participate in the regulation
of the platelet - fibrin plug formation and resolution. The coagulation system
(an impression is shown in Figure 1.1), in particular with the use of activated
platelets as template, generates the enzyme thrombin, which is able to convert
fibrinogen into a fibrin gel. The system of fibrinolysis (Figure 1.1) produces
the fibrin-dissolving enzyme plasmin.
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

XII =-=--:: Activation

HMWK l~ - - t I Inhibition
Xr~CO/lagen
----
TF
Ca2 +
XI ----. Xla Vlla- VII

IX

VIII-+ Villa
~ ... "-

I
/,//, ~ !~ ~":::r":nmb;O ~o
\ Protein CIS ------ TAT
\.,. V -:::;tIVa
',\ / ..... -~ .......
"\ I Prothrombin ~ Thrombin
- + Fragment1+2
-
...... ~ \

,- ,_.....-
1
~ ;'
/
.,."'0'ftlb.o. - - ---:,..1
- I
ftlO<1/JI;f/ I Fibrinogen Fibrin monomer
~
,
FpA 1
Fibrin polymer
"
PAI-1 t-PA --- 1
~~ Fibrin

r
Plasminogen ----+. Plasmin

T PAP 1
Fibrin
Plasmin
HRG degradation
inhibitor
products
Figure 1.1 Some of the key components involved in activation of coagulation and fibrinolysis.
For details see relevant chapter

Several activating stages participate in the conversion of the precursors

prothrombin and plasminogen to the active enzymes. Inhibitors of various
types regulate or block the generation of the active compounds and their action
(Figure 1.1).
At the site of injury (pertubation) of the vessel wall, tissue thromboplastin
is released from damaged cells, and clot-promoting endothelial and subendo-
thelial structures are exposed, which may activate the extrinsic and intrinsic
2
 LABORATORY ASSAYS IN HAEMOSTASIS AND THROMBOSES
pathways of thrombin generation. Similarly, plasmin can also be produced
from different, relatively well-characterized pathways. One of these is the
pathway initiated by the release of plasminogen activator, tissue-type, t-PA,
from the endothelium into the circulation (Figure 1.1).
Furthermore, two pathways (the factor XII-dependent and the pro-urokinase
system) contribute to in-vivo fibrinolysis, but their clinical relevance is more
uncertain, and will not be dealt with in the present manual.

NEW METHODOLOGY

Considerable progress over the past decade, due to the development and
introduction of new assays, has provided new insight into the pathogenesis of
thrombosis and bleeding, but has also resulted in the recognition of a number
of risk indicators of cardiovascular diseases. The refinement of methodology
and increased understanding of pathophysiology has caused a shift from the
use of global clotting assays to the use of synthetic substrates, measurements
of single proenzymes and enzymes, and the use of monoclonal antibodies.
One serious problem has always been the difficulty in converting a given
prolongation in clotting time to percentage or unit activity per mi. Although
some of the problems with the clotting assays have been reduced, new problems
have arisen.
The use of monoclonal antibodies against many of the factors depicted in
Figure 1.1 has its own inborn problems due to formation of enzyme inhibitor
complexes, heterogeneity of proteins, formation of several split products with
the same epitope, also present in the original product, etc.

DEMANDS OF TRUENESS IN MEASUREMENT

Obviously, whenever a laboratory test moves from the research laboratory to

clinical practice there is uncertainty, and demands are placed on the clinical
chemistry laboratory and clinicians. With knowledge concerning haemostasis,
thrombophilia, DNA technology, its use and limitation, epidemiology, and
absolute risk, the clinician can, in cooperation with the clinical chemistry
laboratory, perform good clinical practice, i.e. evidence-based health care. But
there are also demands on the clinical chemistry laboratory, i.e. to perform
'good laboratory practice'.
Analytical reliability has generally not kept up with technical advancesi--{i.
Thus the assay results of an analyte may deviate significantly when they are
produced in different laboratories4 ,5. Despite advances in analytical quality
assurance, constancy of assay results is not assured over longer periods, and
even within the same laboratory variations in reagents from one batch to
another may contribute critically to a high total variance in assay results i. The
Scientific and Standardization Sub-committee under the auspices of the
International Society of Thrombosis and Haemostasis (ISTH) has addressed
this problem over many years. International standards are available. Secondary
standards have been produced and calibrated.
3
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
ACCURACY OF MEASUREMENT

A clinical laboratory test is a reproducible analytical procedure which relates

to the health status of a subject. However, accurately repeated measurements
without error are theoretically impossible7 . According to ISO definitions errors
of a measurement system can be described by trueness of measurement and
precision of measurements.
Trueness of measurements is a systematic component of error (bias) which
is described by 'closeness of agreement between average value and a true value'.
Precision of measurements is a random component of error which is described
by 'closeness of agreement between results of measurement'. The accuracy of
measurement is dependent upon trueness of measurement as well as by precision
of measurement.
Most reports dealing with new haemostatic tests have reported data on
precision of measurement, while much less is usually known with respect to
trueness of measurements. This means that, with a few exceptions, there is little
information available concerning accuracy of measurements. Obviously it is
difficult to make reliable studies on 'trueness of measurement' within the field
of haemostasis. First, measurements of haemostatic quantities are, as described,
based on complex biological systems in which the exact molecular form of the
analyte under study is hardly known. Second, with the exception of prothrombin
time measurements, a comprehensive coherent reference measurement system,
as known from clinical biochemistry, does not exist9--11. Such a standardization
system, which secures the highest possible level of accuracy of measurements
in the laboratory, is characterized by a coupling of the analytical method (defini-
tive method, reference method, field method) with the reference material
(primary reference material, secondary reference material, standardst
Despite these problems it is reasonable to suggest that developers of new
analytical procedures should focus more on 'trueness of measurement' in order
to produce 'meaningful measurement results', as recently stated by Biittner7 •

PROPOSAL FOR MINIMAL REQUIREMENTS

The aim of measuring haemostatic variables in a certain clinical situation is

to relate the measurement result to the health status of the subject under study.
Therefore, it is of great importance, to the highest possible degree, to reduce
the influence of factors which may influence the measurement result but without
being significantly related to the health status of the patient. Such factors,
which can often be controlled to a certain extent, can be separated into
pre-analytical factors and analytical factors.

Pre-analytical factors
The pre-analytical factors which may affect measurement results of haemostatic
variables can be related to the preparation of the patient and to the blood
sampling procedure, and are described in Chapter 3.
4
 LABORATORY ASSAYS IN HAEMOSTASIS AND THROMBOSES

Analytical factors
The analytical method is of great importance for the production of accurate
results in the research as well as in the clinicallaboratory7,9-11. Therefore a
strict and consistent description of a number of analytical factors is neces-
sary12,13 in order to secure transferability of accurate data between laboratories,
and in order to improve transferability from scientific publications to daily
laboratory work (Table 1.1). The principle of the method should be described,
and it should be detailed whether the measurement procedure makes use of
biological or chemical substrates for measurement of the haemostatic process.
It should be noted whether the principle causes a conflict with the application
of the method on some biological fluids. The analytical procedure should be
described in a logical sequence and preferentially in series of small steps. The
procedure of constructing a calibration curve should be detailed separately. It
should be clearly described how the measurement signal is converted to
concentration. The results should, when possible, adhere to international
WHOINIBSC standards and they should be presented in units officially accepted
by ISTH 14. Particular attention should be paid to reliability characteristics of
the method (Table 1.2). Furthermore, it should be elucidated whether the use
of the method in different laboratories gives transferable results. The reference
ranges of the method should be reported, and possible sex and age differences
of the reference ranges should be described. The intra-laboratory traceability
of data should be secured with the use of a well-described quality assurance
system. The specific daily quality assurance procedures should be described.
Finally, the characteristics of the quality control material and the frequency
of analysis should be reported.

CONCLUSION

As reviewed by others, the ideal total diagnostic test evaluation consists of five
phases: phase 1: analytical performance, phase 2: classification performance,
phase 3: clinical performance, phase 4: outcome performance, phase 5: utility
performance 15.
This chapter exclusively deals with a proposal for minimal requirements for
analytical tests, which include the preparation of the subject under study, the
preparation of the samples and the analytical method. A proper description

Table 1.1 Factors which may be detailed regarding the

analytical method

Measurement principle
Analytical steps/instrumentation
Data reduction
Reliability characteristics
Transferability characteristics
Reference ranges
Quality assurance

5
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Table 1.2 Reliability characteristics which may be described for an analytical
method dealing with measurement of haemostatic quantities

Calibration curves and linearity

Precision
Trueness
Analytical recovery studies
Interference studies
Comparison of method studies
Analytical sensitivity
Analytical detection limit

of the analytical method and a consistent documentation on reliability char-

acteristics are necessary if users of new methods within haemostasis testing
should have the opportunity to evaluate whether the method produces accurate
results. The present manual is a contribution aiming at moving in this direction.
Furthermore, it may help to adhere to the suggested guidelines proposed by
the Council of Biology Editors when reporting studies of diagnostic accuracy,
which is a critical part of modem medicine 16•

References
I. Thompson SG, Martin JC, Meade Tw. Sources of variability in coagulation factor assays.
Thromb Haemostas 1987; 58: 1073-7.
2. Thompson SG, Duckert F, Haverkate F, Thomson JM. The measurement of haemostatic
factors in 16 European laboratories: quality assessment for the multicentre ECAT Angina
Pectoris Study. Report from the European Concerted Action on Thrombosis and Disabilities
(ECAT). Thromb Haemostas 1989; 61: 301--6.
3. Thompson SG, Calori G, Thomson JM, Haverkate F, Duckert F. The impact of sequential
quality assessment exercises on laboratory performance: the multicentre ECAT Angina Pectoris
Study. Report from the European Concerted Action on Thrombosis and Disabilities (ECAT).
Thromb Haemostas 1991; 65: 149-52.
4. Gram J, Declerck PJ, Side1mann J, Jespersen J, Kluft C. Multicentre evaluation of commercial
kit methods: plasminogen activator inhibitor activity. Thromb Haemostas 1993; 70: 852-7.
5. Declerck PJ, Moreau H, Jespersen J, Gram J, Kluft C. Multicenter evaluation of commercially
available methods for the immunological determination of plasminogen activator inhibitor-l
(PAl-I). Thromb Haemostas 1993; 70: 858--63.
6. Pyke SDM, Thompson SG, Buchwalsky R, Kienast J, on behalf of the ECAT Angina Pectoris
Study Group. Variability over time of haemostatic and other vascular risk factors in patients
suffering from angina pectoris. Thromb Haemostas 1993; 70: 743--6.
7. Buttner J. The need for accuracy in laboratory medicine. Eur J Clin Chem Clin Biochem 1995;
33: 981-8.
8. International Standards Organisation (ISO) . Accuracy (trueness and precision) of measurement
methods and results. Part I: General principles and definitions (ISOIDIS 5725-1). Geneva:
ISO, 1990.
9. Tietz Nw. A model for a comprehensive measurement system in clinical chemistry. CIin Chem
1979; 25: 833-9.
10. Dybkaer R. Reference materials - a main element in a coherent reference measurement system.
Eur J Clin Chem Clin Biochem 1991; 29: 241--6.
11. Bowers GN. Clinical chemistry analyte reference systems based on true value. Clin Chem
1991; 37: 1665--6.
12. Fraser CG, Geary TO, Worth HGJ. Guidelines (1988) for preparation of laboratory procedure
manuals for clinical chemistry. J Clin Chem Clin Biochem 1988; 26: 45-9.

6
 LABORATORY ASSAYS IN HAEMOSTASIS AND THROMBOSES
13. Stamm D. Recommendation for description of a selected method in clinical chemistry. J Clin
Chern Clin Biochern 1979; 17: 280--2.
14. Committee of the International Society on Thrombosis and Haemostasis (ISTH/SSC) and
the Commission/Committee on Qualities and Units in Clinical Chemistry of the International
Union of Pure and Applied Chemistry - International Federation of Clinical Chemistry
(IUPAC-IFCClCQU (CC)). Nomenclature and units in thrombosis and haemostasis. Thromb
Haemostas 1994; 71: 375-94.
IS. Magid E. Minimal requirements for test evaluation. Scand J Clin Lab Invest 1997; 57(Suppl.
227): 90--4.
16. Bruns DE, ed. Reporting diagnostic accuracy. Clin Chem 1997; 43: 2211.

7
2
Good medical laboratory services
guidelines
J.-C. LIBEER

INTRODUCTION

Medical laboratories must provide a high-quality service by producing accurate,

precise, relevant and comprehensive data that can be applied to the medical
management of patients. In patients with inherited or acquired bleeding and
thrombotic disorders the laboratory has a vital role in the diagnosis and
management of the patients. Any misdiagnosis of the patients is likely to be
serious, e.g. indication and adjustment of long-term anticoagulant therapy.
Therefore the laboratory must follow quality rules that can be described as
'rules for good laboratory practice'. However, this terminology is rather
confusing, as good laboratory practice (GLP) is already defined in a specific
application field of laboratory medicine. GLP is a standard or guideline applied
on laboratory tests from the pre-clinical phase of a clinical trial. The GLP
standards in European countries are based on the OEeD guidelines l . In practice
only a few medical laboratories are involved in GLP, namely laboratories
performing analyses of laboratory tests on animal samples for toxicological,
pharmacokinetic or safety studies. Tests for patient care are mainly performed:
(i) for monitoring of drugs - therapeutic drug monitoring (TDM) - or markers,
for the follow-up of therapy or the health status of a patient and (ii) for
diagnostic purposes such as confirmation or as an aid in clinical diagnosis. For
the management of a correct laboratory outcome more is needed than a correct
analytical result. Therefore the term 'good medical laboratory services' (GMLS),
as used by Haeckel2 , was chosen to describe the quality rules to be followed.
By using the term 'services' the decisive clinical requirements and characteristics
are included.

9
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

EVOLUTION OF QUALITY MANAGEMENT IN MEDICAL

LABORATORIES

In the laboratory sector, medical laboratories, and more especially clinical

chemistry laboratories, have a long tradition of internal and external quality
control. In clinical chemistry laboratories, Levey and Jennings 3 introduced
control charts as used in industrial processes, and known as Shewhart plots or
control charts4 • In 1981 Westgard et al. published their first concept on the
multi-rule Shewhart chart for internal quality control in clinical chemistry5,
the so-called 'Westgard rules'. They produced benchmark work on internal
quality control in clinical chemistry and on total quality management (TQM)6
in medical laboratories by using approaches similar to those used in industry.
The first results of an external quality assessment survey were published by
Belk and Sundermann in 19477 and external quality assessment schemes are
now available in most European countries in the various fields of laboratory
medicine. Recently, new quality concepts were introduced in analyticallabora-
tories extending quality control to TQM. The goal of TQM is the satisfaction
of customers, gained by continued improvement of processes and services.
TQM aims to guarantee the quality of a laboratory result by managing all
aspects: human resources, equipment, housing, reagents, process control, quality
system management, etc. We can consider quality control as a static process
while quality management is a dynamic process. Each effective quality system
aims at continuous improvement. Deming8 describes this process in four steps:
the PDCA (plan-do--check-adjust) elements. The implementation may increase
productivity and reduce avoidable costs caused by errors.
Within the European Union (ED) the EN 45001 9 standard is recommended
for all types of test laboratory. This standard requires not only technical
competence and analytical quality management, but also the management of
system quality aspects (humanresources, structure of the organization, manage-
ment of documents, etc.). The European EN 45001 standard is equivalent to
the worldwide-accepted ISO guide 25. For the moment ISO 25 is under revision
and, according to an ISO-CEN mutual agreement, the new ISO 25 guide will
replace the present EN 45001. The draft for the new ISO 25 10 makes a clear
distinction between quality system requirements and technical requirements.
From a social point of view it would be unacceptable if other laboratories (e.g.
those analysing foods and concrete) were submitted to more stringent quality
criteria than laboratories analysing medical samples from patients. Therefore
the medical laboratory profession had followed this evolution in many countries,
but found the EN 45001 standard insufficient for covering all aspects of good
c1inicallaboratory services:

1. In the medical laboratory it is not sufficient to have only technical competence.

Appropriate medical competence is also required.
2. The medical laboratory may receive infectious patients and infectious samples,
and performs tests using radioactive material or other dangerous substances.
Due consideration should therefore be given to safety aspects with regard
to other patients, personnel, other persons, and the environment.
3. A request in a medical laboratory can be formulated by the requesting
10
 GOOD MEDICAL LABORATORY SERVICES
physician as an enumeration of tests or as an investigation of a well-defmed
syndrome.
Correct sampling also includes patient preparation.
Subcontracting under the responsibility of the primary laboratory is usual
in medical laboratories.
Validation of results includes not only analytical, but also biological, validity,
and diagnostic and therapeutic usefulness.
Some general requirements for reports are not relevant to reports from
medical laboratories.
No interpretation of results is allowed.

All these points have contributed to specific proposals for quality systems for
medicallaboratories l1 . The proposals are intended to cover the pre- and
post-analytical phase, and also include national requirements which are not
always relevant for a quality system. In the meantime international organiza-
tions have tried to propose some more harmonized guidelines. ECLM (European
Confederation of Laboratory Medicine) is in favour of accreditation rather
than of certification of medical laboratories. Medical laboratories must follow
at least appropriate standards and guidelines developed for laboratories. ECLM
has reviewed the ISO/IEC Guide 25 from the point of view of medical
laboratories in an explanatory document12. The EC4 group (European Com-
munity Confederation of Clinical Chemistry) produced a consensus document
with essential criteria for quality systems in clinical biochemistry laboratories
in the EU 13. It comes as no surprise that all these different approaches embarrass
medical laboratory workers who are not specialized in this matter. The medical
laboratory profession has often tried to demonstrate the particular features of
their profession with regard to other laboratory fields, and even to other
countries. However, the same can also be demonstrated for every laboratory
field. If we look more at the common elements than at the differences, it is
possible to develop a harmonized approach on quality systems in accordance
with other laboratory fields. Each national proposal, and even the international
proposals, for a quality system in medical laboratories can be split into four
levels, as presented in Figure 2.1. Medical laboratories can use for their quality
management system, and for the technical and analytical competence, the same
approaches as other laboratory fields. The new ISO 25 guide especially covers
all quality system elements from the ISO 9000 series, and also covers many
aspects for which medical laboratories had been criticized. With the exception
of medical competence for staff members, clinical audit and more security
aspects, the new ISO 25 also covers the requirements from the EC4 document.
In order to maintain a harmonized approach these additional aspects could
be combined in a third level. As social security systems, even in the EU, are
not yet harmonized, we will find some specific national (or regional) require-
ments in a fourth level. Usually, proposals for quality systems in medical
laboratories use an approach based on pre-analytical, analytical and post-
analytical procedures, and require a description of all phases of laboratory
analysis from the moment that a sample enters the laboratory until the sending
of the report to the requesting physician.
11
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

MEDICAL COMPETENCE
(EN?/ISO?)

QUALITY SYSTEM: ISO 9000 (EN 29000)

Figure 2.1 Quality system layers for medical laboratories

Concepts of TQM in medical laboratories will not only consider the quality
of service provided, but will also refer to other linked aspects such as safety
requirements, environmental impact and ethical considerations.

QUALITY MANAGEMENT ELEMENTS IN MEDICAL LABORATORIES

GMLS require that tests requested must be appropriate to the clinical problem,
tests must be analytically correctly performed and the results interpreted
correctly. Requests for tests and interpretation of results belong to medical
quality management. Correct analytical results are based on reagents, equipment,
sample quality and the whole process within the laboratory, including human
resources. Tools for quality management within the laboratory are system
quality and analytical quality management. A part of the quality is defined
outside the laboratory. Medical laboratories frequently use industrially prepared
kits and systems; their quality will be defined by the manufacturer and not by
the user. Sampling is often performed outside the laboratory. All elements
influencing the quality before samples reach the laboratory must be managed
in the pre-laboratory phase. After the release of the protocol with the laboratory
results, the requesting clinician must adequately use the data provided. This
phase of GMLS can be considered as a post-laboratory management layer.

SYSTEM QUAUTY MANAGEMENT

The system quality management is no different in a medical laboratory than

it is in any other organization. The new ISO 25 gives the following elements
for system quality management: quality management system, organization and
management, document and information control, review of request, tender or
contract, subcontracting of tests and calibrations, procurement of services and
12
 GOOD MEDICAL LABORATORY SERVICES
supplies, service to and feedback from the client, control of non-conforming
testing and/or calibration work, corrective action, preventive action, internal
audit and management reviews. Most of these items are also relevant for
medical laboratories.
Examination of several proposals for quality manuals for medical laboratories
in different European countries demonstrates that the 'check' phase from the
CPDA model from Deming is often incomplete, and internal audits, as well as
periodic management review, are not included ll . However, both elements are
essential in each quality system.

ANALYTICAL QUALITY MANAGEMENT

In contrast to other laboratory disciplines medical laboratories demonstrate

some particularities, even at the analytical level:
1. Most medical laboratory tests are performed with industrially prepared kits
and reagents.
2. The method in use is often a compromise between acceptable performance,
economical feasibility and the suitability to produce a result at the right time
(acceptable turnaround time).
3. In most other analytical laboratory fields sample size is not usually a limiting
factor; in medical laboratories sample size (e.g. paediatric samples) often
limits the use of some methods.
Management of analytical quality is possible only if, first, all factors involved
are defined. A schematic presentation of the relevant factors involved in
analytical quality is given in Table 2.1. The permanent external Jactors are
related to the laboratory's choice of analytical principle, equipment, calibrator
and kit reagents, etc. The quality of the permanent external factors will be
defined by the manufacturers of the in-vitro medical devices. The variable
externalJactors are due to variations in batch production (both calibrators and
reagents). The individual laboratory has little influence on the quality of
external factors, and can only set up and apply standard criteria for selecting
the most appropriate devices (management of permanent external factors) and

Table 2.1 Factors involved in analytical quality

Sources External factors Internal factors

Permanent factors Analytical principle, Implementation in Laboratory's choice of

method, equipment the laboratory, quality
instructions
Variable factors Variation in batches Variation in Expected and
performance and unexpected variation
regular errors
Manufacturer's Laboratory's
responsibility responsibility

13
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

use a long-term internal quality control system (management of variable

external factors). A poorly performing system or a kit with poor perform-
ances, or demonstrating a large batch-to-batch variability, will never produce
a reliable result, even in a laboratory with the best-quality management system.

Quality management of external factors

Medical diagnostic manufacturers and distributors are already convinced that
a quality management system can improve the quality of their products and
services. Most companies already have, or are preparing, a certification based
on ISO 9001 or 9002 14,15. In the EU these guides are called EN 29001 and EN
29002; for medical devices the additional standards EN 46001 and EN 46002 16,17
are used.
For better protection of the patient a European directive on in-vitro diagnostic
(IVD) medical devices is in preparation 18 • This directive will impose severe
requirements on manufacturers. The ultimate goal for better patient protection
can be guaranteed only if more stringent quality criteria are also applied by
the users of these kits, in casu the medical laboratories. Each laboratory must
define criteria for the choice of suppliers, and for new equipment and kits, in
such a way that the quality of the input elements is guaranteed.

Quality management of internal factors

The permanent internal factors are related to the implementation of the method
in the laboratory (reagent/sample volumes, commercial calibrators, the number
of calibration points, curve fittings, calculation factors, time factors, etc.).
Control of correct implementation will be done by method validation. The
variable internal factors are linked to all random errors that can occur within
the laboratory, and must essentially be controlled by the laboratory internal
quality control system. The management of internal factors is not limited to
pure analytical quality, but also includes elements from the system quality
management: (i) education and training of staff; (ii) pre-analytical elements
such as sample identification, sample flow through the laboratory, request
processing; (iii) post-analytical steps such as run validation, transmission of
results, protocol generation.

Quality management of the pre-laboratory phase

The pre-laboratory quality layer contains both medical and analytical aspects.
The medical aspects will be dealt with the medical quality management aspects
within the laboratory. Quality management of reagents and equipment have
already been described (see quality management of external factors). Depending
on the type of laboratories, few or a majority of samples are taken under the
responsibility of persons outside the laboratory. GMLS require that before
14
 GOOD MEDICAL LABORATORY SERVICES

analysis the laboratory performs a verification not only on sample identification

but also on the appropriateness of the sample itself. For coagulation tests,
(Chapter 3), underfilling and overfilling of collecting citrate tubes may influence
the results. This information should not only be used in the validation of the
analytical result, but if relevant should also be mentioned in the report. Although
the laboratory has no direct impact on this phase, clear instructions for sampling
and patient preparation, follow-up of suitability of sampling tubes and samples
can be helpful in improving this process. The pre-laboratory phase also includes
sample transport and conservation.

MEDICAL QUALITY MANAGEMENT

Management of system and analytical quality is fundamental for GMLS for

patient care, but does not cover all required aspects. A pure analytical result
cannot be used for patient care. Most guidelines and standards for quality
management in laboratories are primarily focused on analytical quality. They
do not include all features requested by GMLS; consultation and interpretation
of test results are not sufficiently dealt with in some documents. Safety and
ethics are usually completely neglected. The medical quality of a laboratory
includes a pre-laboratory element, the appropriate choice of tests, and a
post-analytical element, the clinical validation and consultancy function of
laboratory staff.

Choice of tests
Appropriateness of tests can be obtained only by dialogue between clinician
and clinical pathologist. A better understanding and collaboration between
the two will cut costs and prove more effective. The ultimate question should
be: 'will the patient find a benefit after undergoing or not undergoing this
laboratory test'. Strategies for diagnosis, prognosis or monitoring of diseases,
and whether an appropriate effect on the health care outcome for an individual
patient can be achieved, are extremely helpful in the choice of tests. In France
biologists have developed a system for helping requesting physicians in the
choice of test selections (references medicales opposables RMO and references
medicales positives)19. This example illustrates the role that can be played by
clinical pathologists in appropriate laboratory tests prescription behaviour.

Clinical validation and consultancy function

After the release of a validated analytical result the clinical validation starts.
Post-analytical validation includes biological and nosological validation as
outlined by Biittner20. In screening tests, biological validation includes
comparison of an individual result with the result of a reference group for
which reference values are determined. In a monitoring situation, results are
15
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
compared with previous results from the same person. Within this range the
laboratory should have knowledge of the intra-individual biological variation
and its own analytical variation. Both values allow calculation of the critical
difference. This critical difference, referred to as the reference change, is the
highest difference between two consecutive measurements that, at a chosen
level of probability, may still be due to the combined effect of the analytical
and biological variations.
It is the responsibility of the clinical pathologist to give an interpretation of
results. This is defined as clinical validation or nosological validation. During
this process, single results are related to other results. This is a very complex
cognitive procedure in which it is necessary to combine clinical data with
laboratory findings in order to discover whether these findings can be explained
causally from the pathogenesis of a disease. Clinical experience plays an import-
ant role in these interpretations. In the future the use of intelligent expert
systems will become a powerful tool for pathologists in this validation process.
Regular contacts between clinicians and pathologists can further improve
GMLS. These contacts must give the clinical pathologist a better insight into
the expectations of clinicians. On the other hand clinicians will be informed
of the fast-evolving possibilities of laboratory medicine. In the future both
parties will have an interest in better understanding and collaboration.
Consultation between pathologists and clinicians will also improve their work.
In this way we can consider a 'post-laboratory' quality layer taking into account
correct interpretations and adequate decisions by the clinicians. All relation-
ships between clinicians and laboratory form the 'clinic-laboratory interface'.

LABORATORY KNOW-HOW

GMLS also requires fundamental and practical analytical and medical know-
ledge. Today this know-how is not always available in some laboratories. As an
example we mention here therapeutic drug monitoring. New technology allows
every laboratory to measure drug levels. A quality system can guarantee a
correct analytical result. However, the interpretation of these results demands
a fundamental and practical knowledge of pharmacokinetics, therapeutic range
and metabolism. Results without clinical validation are worthless to the clini-
cians; but currently many mistakes are observed in the information given to
clinicians, due to lack of knowledge by laboratory staff.
The very rapidly changing techniques, and the introduction of new tests,
will require continuous training of laboratory staff. Although continuous
training is necessary in each profession, a mandatory system has been installed
in several countries for medical staff. These systems require the collection of
a number of CFU (continuous formation units) during a year. However, this
system does not guarantee that medical laboratory staff have real competence
in all those analyses performed in their laboratory. Without this competence
it is impossible to validate both analytical and clinical results and to guarantee
GMLS.
16
 GOOD MEDICAL LABORATORY SERVICES

CONTROL OF GMLS

As described, the clinical outcome of a medical laboratory investigation must

be guaranteed by adequate quality management, taking into account all the
elements of GMLS. Implementation is the most important. In order to maintain
a high analytical quality level, periodical review of the quality system will be
essential. This review must be done within the laboratory (internal audit and
management review) and by an external party (external audit). Just as an author
does not see errors in his own manuscript, Ii laboratory can demonstrate several
failures in his quality system without being aware of this. An external audit
will help to detect these failures and to improve the process. The external
audit must cover all aspects of GMLS in order to be efficient. Efficiency of
the audit will also depend on the qualification of the assessors. Assessors must
combine the analytical, technical and medical knowledge with the capability
for 'good auditing practices'. Most quality audits do not include an audit of
the medical quality management aspects. This audit type is known in the UK
as a 'clinical audit'. Clinical audit involves systematically looking at procedures
used for the diagnosis, care and treatment, examining how associated resources
are used and investigating the effect which care has on the outcome and quality
of life for the patient21 • The clinical audit will check if, within the laboratory,
the medical and analytical knowledge is present for a correct interpretation of
performed tests, and if there is evidence of regular contacts with clinicians.
Clinical auditing is difficult, because medical staff are not familiar with the
procedure; it includes relationships between clinicians and laboratory staff and
touches on the personal capabilities of staff members. However, the clinical
audit may not be an examination of the personal knowledge of laboratory staff
members but must aim to evaluate whether the medical quality management
system, implemented by the laboratory, is adequate.

CONCLUSIONS

Benefits following the implementation of quality management systems have

been widely recognized in industry. Budget restrictions on public health systems,
and on laboratory services in particular, have led to an increasing awareness
of efficient laboratory management. The challenge in this management is to
find the optimum conditions for quality, economy and speed. The implementation
of GMLS offers the possibility for more effective utilization of laboratory resources.
The whole GMLS model as described in this chapter is shown diagrammatically
in Figure 2.2.
The new approaches will also demand a new mentality in medical laboratories.
The laboratory director and clinical pathologists will be forced to review their
management style. Delegation of responsibilities, motivation and persuasiveness
will become more and more important. Laboratory workers will be forced to
change habits and to take more responsibilities. The whole laboratory must be
considered as one team, working together for the same goal- improved quality
of results - and demonstrating this quality to society. Contacts between patholo-
gists and clinicians must be stimulated. They improve the efficiency of the
17
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

pre-laboratory
analytical quality
layer

: ~~t~ ~~
I

,I

mecficallaboratory quality layer

•••
••
laboratory
know-how Post-laboratory
quality layer

t
GMLS
Figure 2.2 The good medical laboratory services model

GMLS. In the model we defined these relations as the 'clinic-laboratory

interface' .
Management of clinical outcome in medical laboratories is a challenge for
the future and is based on two essential requirements: (i) the implementation
of, and keeping up-to-date, a quality system in medical laboratory organiza-
tions and (ii) a reconsideration of the tasks, duties and knowledge of clinical
pathologists.
18
 GOOD MEDICAL LABORATORY SERVICES

References
1. OECD. Principles of good laboratory praxis. Monographs no. 45. Paris, 1992.
2. Haeckel R. Definition of good laboratory services, TQM and (overall) laboratory management,
norms and concepts for the accreditation of laboratories. Abstracts, Medlab 12th IFCC
European Congress of Clinical Chemistry, 17-22 August; Basel, 1997; 15, abstract Ll5.
3. Levey S. Jennings ER. The use of control charts in the clinical laboratory. Am J Clin Pathol
1950;20: 1059-66.
4. Shewhart WA. Economic control of quality of the manufactured product. New York: Van
Nostrand, 1931.
5. Westgard JO, Barry PL, Hunt MR, Groth T. A multi-rule Shewhart chart for quality control
in clinical chemistry. Clin Chem 1981; 27: 493-501.
6. Westgard JO, Barry PL. Cost-effective quality control: managing the quality and productivity
of analytical processes. Washington, DC. AACC Press, 1986.
7. Belk WP, Sunderman Fw. A survey of the accuracy of chemical analyses in clinical laboratories.
Am J Clin Patholl947; 17: 853--61.
8. Deming WE. Out of the crisis. Cambridge, MA: MIT Center for Advanced Engineering
Study, 1982.
9. European standard EN 45001, CEN/CENELEC. The joint European Standards Institutions,
BINIlBN, Brussels, 1989.
10. ISO/IEC guide 25. General requirements for the competence of calibration and testing
laboratories. Geneva: draft 5, 1997.
11. Libeer JC. Total quality management for clinical laboratories: a need or a new fashion? K.lin
Lab Diag 1997; 5: 46--7.
12. EALlECLM. Accreditation for medical laboratories. Guidance on the interpretation of
ISO/lEe. Guide 25. EAL-G25, ECLM-l, 1995.
13. Jansen RTP, Blaton V, Burnett D et af. Essential criteria for quality systems in medical
laboratories. Eur J Clin Chem Clin Biochem 1997; 35: 121-2.
14. ISO 9001. Quality systems - model for quality assurance in design development, production,
installation and servicing. Geneva: ISO, 1993.
15. ISO 9002. Quality systems - model for quality assurance in production, and servicing. Geneva:
ISO, 1993.
16. EN 46001: Quality systems - Medical devices - particular requirements for the application
of EN 29001. CEN/CENELEC, 1993.
17. EN 46002: Quality systems - medical devices - particular requirements for the application of
EN 29002. CEN/CENELEC, 1993.
18. Proposal for a European Parliament and Council directive on in vitro diagnostic medical
devices. Brussels: European Commission, April 1995.
19. Su1klaper, I. References medicales opposables (RMO): les 248 projets pour 1997. OptionlBio
1997; 187: 13-14.
20. Buttner J. Diagnostic validity as a theoretical concept and as a measurable quantity. Eur J
Clin Chem Clin Biochem 1995; 33: AI04-5.
21. Burnett D. Understanding Accreditation in Laboratory Medicine. London: ACB Ventura
Publications, 1996.

19
3
Blood collection and sample
preparation: pre-analytical variation
I. D. WALKER

INTRODUCTION

It is widely accepted that for reliable, accurate and reproducible results,

laboratories must establish standardized analysis systems, but there is less
widespread acceptance of the requirement to standardize specimen collection
and handling procedures.
As soon as a blood vessel is entered for sampling, changes occur - tissue
thromboplastin factor is released, platelets, clotting factors and natural
anticoagulants are activated, active enzymes appear transiently in the circulation
and enzyme-inhibitor complexes form. For some types of assay the aim in collecting
and handling blood specimens must be to preserve the components to be analysed
in a state as close as possible to that in vivo. For other assays this attempted
preservation of the in-vivo state is less necessary unless the assay is specific for, or
discriminates between, molecular forms which may change ex vivo.
Clearly defined procedures for blood sampling and pre-analytical handling
are necessary for all haemostasis laboratories but these procedures may vary
according to whether the laboratory is a research laboratory, or a routine clinical
laboratory accepting specimens from a wide variety of sources. However, even
routine service laboratories must define for their users what the minimum
acceptable standards for sample collection and handling are, and must ensure
that these standards are adhered to.
The blood sampling and handling procedures developed for diagnostic or
for research use must be the same as those used in collecting specimens to
establish reference ranges. The criteria for sampling reference population
individuals and for handling and assaying their blood should be no more nor
less stringent than can be applied in diagnostic or research practice. From a
practical point of view it is highly desirable that the number of separate
21
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
procedures which a laboratory defines for sample collection and handling is
minimized. The ideal would be to unify the procedures for collection and
handling of specimens for all variables, but whilst this may be practicable for
laboratories offering a limited range of coagulation tests, it is unfortunately
unlikely to be possible for services offering tests and assays of a wide range of
haemostasis components, including not only coagulation factors, but also
fibrinolytic variables and platelet tests.

PATIENT REQUIREMENTS

The haematocrit varies according to posture, and differences are noted even
between sitting and lying - the haematocrit increasing significantly as subjects
assume a more erect position 1. The effect which changes in haematocrit make
to the plasma concentrations of haemostasis components are further accentuated
by the effects of dilution with liquid anticoagulant in the sample tube. The
change in haematocrit which follows alteration of position takes around 20-30
min to complete. Most hospital inpatients are venipunctured as they lie in or
on a bed, but most outpatients will have their blood sampled whilst they are
sitting, and often within less than 20 min of having walked into the consulting
room. Whilst it is impractical to suggest that, to gain uniformity, all patients
should be sampled after lying down for a period of at least 30 min - for some
tests this may be preferable or even mandatory, and in some studies it may be
essential to make a definite choice to sample participants in a particular posture
- either sitting or lying.
Some haemostasis components show distinct circadian variation and
although, for most routine diagnostic purposes, blood specimens are collected
at random times of day, for those variables which show a marked circadian
variation, standardizing the timing of blood collection may be necessary. In
some instances it may also be important to standardize blood sampling to a
particular season or a particular point in the menstrual cycle.
Frequently where laboratories standardize the timing of specimen collection
they seek to standardize other factors which may also influence haemostasis
variables - for example food and alcohol intake, caffeine consumption, cigarette
smoking and stress - all of which have documented effects on haemostasis,
particularly on factor VIlle and von Willebrand factor and on fibrinolytic
system components.
Patients are sometimes given detailed instructions in an effort to avoid the
effects of smoking, food intake, caffeine or alcohol on test results. However,
unless the intent is to study the effect of these 'lifestyle'variables, over-zealous
preparation of patients may result in considerable deviation from the individual's
normal habits and risk producing unrepresentative results or, worse, results
which have been influenced by the effects of abstention. It is important in
preparing patients for venipuncture that care is taken to ensure that the sampling
conditions meet the requirements of the diagnostic or research questions to be
answered. Once it has been decided which clinical variables must either
be eliminated or accounted for, and how, most simply, this objective is to be
achieved, then it is a relatively simple step to define the patient preparation
22
 BLOOD COLLECTION AND SAMPLE PREPARATION

required. In general, in the interests of practicality, the fewer requirements

which are placed on the patient and the sampling conditions the better.
For most routine diagnostic tests of haemostasis, few restrictions are put on
the patient prior to blood sampling, although it may be relevant to take into
account certain aspects of the patient's lifestyle in interpreting results. However,
for investigations of fibrinolysis and perhaps also for measurement of the
components of the factor VIII complex (factor VIlle and von Willebrand
factor) more restrictive preparation is often recommended - sampling at a
standardized time of day (usually between 8 a.m. and 10 a.m.), after an overnight
fast or a standardized light meal (to avoid any possible influence on the results
of blood volume distribution changes during digestion), avoidance of alcohol
for 18-24 h and abstention from smoking and caffeine for at least 1 h prior to
venipuncture. For these investigations venipuncture is usually performed with
the patient in a specified posture (either sitting or lying) and after a 20--30 min
period of relaxation in the chosen posture. For research purposes very detailed
instructions may be given to patients, which may be acceptable and practical
for the limited purpose of the study.
However flexible or restrictive the patient sampling requirements are, the
basic essential is that for each purpose the sampling procedure is standardized,
and phlebotomists, clinicians and researchers performing venipunctures know
the relevant standards and understand the rationale behind them.

VENIPUNCTURE TECHNIQUE

In general it is recommended that blood specimens for haemostasis tests are

collected from the upper limb, from the antecubital fossa. Differences may be
observed in results obtained from specimens collected from upper and lower
limbs and in the upper limb from the antecubital fossa and the hand. A number
of techniques are possible - the traditional two-syringe and needle technique,
evacuated sample tube and needle, butterfly cannula and syringe or an indwelling
cannula and syringe.
In general indwelling cannulae are least acceptable, and for haemostasis tests
should be avoided if possible. Samples collected through indwelling catheters
may be contaminated by exogenous heparin. If blood must be drawn through
an indwelling catheter, the line should be flushed with saline and the first 5 ml
of blood discarded or used for other laboratory tests. If the result of any
haemostasis test is unexpectedly abnormal, ideally a new specimen should be
obtained from a different site. If this is not possible then the test should be
repeated after treating the sample with a resin which removes heparin. Blood
specimens for laboratory monitoring of heparin treatment must not be collected
through an indwelling catheter. Even butterfly cannulae have been shown to
be unsatisfactory for the collection of blood specimens for some haemostasis
tests. They may, however, be the only possible method of obtaining blood from
small children, or even adults with 'difficult' veins.
The ideal needle gauge for coagulation tests ranges from 21 to 19. For
paediatric patients 23-gauge needles may have to be used. If butterfly needles
are used, they should be of the same gauge. The sample for haemostasis tests
23
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
should be part of a blood specimen which is in total no more than 25 ml if a
20-gauge needle is used, or 50 ml if a 19-9auge needle is used2 •
Many clinics and large studies nowadays use evacuated tubes because in
general they are safer, and have the advantage of ensuring that the relative
volumes of blood and anticoagulant are correct. However, even evacuated
tubes may be incorrectly filled if their caps have been loosened and the vacuum
disturbed. Theoretically, simple syringes and needles cause less activation of
the blood specimen, but they demand care in ensuring that an accurate volume
of blood is dispensed into the sample tube and that the relative volumes of
anticoagulant and blood are correct.
The patient's arm should be supported in a downward position. Prolonged
tourniquet application and resulting stasis should be avoided. If a tourniquet
is required then it should be applied for the minimum period of time possible,
and for no more than 1 min. Expert phlebotomists can obtain blood specimens
in a much shorter time, and in patients with good veins it may be possible to
obtain the specimen without the use of a tourniquet. It has been recommended3
that the maximal pressure for the tourniquet should not exceed + 2 kPa. Good
'clean' entry into the vein is required and the needle should not be moved once
it has entered the vein. Blood must be drawn gently without any forced pressure.
If an evacuated tube is being used the tubes must be held steady until the
vacuum is exhausted and blood flow ceases before gently changing the evacuated
tube. In general as soon as blood enters the syringe or the first collection tube
the tourniquet should be released.
The first few millilitres may be contaminated with tissue thromboplastin
factor and optimally should not be used for haemostasis tests. Usually the first
5 ml of blood collected from an adult are discarded or at least not used for
haemostasis tests. Ideally, if syringes and needles are being used, the syringe
should be changed after the first 5 ml have been collected, and if evacuated
tubes are being used the first tube should not be used for haemostasis tests2 •
No significant differences, however, were observed in prothrombin times (PT)
or activated partial thromboplastin times (APTT) of normal blood samples
between the first and second tubes collected using an evacuated system4 , but
in samples collected from heparinized patients the APTT of the first tube may
be shorter than that of the second tube 5 • Where a patient requires only an INR
(international normalized ratio) for monitoring of oral anticoagulant therapy
it is not necessary to draw a preliminary sample4 • For APTT and coagulation
factor assays ideally a preliminary tube should be collected and not used for
haemostasis tests, but if no specimen for other tests is necessary, and providing
the venipuncture is simply and cleanly performed and the blood flows easily,
it may be possible to dispense with the requirement for a preliminary tube.
If a syringe has been used to collect the blood specimen the syringe should
be emptied by placing its tip against the inside of the sample tube and allowing
the blood to run down the side of the tube so that frothing and excessive
turbulence is avoided. After filling, the tubes must be mixed immediately, and
it is suggested that to do this the tubes are inverted carefully three to five times.

24
 BLOOD COLLECTION AND SAMPLE PREPARATION

BLOOD SAMPLE TUBES AND ANTICOAGULANTS

Blood specimens must be collected and stored in containers made from

non-reactive materials such as polypropylene or siliconized glass. A variety of
anticoagulants including ethylenediamine tetra-acetic acid (EDTA) and citrate
have been used, but not all analyses are possible on specimens anticoagulated
with all types of anticoagulant, and different anticoagulants may give signifi-
cantly different results in particular tests or assays. The most widely recom-
mended anticoagulant is trisodium citrate. The National Committee for Clinical
Laboratory Standards (NCCLS) recommends citrate concentrations of 0.109
or 0.129 moUL (3.2% or 3.8~W. In Europe the European Committee for Clinical
Laboratory Standards (ECCLS) recommends citrate concentrates of 0.100-
0.120 moUL 6. Whilst there is widespread agreement that an international
standard of sample tube anticoagulant should be adopted, there is no consensus
at present on what that standard should be.
Non-buffered sodium citrate solution has a pH which is greater than 8.0
and if it is stored in siliconized glass tubes the silicon layer may become unstable
due to changes in the silica skeleton of the glass7 • This can be avoided by using
buffered sodium citrate. At high pH, labile clotting factors such as factor V
and factor VIII deteriorate, thus commercial manufacturers of sample tubes
for haemostasis tests in general use a mixture of sodium citrate and citric acid
with a pH of 5.8.
It is important that the relative volumes of plasma and citrate are standardized.
All of the calcium which is to be bound by the citrate is in the plasma. Thus if
the patient's haematocrit is very high, and the usual relative volumes of blood
to citrate (9:1) are employed, the plasma will be over-citrated and the PT and
APTT will be prolonged. In polycythaemic adults, and in neonates, wherever
possible, the actual haematocrit should be determined and the volume of citrate
solution added to the specimen tube adjusted8 • In these patients the specimen
must be collected by syringe since in evacuated tubes the blood:citrate ratio is
fixed.
For some tests and assays it is desirable to stabilize the platelets in the blood
specimen as quickly as possible after collection. Stabilyte® (Biopool) tubes,
which contain citrate at reduced pH, were developed to avoid in-vitro activation
of plasminogen at elevated tissue plasminogen activator (t-PA) levels, and have
been used successfully for assays of t-PA activity (Chapter 24) and antigen
(Chapter 25) and for plasminogen activator inhibitor-l antigen (Chapter 26).
However, citrate at reduced pH affects the results of some assays, and results
from tests on blood specimens collected into reduced pH citrate should not be
compared with results of specimens collected into normal citrate9 .
It may be preferable to use citrate with supplements of platelet-stabilizing
agents. For some purposes 'home-made' mixtures such as citrate with prost-
aglandin El and theophylline may be suitable but commercially available tubes
(e.g. CTAD®) containing citrate, theophylline, adenosine and dipyridamole
are widely available. Problems with inter-batch variation in some of these
commercially produced tubes have been reported 1O.

25
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
SPECIMEN PROCESSING

All specimens for haemostasis tests must be labelled with the patient's full name
and identification details. The labels should be placed on the sample tubes at
the time of sample collection and in the patient's presence.
Even small fibrin clots in the specimen render it unacceptable for investigation.
Specimens which have less than 90% expected fill of the collection tubes must
be rejected and another specimen requested. Haemolysis may affect the APTT
result but apparently not the PT result4 . After centrifugation the plasma should
be inspected for haemolysis, and haemolysed plasma should not be used for
APTT tests. Coagulometers using an optical detector may have problems with
end-point determinations on samples which are icteric or lipaemic, and alter-
native methods should be used.

Specimen transport

As soon as the specimen has been withdrawn from the patient's vein it is
beginning to deteriorate. It must therefore be transported as quickly as possible
to the laboratory. During transport specimens should be handled gently to
minimize trauma - in particular vibrational trauma - which may cause hae-
molysis and activation of coagulation. Sample tubes must remain tightly capped
to prevent loss of carbon dioxide and a subsequent rise in pH. Ideally tubes
should be held upright during transport.
There is conflicting advice about the temperature at which specimens should
be held prior to centrifugation. Previously it was recommended that blood
specimens for haemostasis tests should be immediately placed on melting ice
or the equivalent ll . Whilst this is certainly possible in some circumstances it
is not always practical, and indeed in view of the well-documented effect of
cold activation of factor VIIC it may be preferable to maintain blood specimens
at room temperature (15-20 0c) if factor VIIC activity assay (Chapter 10) is
required or for a PT I2 (Chapter 6).
Koepke et al. 13 demonstrated that there is no significant change in either PT
or APTT in specimens stored at room temperature for 6 h. However, the APTT
has been shown to shorten during storage at room temperature in specimens
taken from patients receiving heparin 14. This effect can be minimized by collecting
blood into CTAD to inhibit the release of platelet factor 4, a heparin ant-
agonist l5 .
The rate of factor VIIIC deterioration at room temperature is similar to
that at 4 °C I6 . Specimens for factor VIIIC assay (Chapter 11) therefore do not
need to be maintained at 4°C, but it is im~ortant that specimens for VIIIC
assay are analysed within 2 h of collection 1 •

Centrifugation and plasma separation

Platelets contain a number of blood coagulation factors, platelet factor 3 and

platelet factor 4. It is therefore important to consider whether or not residual
26
 BLOOD COLLECTION AND SAMPLE PREPARATION

platelets in the plasma sample following centrifugation may interfere with the
planned test or assay, and to control and standardize the speed and duration
of centrifugation according to the particular requirements of each test to be
performed.
For most tests and assays it is suggested2 that the minimum centrifugation
criteria should be 2500g for 15 min. For some tests it is essential to ensure that
as far as possible the plasma sample is rlatelet free. For this purpose a double
centrifugation may be recommended 1 • The centrifuge brake should not be
used. After centrifugation plasma must be drawn off carefully using a non-
contact pipette. Great care must be taken to avoid disturbing the platelet layer.
Deterioration of clotting factors can be minimized by rapid processing and
double centrifugation to eliminate most cells 17 . Samples which will not be
tested immediately but which are to be stored frozen may be improved by double
centrifugation.

Sample storage
The maximum acceptable time interval between venipuncture and testing of
the sample will depend on the temperature maintained during transport and
storage. NeeLS recommend the following: 2 h for samples stored at 22-24 DC,
4 h for samples stored at 2-4 DC, 2 weeks for samples stored at -20 DC and 6
months for samples rapidly frozen and stored at -70 De 2• In general clotting
factors in frozen plasma will be more stable at lower temperatures. Frozen
samples should be stored in small aliquots in well-identified containers which
are tightly capped. Ideally samples which will not be tested fresh should be
snap-frozen. If liquid nitrogen is available this should be the method of choice
for rapid freezing, but the process can also be achieved by using a refrigerated
bath at -50 DC or a mixture of dry ice and acetone (provided the containers
are resistant to acetone). If none of these methods is available plasma aliquots
may be placed in the deep-freeze at the lowest possible temperature available
to ensure rapid cooling and freezing.
The temperature during frozen storage must be carefully controlled. Frozen
plasma samples should be thawed rapidly at 37 DC according to a standardized
protocol and tested immediately after thawing before the sample warms. If
testing cannot be performed immediately the sample may be held for a maximum
of 2 h at 4 DC until tested2 •

PREPARATION OF BLOOD FOR DNA ANALYSIS

The availability of genotype confirmation of the diagnosis of haemostasis

abnormality is expanding rapidly, and frequently haemostasis laboratories are
responsible for the reception and initial handling of blood specimens for this
purpose.
Samples for DNA analysis may be prepared from either citrate stabilized or
EDTA specimens. After centrifugation of a 10 ml specimen the plasma is
removed, leaving approximately 500 ~ of plasma in the tube used for blood
27
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
collection. This 500 J.11 of plasma is then mixed carefully with the buffy coat
(the layer of leukocytes and platelets on top of the red cells) and this mixed
top layer (approximately 500 J.Ll) is subsequently transferred to a cryo tube and
frozen at -70°C to -80 °C.

References
1. Leppanen EA, Grasbeck R. Experimental basis of standardised specimen collection effect
of posture on blood picture. Eur J Haemato11988; 40: 222-6.
2. NCCLS. Collection, Transport, and Processing of Blood Specimens for Coagulation Testing
and Performance of Coagulation Assays. Villanova, PA: NCCLS Document H21-A2, Vol.
11, No. 23, 1991; 1.
3. KIuft C, Meijer P. Update 1996: Blood collection and handling procedures for assessment of
plasminogen activators and inhibitors. Fibrinolysis 1996; 10 (SuppL 2): 171-9.
4. Gilmer PRo Preanalytical variables in blood coagulation testing. In Triplett DA, ed.
Standardisation of coagulation Assays: an overview. Skoki Illinois: College of American
Pathologists, 1982; 1-8.
5. van Putten JJ, van de Ruit M, Beunis M, Hemker He. Heparin neutralisation during collection
and processing of blood inhibited by pyridoxal5'-phosphate. Haemostasis 1984; 14: 253-61.
6. ECCLS. Standard for Specimen Collection Part 1: Blood Containers, Vol. 3, 1983: 1.
7. van den Besselaar AMHP, van Hallem-Vissei LP, Loeliger EA. The use of evacuated tubes
for blood collection in oral anticoagulant control. Thromb Haemostas 1983; 50: 676-7.
8. Ingram GIC, Brozovic M, Slater NGP. Bleeding Disorders Investigation and Management,
2nd edn. Oxford: Blackwell Scientific Publications 1982; 244-5.
9. Meijer P, van der Ham F, KIuft C. The use of Stabilyte® plasma may cause changes in pH in
the assay of some fibrinolysis analytes and might affect results. Fibrinolysis 1996; 10 (SuppL
2): 155-7.
10. Polak B, Barro C, Mossuz P, Pernod G. Inadequate quality of a blood collection tube containing
an anticoagulant/platelet inhibitor mix. Thromb Haemostas 1997; 77: 1035-6.
11. NCCLS. Tentative guidelines for the standardised collection, transport and preparation of
blood specimens for coagulation testing and performance of coagulation assays. Villanova,
PA: NCCLS Publication, Vol. 2, No.4, 1982, 103-28.
12. Thomson 1M. Specimen collection for blood coagulation testing. In Koepke JA, ed. Laboratory
Haematology. New York: Churchill Livingstone, 1984; 833-63.
13. Koepke JE, Rodgers JL, Ollivier MJ. Preinstrumental variables in coagulation testing. Am J
Clin Patho11975; 64: 591-6.
14. van den Besselaar AMHP, Meeuwisse-Braun J, Jansen-Gruter R, Bertina RM. Monitoring
heparin therapy by activated partial thromboplastin time. The effect of preanalytical condi-
tions. Thromb Haemostas 1987; 57: 226-31.
15. van den Besselaar AMHP, Bertina RM. Standardisation and quality control in blood
coagulation assays. In: Lewis SR, Verwilghen RL, eds. Quality assurance in haematology.
London: Bailliere Tindall, 1988; 119-50.
16. Rotbalt F, Tuddenham EGD. Immunologic studies of factor VIII coagulant activity (VIIIC)
assays based on a haemophilic and an acquired antibody to VIlle. Thromb Res 1981; 21:
431-45.
17. DIN. Einstufenmethode zur Bestinunung der Faktor VIII-gerinnungsaktivitiit (FVIIIC).
Deutsches Institut fUr Normung 58909, Teill, 2. Berlin: Beuth Verlag, 1987.

28
4
Quality assessment of haemostatic
assays and external quality
assessment schemes
T. A. L. WOODS, S. KITCHEN and F. E. PRESTON

INTRODUCTION

The coagulation laboratory has a vital role in the diagnosis and management
of patients with familial and acquired haemorrhagic and thrombotic disorders.
Wherever possible the laboratory methods employed must reflect the state of
the art, and the results generated, including locally derived reference values,
should be reliable, reproducible and unambiguous. These issues are of particular
importance when investigations are undertaken for the diagnosis of a possible
familial disorder. When an error occurs, there is possibility of misdiagnosis
and, irrespective of whether this causes an individual to be misdiagnosed as
either having or not having a familial defect, the clinical consequences of such
a mistake are likely to be serious. Within recent years the workload and scope
of the routine laboratory has increased substantially, and this has been accomp-
anied by the introduction of automated equipment employing a variety of
different technologies. The potential for laboratory error is therefore considerable,
and monitoring of laboratory performance through a programme of quality
assurance is an essential laboratory requirement.

CONCEPT OF QUALITY ASSURANCE

Quality assurance (QA) is an overall term that may be used to describe all
measures that are taken to ensure the reliability of laboratory testing and
reporting. This will range from the choice of test, the collection of a valid
sample from the patient, analysis of the specimen and the recording of results
29
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
in a timely and accurate manner, through to interpretation of the results, where
appropriate, and communication of these results to the referring clinicians.
Internal quality control (IQc) and external quality assessment (EQA) are
two distinct, yet complementary, components of a laboratory QA programme.
IQC is used to establish whether a series of techniques and procedures are
performing consistently over a period of time. It is therefore deployed to ensure
day-to-day laboratory consistency. EQA is used to identify the degree of agree-
ment between one laboratory's results and those obtained by other centres. In
large EQA schemes retrospective analysis of results obtained by participating
laboratories permits the identification not only of poor individual laboratory
performance, but also those reagents and methods that produce unreliable or
misleading results.

INTERNAL QUALITY CONTROL

IQC is used to establish whether a series of techniques and procedures are

performing consistently over a period of time. The expression 'quality control'
is commonly used to describe the set of procedures used to check that the
results of laboratory investigations are reliable enough to be released to assist
clinical decision-making, monitoring of therapy and diagnosis of haemostatic
abnormalities. Quality control procedures should be applied in a way which
ensures immediate and constant control of result generation. Within a laboratory
setting the quality of results obtained is influenced by many factors, which
include the following: appropriate sample collection and handling; selection
of suitable techniques and maintenance of an up-to-date manual of standard
operational procedures; use of reliable reagents and reference materials; selection
of suitable automation and adequate maintenance; adequate records and report-
ing system for results. In addition, the quality of results obtained in routine
practice is highly dependent on the selection, training and motivation of an
appropriate complement of suitable personnel.
IQC is particularly useful to identify the degree of precision of a particular
technique, precision being the degree of agreement between repeat measure-
ments on one sample. It is important to recognize that a precise technique is
not necessarily accurate, accuracy being a measure of the closeness of an esti-
mated value to the true value.

QUALITY CONTROL MATERIALS

In order to assess the precision of a particular method it is necessary to perform

repeated analyses of aliquots of the same sample. It is important to include
QC samples with normal and abnormal values, to ensure that a method is
under control at different levels of a particular analyte, since relatively minor
changes in an analytical process may be more apparent when testing an abnormal
control. The control material should be similar in properties to test samples
and be analysed concurrently. Quality control materials of human origin are
more likely to closely resemble human test samples. All vials or aliquots of the
30
 HAEMOSTATIC ASSAYS AND QUALITY ASSESSMENT
control material should be practically identical, so that any variation in test
results is not a consequence of vial-to-vial variation. The QC material should
also be stable for its intended period of use. In respect of haemostatic tests and
assays, plasma samples have to be deep frozen (preferably at -40 °C or lower)
or lyophilized in order to ensure adequate stability for use as QC material. For
reconstitution of lyophilised samples it is important to use distilled water with
pH 6.8-7.2 and to allow at least 5 min for reconstitution. If commercial QC
material is used, this should bereconstitutedaccordingtomanufacturersinstruc-
tions using an accurate pipetting system. If deep-frozen QC material is used
this should be thawed rapidly at 37°C for 5 min. In the selection of QC material
the risk of transmission of blood-borne viruses should be considered, and
material should be negative for tests to detect the presence of human immuno-
deficiency virus (HIY), hepatitis C (HCy) and hepatitis B virus.
At least one QC material should be included with each group of screening
tests or assays. Where continuous sampling is in operation a QC sample should
be included after every 20 test samples. For screening tests it may be most appro-
priate to include a normal QC in this way, and to test abnormal QC materials
once per day or shift, or when doubt exists about whether a method is under
control. For assays of individual components the level of analyte in the QC
sample should reflect the samples under investigation, for example a QC material
with a reduced level should be included with tests used for the diagnosis and
monitoring of congenital deficiency states associated with bleeding or thrombosis.
A QC sample with an elevated level should be included with tests used to
identify the presence of raised levels (for example of factor VII:C or fibrinogen)
in epidemiological or other studies.
In all cases the control material must be treated exactly like test samples if
possible. Since some variation will necessarily occur as a result of biological,
technical and analytical variation, each QC result should be recorded and
assessed against the range considered to be acceptable as described below.

ACCEPTABLE LIMITS OF VARIATION

The variation of clotting factor results is influenced by method errors, between

batch variation, within-person variation and between-person variation!. The
purpose of IQC is to identify the degree of analytical variation. This requires
suitable IQC materials. For commercial IQC samples manufacturers often
provide a target range of acceptable values. In the case of screening tests, and
occasionally assays, the results obtained will normally be dependent on the
reagents and end-point detection system used to perform the tests. The target
range must take account of these effects. Where a target range is not available
for a particular technique, this can be established locally. The IQC material is
tested repeatedly (minimum 10 times) on different days when the method is
known to be under control (as indicated for example by within-target results
on an alternative QC material). The mean and standard deviation (SO) of
these results are then calculated. The SO is the square root of the sum of J2
divided by n - 1, where d is the difference of individual results from the mean
and n is the number of determinations. The SO is a measure of the spread of
31
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
results; the larger the SD, the greater is the spread of results. Another important
parameter is the coefficient of variation (CV), which is the SD expressed as a
=
percentage of the mean (CV SD divided by mean, multiplied by 100%). The
CV of results of prothrombin time and activated partial thromboplastin times
of a QC sample by one particular technique was 6%2 for assays on different
days. For many coagulometers and techniques, between-assay CV values of
<5% are commonly obtained. This is also the case for assay techniques such
as amidolytic assays of antithrombin, which are commonly more precise than
assays of clotting factors 3. For assays such as factor VIII:C and IX, between-day
CV values of less than 10% should be attainable.
In most cases repeated results obtained for an IQC sample will show a
normal (Gaussian) distribution. It is common practice to set the target range
for IQC results as the mean ± 2 SD, since this should include 95% of values.
Individual results should be recorded on a chart which identifies the target
range. An example is shown in Figure 4.1. Any future values which lie within
these limits are considered acceptable. Results outside this range indicate that
the QC material has deteriorated or been handled incorrectly, or that the method
is not properly controlled. Repeat testing of further QC material will then
differentiate between these two possibilities, further out-of-limits results
confirming that the test system is out of control. Medium- or long-term drift
in QC test results, for example as a consequence of instrument or reagent
deterioration or change, will become apparent by scrutiny of cumulative record
charts.

EXTERNAL QUALITY ASSESSMENT

The ultimate objective of EQA is the improvement of health care through

improved laboratory performance. In recognition of this concept EQA in blood
coagulation have been established in a number of different countries including
Denmark, France, Italy, the Netherlands, Switzerland, UK and the USA. In
all of these schemes, samples are distributed to participating centres on a per-
iodic basis. However, the individual schemes vary considerably in respect of
the number of analytes offered, the number of participants, the source of
samples and the frequency of distribution.
The primary function of EQA is proficiency testing of individual laboratory
testing. This should embrace all aspects of coagulation, and analytes offered
should represent the current 'state of the art'. In conjunction with their primary
function, the larger EQA schemes can also provide information concerning
the relative performance of analytical procedures, including the method principle,
reagents and instruments. EQA schemes may also identify method-dependent
variability of results. Examples of this, identified by UK National External
Quality Assessment Scheme (UKNEQAS), are the method-related differences
in functional protein C and VWF:RiCof assays.
Continued participation in EQA schemes has been clearly linked to improved
laboratory performance. This has been seen not only in the overall performance,
evidenced by a reduction of the variability of results between laboratories, but
also in respect of individual laboratories.
32
 HAEMOSTATIC ASSAYS AND QUALITY ASSESSMENT

90~-------------------------------------,

80 I- ..•...............................•.•.......

..

~
-s:
~
ctI 70 - .
()

:>
L-
o
~
LL

60 I- .................................. - . - ... - - ... .

Date of test
Figure 4.1 Results of factor VIII:C assays on an internal quality control sample assayed on
different days_ Each point is a different assay on the same material. The solid lines represent the
mean and 2 standard deviations of 20 assays on this material, considered to represent the limits
of acceptable results

33
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Increasingly effective EQA scheme designs have evolved through the combined
effects of experience gained over many years, and the continuing and necessary
interactions that take place between participants and the scheme organizer. A
number of criteria have emerged as the fundamental building blocks of a success-
ful scheme. These may be summarized as:

1. Adequate topical data; for example, sufficient data returns for statistical
analyses. This will be influenced by: (a) adequate number of participants;
(b) prompt return of results from participants; (c) sufficient number of
analytes in each distribution; (d) frequency of distribution.
2. Effective communication of performance data, through: (a) well-composed,
informative and easily understood reports; (b) sequential historical grades.
3. Relevant structure for performance assessment: (a) samples that are stable
and consistent; (b) matrix that is compatible with clinical samples.

The assessment of individual laboratory performance is an essential component

of all EQA schemes. A number of different methods of assessment have been
adopted by different scheme organizers, and for quantitative assays individual
laboratory results may be assessed against so-called true values or consensus
values. The former are derived from results obtained by expert reference labora-
tories using reference methods under highly standardized conditions. For EQA
coagulation schemes laboratory performance is more frequently determined
by reference to consensus values. For many coagulation assays there are
undoubted reagent-related differences, and it is therefore important that, when
consensus values are used for performance analysis, this should relate to results
obtained using the same reagent. In UKNEQAS, individual performance
analysis for specific assays is assessed against either the reagent median, where
the number of users of a reagent is greater than or equal to ten, or against the
overall median where the reagent group has less than ten users.
Whichever method is adopted by the EQA scheme director for laboratory
performance analysis, it is vitally important that individual laboratories are
aware of the criteria for the assessment of their results, and also the mathematical
methods deployed for the statistical analysis.
The educational aspects of EQA schemes require particular emphasis. There
are many possible reasons why a laboratory might produce results that are
considered unsatisfactory, or outside consensus and, whilst the cause for this
might be immediately apparent, it is our experience that the identification of
the underlying problem is not always simple. When a laboratory is persistently
outside consensus then it is the responsibility of the scheme director to direct
the problem to the attention of the head of the laboratory, and to offer any
assistance that may be required. It is important to stress that in this respect the
role of the scheme director is supportive and educational rather than punitive.
Larger schemes are able to identify performance problems which relate specifi-
cally to reagent differences or differences of methodology. With respect to differ-
ences in results arising out of reagent differences the scheme director should draw
the attention of participants to these anomalies, but where consensus values are
used for data analysis, he/she should refrain from giving a value judgement as to
which reagent is preferred.
34
 HAEMOSTATIC ASSAYS AND QUALITY ASSESSMENT

STANDARDIZATION OF EQA SCHEMES

It is essential that participants in EQA schemes should have complete confidence

in their efficiency and effectiveness. The International Council for Standardi-
zation in Haematology (ICSH) has prepared guidelines for the organization
and management of EQA schemes using proficiency testing. These guidelines
are intended to help maintain a meaningful standard in the organization of
schemes and to harmonize the way in which they function. They include a
number of important principles and criteria which are detailed below.
Surveys should be sufficiently frequent to make sequential performance
analysis meaningful in order to identify participants whose performance is
persistently unsatisfactory (outside consensus) as soon as possible. Tests for
haemostasis and thrombosis need to be distributed at least quarterly; more
frequently where the clinical importance and reliability of the analytical methods
are of concern. In such cases two specimens per distribution may be required,
with values at diagnostically critical levels.
To ensure that EQA relates to clinical practice, the survey samples should
simulate natural samples as closely as possible, with instructions being given
to participants regarding sample handling, and for laboratory testing to be
carried out in the same way as for routine specimens.
The material used in surveys should be stable, at least until the closing date of
the survey. The survey specimens must be negative for HIV and HCV antibodies
and for hepatitis B surface antigen, and must be labelled in accordance with
national guidelines for packaging and transportation for biological materials.
Data processing must be as rapid as possible, with prompt returns to participants.
Total confidentiality is an important feature of all EQA schemes, and infor-
mation regarding individual laboratory performance is not divulged to anyone
other than the nominated head of the department. Any information on an
individual's results to a third party (e.g. licensing authority) would be provided
by the participant, and this must not be the responsibility or duty of the NEQAS
director or organizer. EQAS must be professionally led and should function
independently of government health authorities.
Industry may provide a useful service by organizing a scheme for users of
their apparatus, but a national scheme should always be independent of industry.
Above all, EQAS itself must not be a licensing authority or a policing body.
As an indication of the structure of a national EQA scheme for haemostasis
and thrombosis, an outline of the UKNEQAS for blood coagulation follows.

Analytes covered by the scheme

International Normalized Ratio (INR) based on: (a) Quick's one-stage method,
(b) capillary reagent method.
Prothrombin time (diagnosis).
Activated partial thromboplastin time (APTT).
Thrombin time.
35
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

Heparin dosage assessment.

Fibrinogen determination.
Fibrin(ogen) degradation products (FDP)/D-dimer.
Assays of factors II, V, VII, VIII, IX, X, XI, XII.
Factor XIII screen test/assay.
Lupus anticoagulant detection.
Factor VIII inhibitor determination.
Assays of: vWF:Ag and vWF:RiCof, antithrombin antigen and activity, protein
C antigen and activity, protein S total and free antigen, protein S activity, plas-
minogen activity.
Activated protein C resistance.
Material distributed - human, single-donor, plasmapheresed, freeze-dried
plasma.
Number of distributions: six annually.
Performance assessment:
1. International normalized ratio (INR). Results assessed against consensus
median INR obtained with a particular thromboplastin reagent. A per-
formance outside consensus is defined as a deviation exceeding 15%from con-
sensus median.
2. Specific assays. Performance analysis is performed by the ranked grading
analysis. The general reference point is the overall consensus median.
Participants' results are compared against this figure by calculating percentage
deviations. Outside consensus is defined as two consecutive exercises for any
given test. Persistently outside consensus is defined as two consecutive outside
consensus performances for any particular test.
Participants receive their individual reports no later than 2 weeks after the
closing date for any survey. At a later date, dependent on whether reports from
one or two surveys are published together, participants receive a detailed
analysis, with an accompanying commentary, of the results arising out of each
survey.

References
1. Thompson SG. Martin JC, Meade Tw. Sources of variability in coagulation factor assays.
Thromb Haemostas 1987; 58: 1073-77.
2. Dot D, Miro J, Fuentes-Arderiu X. Within-subject and between-subject biological variation
of the prothrombin time and activated partial thromboplastin time. Ann Clin Biochem 1992;
29: 422-5.
3. Costongs GMPJ, Bas BM, Janson PCW Short-term and long-term intra-individual varia-
tions and critical differences of coagulation parameters. 1. Clin Chern Biochem 1985; 23:
405-10.

36
5
Activated partial thromboplastin time
(APTI)
L.POLLER

INTRODUCTION

The term partial thromboplastin is used to distinguish the reagent from that
used in the prothrombin time, since the APTT reagent lacks the apoprotein
component of the complete tissue thromboplastin. The APTT is the main test
for screening for intrinsic clotting defects including haemophilia. It is also used
for detection of lupus anticoagulant and for laboratory monitoring of heparin
administration. The presence of an activator in the test system, to accelerate
the PIT test by effecting maximum activation, also increases its precision and
reproducibility by eliminating the variable effects of contact with glass surfaces.

APTT PHOSPHOLIPIDS

Thepartialthromboplastinconsistsofthephospholipidcomponentofthrombo-
plastin and is prepared from animal tissue or from vegetable sources. The
phospholipid acts as a platelet substitute in the intrinsic system. The lipid
composition of different APIT reagents, however, varies considerably. The
total concentrations of phospholipid and fatty acid in some widely used APTT
reagents have been shown to differ by as much as 300 times!. These discrepancies
markedly affect responses to coagulation defects and inhibitors of coagulation.
The requirements for the phospholipids in the test system may also vary
according to the nature of the clotting defect being measured. For example,
the concentration of negatively charged phospholipids, e.g. phosphatidyl serine,
has been shown to be criticae.

37
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
OTHER COMPONENTS OF THE APTT TEST

Factors which affect the clotting response include the type of activator, length
of incubation with the plasma and the presence of buffers 3 •4 • Particulate
activators include kaolin, celite and micronized silica, whereas a commonly
used activator, ellagic acid, is non-particulate. The amount of activator present
in the various commercial techniques, and the length of incubation time
employed with the plasma, show considerable variation. The effectiveness of
an activator in the APTT is governed by many considerations, e.g. its con-
centration, the incubation time and the composition of the phospholipid. It is
the combination of the activator with the other components which appears to
determine the reliability of the tese. The trend is to use less opaque activation
to avoid interference with newer types of coagulometers. The use of different
types of coagulometer can also have a considerable effect on the clotting time6 .

SENSITIVITY OF THE APTT

A reliable APTT reagent should be sufficiently sensitive to record an abnormal

result when the level of any single or combined intrinsic clotting factor deficiency
is reduced to a level which may cause spontaneous bleeding, or haemorrhage
following a haemostatic challenge. Proctor and Rapapore stated that the APTT
should be able to detect factor deficiencies of 30% or less. It should also be
sufficiently responsive to low concentrations of heparin whilst giving a linear
response to graded concentrations of heparin, spanning a clinically relevant
range of 0.05-0.5 units/ml of heparin. The importance of assessing heparin
response to patients' heparin-treated samples rather than normal plasma 'spiked'
with heparin was emphasized in the ISTHlICSH Study8 as the response to
heparin-treated patients may be markedly less with some APTT reagents com-
pared with the same concentration of heparin added in vitro, particularly after
a recent thrombotic episode. The variable effects on heparin response of APTT
reagents from coagulometers used to perform the test was observed in the
ISTH/ICSH study and a UK survey from the National External Quality
Assessment Scheme9 . The sensitivity to lupus-like anticoagulant of anAPTT
system should be established in comparison with a responsive formulation of
the viper venom test.

CLINICAL USE

The main use of the APTT is for the screening of coagulation defects and the
presence of inhibitors. The test is prolonged by deficiencies of factors VIII,
IX, X, XI and XII and defects of the contact phase, e.g. prekallikrein, high
molecular weight kininogen. It also may be prolonged by gross defects of
factors II, V and fibrinogen. With reliable APTT systems, specific and
non-specific inhibitors of intrinsic clotting factors are detected. The degree of
abnormality depends upon the responsiveness of a particular APTT method
to a specific defect. When used as a screening test for lupus anticoagulant (LA)
38
 ACTIVATED PARTIAL THROMBOPLASTIN TIME

(Chapter 19) the detection rate is greatly influenced by the concentration and
type of phospholipid content of the reagent. Reagents containing the highest
concentration of phospholipid tend to be less responsive to LA. The APTT is
also the most widely used method for the laboratory monitoring of heparin
administration I 0, II. Although other more specific techniques have been advo-
cated, e.g. anti-Xa assays, the APTT has been almost universally preferred as
it is regarded as a global test of coagulation which assesses the overall effect
of heparin on clotting. The APTT may also be of value in case of unexpected
bleeding during anticoagulant therapy as there may be disproportionate
depression of factor IX. The APTT is also used as in the ECAT study as an
overall assessment of the intrinsic clotting system. Accelerated APTT have
been reported after operations, oral contraceptive administration and withdrawal
of oral anticoagulation.

STANDARDIZATION OF HEPARIN MONITORING

The need for standardization of the APTT has been demonstrated in many
reports, partic~arly with regar~ to th.e ~de~pread a.ppli~a~ion of the test in
laboratory momtonng of hepann admimstratlOn 3,12-1 • Ongmally Basu et al. 18
recommended that the therapeutic range for heparin should be given by an
APTT ratio between 1.5 and 2.5. The intensity of treatment corresponding to
this range varies, however, with the local APTT test system. The variable
responses reagents in the US and UK demonstrated by national external quality
assessment surveys are illustrated in Figure 5.1. The different responses on
fresh samples from heparin-treated patients resulting from the use of different
coagulometers is shown in Figure 5.2. A calibrations constant was devised for

3.0

o
~
....
~2. D·
c..
« o·

.-""'=----+-+-r ----j-+---,----------,
0.2 0.4 0.8
M GOO
iu/ml
Figure 5.1 AP1T ratios from surveys from the College of American Pathologists aod UK external
quality assessment showing responses to in-vitro heparinized samples. The dotted line indicates
the Manchester AP1T response on flesh plasma from heparinized patients. The concentration of
heparin giving a 1.5 APTT ratio with the individual reagents is indicated by vertical arrows: M =
Manchester; G = GD Automated; D = Dade; 0 = Ortho

39
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

2.01····· 1: Electra

~ 1.5+ . .
2: KC
3: ACL
m I • 4: Coag·a·mate
1ii
c: I
I
• 5: Schnitger & Gross
!- ...........................................
~ 1.0 ............, ........... : ............ ...................................... 6:· (manual)
o . I : • 7: Cobas·Fibro
.~ 8: Fibrintimer
10: coagulometers in
~ 0.5 use in 1 Laboratory
(,)

o 2 3 4 5 6 7 9 10
Type of coagulometer

Figure 5.2 Calibration constant (b) by coagulometer. Reagent A versus reagent B

the ISTHlICSH study based on the orthogonal regression slope (as in proth-
rombin time standardization) of the plots of the log APTf results of the same
two reagents at different centres using different coagulometers8 . The marked
effect of the coagulometers is seen. Hirsh et al. 19 gave the equivalents for an
APTT ratio of 1.5 for the protamine titration as 0.2 units/ml and for the
anti-factor X assay as 0.3 units/mL The difficulty of employing these alterna-
tives for standardization is that the vast majority of centres prefer to use the
APTf for heparin control as it is regarded as a measure of the overall biological
response to heparin. As far as the heparin response is concerned anti-Xa levels
of 0.3 to 0.7 ulml have been recommended to demarcate the APTT ratio
prolongation 19. A problem with attempts to standardize heparin monitoring
by APTT is that APTT test systems differ in response to components of the
clotting mechanism other than heparin, e.g. anti-heparin activities. The
difficulties of standardization of APTT heparin monitorinl were highlighted
by the recent ISTHIICSH international collaborative study. APTT responses
of patients with recent thrombosis were much less to the same concentrations
of heparin. A calibration constant was determined for each local APTf system
from the orthogonal regression slope of plasmas from patients and healthy
subjects. It was found that the use of a single slope value for a brand of reagent
for all laboratories was not advisable, and each laboratory should perform its
own local calibration. The study recommended that the adoption of a reference
APTT reagent would be a useful step to initiate heparin monitoring standardi-
zation and to develop safe and effective local therapeutic ranges.

PATHOPHYSICAL VARIATION IN THE APTT

Stress, exercise, pregnancy, the post-partum state and surgical operations result
in acceleration of the test. Pro thrombotic changes associated with recent
deep-vein thrombosis, thromboembolic disorders and oestrogen administration
may result in accelerated clotting times4 • Acquired pathological states such as
liver disease, disseminated intravascular coagulation (DIC), and drug toxicity
cause prolongation. Certain drugs, including oral anticoagulants, heparin and
thrombolytic agents, prolong the APTT. The test is also prolonged in the new-
born20 •
40
 ACTIVATED PARTIAL THROMBOPLASTIN TIME

THE APTT IN THE EUROPEAN CONCERTED ACTION ON

THROMBOSIS (ECAT) STUDIES

A single APTT reagent and technique was selected by the executive committee
to be employed at all participant centres in the BCAT studies because of the
variability of performance of the different APTT methods summarized earlier
in this chapter, due to lack of standardization. The Manchester APTT was
chosen as it has been shown in a number of published studies to give good
sensitivity to a wide range of coagulation disorders and inhibitors 17 • The
manual technique was selected for the BCAT studies because automated
techniques may have marked and variable coagulometer effects on APTT
results. To gain familiarity with the method, and to reduce the influence of
inter-laboratory variation in technique, representatives from all participant
centres attended training courses organized by BCAT. On-going programmes
of external quality control were conducted subsequently to monitor their
performance. As stated earlier, a variety of activators and incubation times are
employed in current conventional preparations. The trend is to use non-opaque
activators and shorter incubation times (3-4 min). In the absence of a recognized
standard or reference preparation for the APTT the Manchester APTT may
be regarded as a reference method to assess the responsiveness of other APTT
procedures to concentrations of heparin and specific clotting factors (particularly
VIII and IX). The responsiveness to lupus anticoagulant is an additional
criterion2 . Some APTT reagents are relatively insensitive to the lupus anti-
coagulant.

THE MANCHESTER APTT METHOD

The reagent is a phospholipid extract of rabbit brain tissue which is prepared

in lyophilized form for long-term stability. It is reconstituted with physiological
buffer (Owren's buffer, pH 7.35). The activator is light kaolin. Optimum sen-
sitivity is achieved with an activation time of 10 min.

Details of technique
1. Lyophilized APTT reagent (store at -20°C). Reconstitute by adding exactly
10 ml of cold Owren's buffer (see item 2, below). The stopper is replaced
and is mixed gently to resuspend. The diluted suspension is dispensed into
small volumes, e.g. 1.0 ml amounts, using non-wettable containers which
are tightly capped and immediately frozen (-20 to -40 0C). This procedure
should be accomplished with minimum delay, preferably within 15 min from
reconstitution. The frozen aliquots of the reagent are stable for at least 3
months.
An aliquot of the diluted suspension should be thawed out rapidly when
required by placing in a water bath at 37°C for 1-2 min immediately before
use. This is maintained in crushed ice prior to use. The reagent is stable for
at least 2 h under these conditions. Any remaining reagent is discarded after
use and is not refrozen.
41
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
2. Owren's buffer (sodium diethylbarbiturate (11.75 g) and sodium chloride
(14.67 g) dissolved in 1570 m1 distilled water and 430 m1 0.1 N hydrochloric
acid (PH 7.35) for reconstitution of APTT reagent (store at + 2 to 8°C).
3. Light kaolin (0.25 g/100 ml) is suspended in Owren's buffer (store at + 2 to
8°C). It is mixed well before use. An aliquot is decanted when required into
a separate container and any remainder is discarded.

Method
Testing should be performed on fresh plasma, collected and stored as indicated
in Chapter 2, as soon as possible after collection and no longer than 1 h after
venipuncture. The test plasma is kept in a stoppered polystyrene container at
room temperature.
The kaolin suspension and calcium chloride are warmed in separate test
tubes in a water bath at 37°C. The following reagents are added in the order
indicated without delay into a glass tube, pre-warmed in a water bath:
0.1 m1 test plasma
0.1 m1 APTT reagent
0.1 m1 warmed kaolin suspension (after resuspension) and start stopwatch.
The test tube is tilted three times immediately, and subsequently at approximately
1 min intervals in order to resuspend the kaolin.
At exactly 10 min, 0.1 m1 warmed calcium chloride (0.025 moUL) is added.
The tube is tilted gently three times and left undisturbed in a water bath for
exactly 20 s from the addition of the calcium chloride. It is then tilted gently
until a solid clot forms. The clotting time (seconds) is recorded and the test is
performed in duplicate.
The normal range for the Manchester APTT is based on results from a large
normal population of both sexes, spanning the adult age range. These extend
from 36.0 to 48.0 S, based on two standard errors from the mean with correction
for non-Gaussian distribution.

SOURCES OF ERROR

Although it is a relatively simple test to perform, the APTT is subject to a

number of possible sources of error. The first can arise from faulty blood
collection. Contamination by tissue juices, causing acceleration of the test, can
result from traumatic and/or difficult venipuncture. In addition delay in mixing
with the anticoagulant may cause partial clotting, which is not always obvious
visually. In the ECAT studies this possibility has been minimized by the use
of siliconized evacuated tubes. The use of test tubes for the performance of
the test, which are not chemically clean, can also have a deleterious effect. With
the manual technique, employed with the Manchester reagent, it is important
to control contact activation in order to facilitate the reading of the end-point.
During the incubation time the kaolin should be kept suspended in the reaction
mixture by gentle mixing at regular intervals.
42
 ACTIVATED PARTIAL THROMBOPLASTIN TIME

COAGULOMETER TESTING

Some of the newer types of coagulometer require a much less opaque reagent
than the above procedure with kaolin. Non-opaque liquid activation (sulphatides,
ellagic acid, etc.) and other activators of low opacity, e.g. micronized silica, are
incorporated into commercial products and are more compatible with some
coagulometers than is kaolin. The variable effects of different coagulometers
on the APTT response to heparin with a modification of the Manchester APTT
method employing sulphatides as the activator have been reported8 •

THE NEED FOR STANDARDIZATION OF THE APTT

Because APTT results are influenced by changes in various different stages of

the coagulation mechanism they are subject to more variables than specific
clotting assays. The requirement for standardization of the APTT method has
been emphasized in many reports 3,12,13,16--18,21-23. The variable response to
heparin and factor VIII is well established. No agreement on a reference reagent
and a method has, however, proved possible.

References
1. Stevenson KJ, Easton AC, Curry A, Thomson JM, Poller L. The reliability of activated partial
thromboplastin time methods and the relationship to lipid composition and ultrastructure.
Thromb Haemostas 1986; 55: 250-8.
2. Kelsey PR, Stevenson KJ, Poller L. The diagnosis of lupus anticoagulants by the activated
partial thromboplastin time - the central role of phosphatidyl serine. Thromb Haemostas
1984; 52: 172-5.
3. Barrowcliffe TW, Gray E. Studies of phospholipid reagents used in coagulation II: factors
influencing their sensitivity to heparin. Thromb Haemostas 1981; 46: 634-7.
4. Thomson JM, Poller L. The activated partial thromboplastin time. In: Thomson JM, ed.
Blood coagulation and haemostasis: a practical guide. Edinburgh: Churchill Livingstone,
1985; 301-9.
5. Triplett DA, Smith C. Sensitivity of the activated partial thromboplastin time: results of the
CAP survey and a series of mild and moderate factor deficiencies. In: Triplett DA, ed.
Standardisation of coagulation assays: an overview. Skokie, IL: College of American Patholo-
gists, 1982; 137--63.
6. Koepke JA. The use of proficiency testing results in the coagulation laboratory. In: Triplett
DA, ed. Laboratory evaluation of coagulation. Chicago, IL: American Society of Clinical
Pathologists, 1982; 368-87.
7. Proctor RR, Rapaport SL. The partial thromboplastin time with kaolin. A simple screening
test for first stage plasma clotting deficiencies. Am J Clin PatholI961; 36: 212-19.
8. van der Velde EA, Poller L. The APTT monitoring of heparin. The ISTHlICSH collaborative
study. Thromb Haemostas 1995; 73: 73-81.
9. Jennings I, Kitchen S, Woods TA, Preston FE. Monitoring heparin therapy. Instrument effects
on reagent sensitivity. Thromb Haemostas 1997 (Suppl.): 279, abstract 1141.
10. Thomson JM. The control of heparin therapy by the activated partial thromboplastin time:
sensitivity of various thromboplastins to heparin. In: Triplett DA, ed. Standardisation of
coagulation assays: an overview. Skokie, IL: College of American Pathologists, 1982; 195-206.
11. Triplett DA. Heparin: Clinical use and monitoring. In: Triplett DA, ed. Laboratory evalu-
ation of coagulation. Chicago, IL: American Society of Clinical Pathologists, 1982; 271-313.
12. Poller L, Thomson JM. The partial thromboplastin (cephalin) time test. J Clin PathoI1972;
25: 1038-44.

43
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
13. Koepke 1. The partial thromboplastin time in the CAP survey programme. Am J Clin Pathol
1974; 63: 990--4.
14. Shapiro GA, Huntzinger SW, Wilson JE. Variations among commercial activated partial
thromboplastin time reagents in response to heparin. Am J Clin Patho11977; 67: 477-80.
15. CISMEL (Italian) Study Group. Activated partial thromboplastin time - a multicentre evalu-
ation of commercial reagents in the diagnosis of haemophilia. Scand J Haematol1980; 25:
308.
16. Brandt IT, Triplett DA. Laboratory monitoring of heparin. Effects of reagents and instru-
ments on the activated partial thromboplastin time. Am J Clin Patho11981; 76: 530-7.
17. Mannucci PM. A multicentre evaluation of partial thromboplastin time reagents in the
detection of mild haemophiliacs. In: Triplett DA, ed. Standardisation of coagulation assays:
an overview. Skokie, IL: College of American Pathologists, 1982; 165-77.
18. Basu D, Gallus A, Hirsh JA. A prospective study of the value of monitoring heparin treatment
with the activated partial thromboplastin time. N Engl J Med 1972; 287: 324-7.
19. Hirsh J, Raschke R, Warkentin TE, Dalen JE, Deykin D, Poller L. Heparin: mechanism of
action, pharmacokinetics, dosing considerations, monitoring, efficacy and safety. Chest (Suppl.)
1995; 108: 258-75S.
20. Hathaway WE, Bonnar J. Physiology of coagulation in the fetus and newborn infant. In:
Perinatal coagulation. New York: Grune & Stratton, 1978; 69.
21. Refsurn N, Abildgaard U. 'Cephotest' in monitoring heparin therapy. Evaluation of a
standardised activated partial thromboplastin time (AP1T) test. Farmakoterapi 1976; 1-2:
27-35.
22. Shapiro GA, Huntzinger SW, Wilson JE. Variations among commercial activated partial
thromboplastin time reagents in response to heparin. Am J Clin PathoI1977; 67: 477-80.
23. Van den Besselaar AMHP, Meeuwisse-Braun J, Jansen-Gruter R, Bertina RM. Monitoring
heparin therapy by the activated partial thromboplastin time - the effect of pre-analytical
conditions. Thromb Haemostas 1987; 57: 226-31.

44
6
Prothrombin time (PT)
L.POLLER

INTRODUCTION

The prothrombin time (P1) is the screening test for the extrinsic (tissue) clotting
system. The PT was originally introduced by Quick l as a measure of a
coagulation defect in the newborn and in jaundiced patients, and subsequently
adapted for its present principal uses in the screening of extrinsic clotting and
in monitoring of oral anticoagulant dosage. Although specific methods were
subsequently developed designed for anticoagulant control, e.g. the P&P test2
and Thrombotese the Quick test continued in most countries to be the exclusive
control method. The PT is also the screening test for defects of extrinsic (tissue)
clotting. Its responsiveness to depression of the extrinsic clotting factors depends
on the source and type of tissue factor (TF) thromboplastin extract. TF is a
combination of an apoprotein and phospholipid. It has been purified and its
chemical structure has been defined4 ,5. Human recombinant TF relifdated
with natural or synthetic phospholipids is now being widely employed 8. The
EeAT study has been concerned principally with its application of the PT in
assessing extrinsic clotting function. The PT reflects changes in three of the
vitamin K-dependent clotting factors (factors II, VII and X), and the non-vitamin
K-dependent factor V. In the PT the recalcification time of citrated plasma is
accelerated by a TF extract. The test depends on activation by TF of factor X
in the presence of factor VII and the bypassing of the intrinsic pathway. The
speed of this reaction and the responsiveness of the PT to deficiencies of
clotting factors depends upon the properties and concentration of the TF, as
well as to the clotting factor concentration.

45
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
SOURCES OF TISSUE FACTOR (TF)

Many modifications of the test have been introduced since Quick's original
rabbit brain extract. Other tissues in addition to rabbit brain have been used
as a source of TF in routine work, e.g. human brain, ox brain, monkey brain,
rabbit lung and human placenta. TF is a combination of an apoprotein and
phospholipid and is a complete thromboplastin as opposed to the partial
thromboplastin of the activated partial thromboplastin time (APTT) which
lacks the apoprotein. It has been purified and its chemical structure has been
defined4 ,5. Human recombinant TF is now being produced which, when re-
lipidated, can be employed as the TF in the PT test6 ,7. The dangers of virus
transmission from human tissue led to the most widely used preparations being
derived from rabbit brain, but there is now increasing demand for the human
recombinant preparations because of the avoidance of virus transmission
combined with high responsiveness. TF extracts vary in their procoagulant
properties and their ability to detect alterations in individual clotting factors
or systems. The reagent used in the ECAT study was a responsive rabbit brain
extract with a similar low International Sensitivity Index (lSI) to the new human
recombinant reagents.

THE NORMAL RANGE AND THE EFFECT OF COAGULOMETERS

The reference range

The reference (normal) range for the PT test varies with pre-test and analytical
variables including the species of TF origin, the potency of the extract as well
as the method of end-point detection. The reference range should be established
on a minimum of 20 healthy adults including both sexes over a wide age range
of 20-80 years. The mean normal prothrombin time (MNPT) (geometric mean
of 20 healthy subjects) is then calculated. A larger number than 20 will give a
more reliable normal mean, and a total of 50-100 healthy subjects is recom-
mended for a more reliable reference range which should be calculated as the
MNPT ± 2 SD. Reference values with automated systems may be appreciably
shorter than with the manual technique.
Normal values have been investigated in European Concerted Action on
Anticoagulation (ECAA) field studies with two ECAA reference reagents.
Coagulometers which shorten the normal PT compared to prolonged
PT, disturb prothrombin ratios and hence the international normalized ratio
(INR)9.

THE PT IN DIFFERENT CLINICAL STATES

The PT is prolonged by oral anticoagulation, impaired liver function, biliary

obstruction and congenital deficiencies of factors II, V, VII and X. The PT
with many reagents may be prolonged by heparin administration, although
usually only with excessive doses. Some thromboplastins contain anti-heparin
46
 PROTHROMBIN TIME
agents, e.g. polybrene, making them unresponsive to heparin at therapeutic
levels, which is important in the view of the increasing trend to combined
heparin and warfarin therapy. The PT is considerably prolonged in the neonatal
period and to a lesser extent throughout childhood. On the other hand, PT
may be accelerated in certain pre-thrombotic states, e.g. after administration
of high-dose oestrogen oral contraceptives, after surgical operations and
post-partum. In adults the PT is shorter than in juveniles and children.

LABORATORY VARIATION IN PT TESTING

The responsiveness of the PT test to a specific coagulation deficiency is

dependent on the source and type of tissue thromboplastin. When the PT is
used as a measure of the depression of vitamin K-dependent clotting factors,
e.g. following oral anticoagulant administration or in liver function testing, the
PT results with the different thromboplastins vary greatly. The overall sensitivity
to the depression of factors II, VII and X during oral anticoagulant treatment
is quantified numerically as the International Sensitivity Index (lSI) using the
World Health Organization (WHO) international system of PT standard-
ization 10--12 • This depends on a calibration of the individual thromboplastin
extract against the WHO international reference preparation (lRP) for
thromboplastin or a secondary IRP reagent calibrated in terms of the first
primary WHO IRP. The responsiveness of the first WHO IRP is by definition
1.0. The slope of the orthogonal regression line obtained when the log PT
with the local thromboplastin of 20 normal and 60 coumarin-treated patients
samples are plotted on a logarithmic scale against the PT (log) with an IRP,
is a measure of the responsiveness of the thromboplastin. The calibration
=
slope provides the lSI of the IRP (i.e. slope x lSI IRP lSI of test reagent).
Sensitive thromboplastins have an lSI value close to 1.0, whereas less
responsive thromboplastins have higher values. Traditionally North American
reagents of rabbit origin produced since the 1950s had lSI values between
2.0 and 2.8.
The provision of a low lSI thromboplastin with good sensitivity to factor
VII is essential for a reliable assessment of liver function. The progress of
patients with liver disease and liver impairment secondary to other diseases
and drug overdo sages (e.g. paracetamol, aspirin, etc.) may then be monitored
by the PT.

Standardization of the prothrombin time

It has long been recognized that PT results show poor agreement between
centres mainly due to differences in performance of the thromboplastins used
in the test to accelerate clotting. This is aggravated by different methods of
reporting results, e.g. activity %, index or ratio.
The lack of uniformity of PT results between centres has contributed to
continuing differences in the intensity of anticoagulation on a world-wide
scale 13 • Thus the result on a given plasma may signify overdosage at one hospital
or underdosage at another, dependent entirely on the laboratory technique.
47
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
The WHO standardization scheme
The establishment by the WHO of the first primary reference thromboplastin
(WHO Technical Series 1977)14 provided the basis for international
standardization of the PT.
The WHO thromboplastin calibration scheme (WHO Technical Series 1983i 5
emerged from an international study (ICTHIICSH 1979)16.
After a series of previous attempts, an acceptable PT standardization
statistical model was developed by KirkwoodlO. The logarithms of PT values
with different thromboplastins are related linearly, and this linear relationship
has been used to calibrate thromboplastins in the new procedure. The linear
relationship between the logarithms of PT is estimated using a symmetric
procedure of least-squares estimation of a functional relationship I7 described
as orthogonal regression. This procedure allows for the fact that an individual
patient's prothrombin times determined using both thromboplastins (the IRP
and test reagent) show biological and experimental variation about the
calibration line. The details of the calibration procedure are given in the
Appendix.

International Sensitivity Index (lSI)

In order to standardize results with different types of thromboplastin, so as
to be able to interpret the results of the prothrombin time test on a common
scale, it is essential that each reagent is correctly calibrated with an lSI.
Prothrombin time results are expressed on a common scale, i.e. the INR, if the
lSI of the thromboplastin used is known. For any given thromboplastin the
lSI is derived from the slope of the calibration line when the logarithms of PT
with the IRP are plotted on the vertical axis against the logarithms of PT
obtained with the test thromboplastin on the same plasmas from healthy
subjects and anticoagulated patients (see Figure 6.1).

International Normalized Ratio (INR)

This is the prothrombin ratio which, it is calculated, would have been obtained
if the first primary WHO reference thromboplastin (67/40 human combined)
had been used to perform the test with the manual technique.
INR = (observed ratio )ISI
For example, a working thromboplastin with a prothrombin ratio of 1.5 and
an lSI of 2.0 gives an INR of 2.25, i.e.:
=
INR 1.52.0 2.25 =
The average sensitivity of a reagent to the depression of factors II, VII and X
during oral anticoagulant treatment is quantified as the lSI using the WHO
international system of PT standardizationI0-12.18.
The lSI depends on calibration of the individual thromboplastin extract in
48
 PROTHROMBIN TIME

2.0

1.8

t="
e..
Ol
g
c 1.6
~
..!l!
Co
.8 0
E
e
J::
1.4 o Patients (n=60j
f- + Normals (n=20j
e..
9;

1.2

1.0 -1----..-------.------.------.------.
1.0 1.2 1.4 1.6 1.8 2.0

Local Thromboplastin (log PTj

Figure 6.1 Example of orthogonal regression slopes in lSI calibration showing good coincidence
of patient/normal and patient-only lines

terms of the first primary WHO IRP for thromboplastin (human combined
67/40), and as this reagent is no longer available against a secondary reference
preparation calibrated in terms of the first primary WHO, the responsiveness
of which was by definition 1.0. The slope of the orthogonal regression line
from which lSI are derived is obtained from plotting the log PT of 20 healthy
subjects and of 60 long-term-stabilized coumarin-treated patients' samples with
the local thromboplastin against the log PT with the relevant IRP (see Figure
6.1) as experience has shown that these numbers give reliable results within the
3% limits of precision of the CV of the slope of the calibration line. Responsive
thromboplastins have an lSI close to 1.0, whereas less responsive reagents have
higher values.
To enable manufacturers of thromboplastin to relate their materials to the
WHO reference thromboplastin standard, three further reference thromboplastins
were later designated IRP. These were BCf/099 (human, plain), OBT179 (bovine,
combined) and RBT179 (rabbit, plain) to be used with the relevant species of
routine thromboplastin.
Manufacturers of thromboplastin are expected to provide the lSI of their
reagent. Each package should contain a table or graph to convert into INR
results usually expressed as prothrombin ratios obtained with the local
thromboplastin.
The WHO calibration scheme is hierarchical (see Figure 6.2). Starting with
49
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

WHO International Reference Materials for Thromboplastins

1st International Reference Preparation

human, combined 67/40 (1976)

1st IRP, bovine 1st IRP, rabbit, plain

combined 70/178 (1978)
68/434 (1978)

2nd IRP, bovine,

combined, 2nd IRP, human, plain 2nd IRP, rabbit, plain
BCT1253 (1983) RBT179 (1982)
OBT179 (1983)

3rd lSI, recombinant 3rd IRP rabbit, plain

human, plain RBT190 (1995)
rTF195 (1996)

Figure 6.2 Hierarchical relationship of IRP

the first WHO primary IRP, the calibration model proceeds through second-
and third-generation IRP down to house standards and manufacturers'
production lots. Manufacturers of thromboplastin are expected to produce
their own house standard, which should be a lyophilized batch of thromboplastin
set aside for the calibration of production batches. The WHO IRP* are in very
limited supply, and are intended only for the calibration of national reference
preparations or working reference preparations, at 'national control laboratories' ,
not manufacturers' reagents.
The European Commission Bureau Communautaire de Reference (BCR)
currently provides, at not-insignificant cost, certified reference thromboplastins
CRM/149S (rabbit plain)'9 and OBT179 (bovine combined)20. The human
plain reagent (BCT/099) has been discontinued and not yet replaced. The CRM
are intended to be available to manufacturers of commercial thromboplastins.
The International Council for Standardization in Haematology (ICSH) made
available an additional human plain IRP (BCT/441i'. A WHO human plain
IRP (recombinant human) replacement for BCT/441 human IRP is now
approved (RTFI95).
The precision of a calibration is greater when similar reagents and techniques
are compared. The WHO scheme states that each step of the calibration should
be made, whenever possible, against a reference preparation which is most
similar in terms of the species of tissue extract. It is also important to compare
'plain' reagents, i.e. used with a PT technique with reference reagents that are
also plain. Similarly combined (with added factor V and fibrinogen) reagents
(e.g. Thrombotest) should be calibrated with a combined reference prepara-
tion22 . Calibration of successive batches of thromboplastin should be carried

*WHO Custodian Laboratory is the Central Laboratory for Blood Transfusion, Plesmanlaan 125,
1066 ex, Amsterdam, The Netherlands.

50
 PROTHROMBIN TIME

out by comparison with a national or in-house reference preparation using the

WHO calibration protocol22 •
A calibration exercise consists of a sequence of parallel PT with the respective
thromboplastins on fresh plasma samples from healthy subjects and coumarin-
treated patients. The number of plasma samples included in a calibration should
be as large as necessary to ensure that the precision of the slope of the orthogonal
regression calibration line (b), expressed as a coefficient of variation (CV), is
=
less than 3% (CV of slope(b) 100 x SD(b)/b). In practice blood samples from
20 healthy subjects and 60 stabilized patients receiving oral anticoagulants have
invariably achieved this precision. This fixed number of constituent plasma
samples has been a constant feature of multicentre exercises in IRP calibration,
and allows better advance planning of a calibration exercise. It is customarily
referred to as a 'full fresh calibration'. Testing on several separate days is advised
to reduce the effects of the day-to-day technical (intra-laboratory) variation
in the PT determination. The WHO Technical Series (1983)15 protocol states
that the coagulation end-point may be determined by a manual technique or
with the aid of a semi-automated end-point recorder. With recognition of the
modification of PT results by coagulometers even of the same brand it has
become accepted that only manual PT calibrations are acceptable for lSI
calibration, and this has been proposed in the revision of the 1983 WHO
procedure. The manual PT technique has been used in all multicentre calibra-
tions of the thromboplastin IRP and the 'true' INR has become accepted as
the equivalent of the manual PT result with the first WHO IRP in view of
coagulometer effects on the prothrombin ratio (PR).
Figure 6.1 provides an example of a single-centre calibration exercise with
PT expressed as log values. Recommended specifications in lSI determination
include:
1. the relationship between the logarithms of the PT values should be linear;
2. a line drawn through the logarithms of PT values of the patients should
intersect the logarithms of PT values from the healthy subjects;
3. the variability of points about the line should be constant over its whole
length;
4. outlying points (more than 3 SD from the line) should be removed.

Nomogram for INR derivation

Figure 6.3 is a nomogram for correcting prothrombin ratios to INR, and can
be used for any thromboplastin reagent where the lSI is known, without the
need for calculations23 . It is a common and fixed misconception that for an
individual patient's plasma sample the INR will be identical with different
thromboplastins. There is necessarily a degree of imprecision in lSI calibration,
as it is based on comparison of varying PT responses of the two different
thromboplastin reagents to the plasma of patients with anticoagulant-induced
coagulation defects of varying intensity and clotting factor content. It is therefore
emphasized that lSI calibration is based on this average response used in the
empirically demonstrated linear relationship between the logarithms of PT. A
51
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

INR

9.0

8.0

7.0

-cQ)
C)
as 5.0
8.0

5.0
Q)
a::: 4.5
~0 4.0 4.0
~
3.8
-I 3.6 E
0 3.4 :::i
~ 3.2 0
a:::
3.0 3.0 ~
:::I
C 2.8 Q)
:c 2.6
~
I!!
E
e 2.4 Q)

r=.
e
.r:. 2.2

2.0 2.0--
D.. 1.9
1.8
1.7
1.8
1.5
1.4
1.3
1.2
1.1
1.0 I---~--~--l---~-~::__--+ 1.0
1.0 1.1 1.2 1.3
Intemational Sensitivity Index (lSI)
Figure 6.3 Nomogram for INR derivation. Example: observed ratio 2.65, with thromboplastin
lSI 1.3; INR = 4

good correlation does not necessarily imply the interchangeability of two

methods of measurement24 • A demonstration of such variability, based on
average response, is that there is a strong correlation between height and shoe
size, but most people would find it extremely uncomfortable wearing shoes
predicted on the basis of their height! When it is appreciated that different
thromboplastins also vary greatly in responsiveness to the individual coumarin-
dependent clotting factors, i.e. II, VII and X, as well as some non-coumarin-
dependent clotting factors, e.g. factor V, discrepancies between INR with
different working thromboplastins arising from these biological variations and
additional technical errors are therefore not unexpected.
The stated within-centre precision requirement is for a CV of the orthogonal
regression slope of less than 3%. The CV of the lSI defined as:
CV of lSI =100 x SD(ISI)IISI,
where:
SD(ISI) = v'(lSIref X SD(b»2 + (b x SD(ISIref»2
52
 PROTHROMBIN TIME

is also used in calibration programmes takes into account also the imprecision
of the earlier calibration(s) of the IRP and in-house reference preparations
against which a new reagent is being calibrated. The CV of the lSI is therefore
greater than the CV of the slope.

Coagulometer effects
In view of the coagulometer effect on the ISe5-28 some manufacturers provide
a 'system lSI' for a particular thromboplastin!coagulometer combination. The
limitations of the procedure have been demonstrated by the variability of the
system lSI with the same reagent and coagulometer at different centres found
in collaborative studies29,3o.
Coagulometers have now largely replaced the manual PT technique in most
countries. They may cause a disproportionate shortening of PT results from
healthy subjects compared to the shortening effects on patients' plasmas, thus
artificially increasing the prothrombin ratio and reducing lSI. In view of this
some manufacturers provide a 'system lSI' for a particular thromboplastin!
coagulometer combination. This may assist, but the limitations of the procedure
have been demonstrated by the variability of system lSI with the same reagent
and coagulometer at different centres found in collaborative studies29,3o. The
need for local system lSI at individual laboratories has therefore been demon-
strated, as well as the need for multiple calibrations where more than the
coagulometer is in use in a single laboratory. As conventional local lSI calibration
necessitates the performance of the generally discarded manual PT technique
with the thromboplastin IRP it is therefore not now usually feasible. The use
of certified lyophilized plasma calibrants has been recommended as an alternative
for local lSI determination29- 32 • The reliability of the procedure and the
minimum requirement of numbers of certified lyophilized plasmas for a
calibration is currently being studied. A European Concerted Action on
Anticoagulation (ECAA) exercise33 confirmed that lyophilized plasmas give a
reasonable approximation to fresh plasmas in a full calibration of 60 lyophilized
versus 60 fresh plasmas from stabilized treated patients. In an international
collaborative calibration at 15 centres there was a demonstrable but small
difference (mean 4.5%) from a fresh plasma calibration of the lSI of low lSI
ECAA human recombinant reference reagent when such plasmas were used,
presumably due to a freezing or lyophilization effect. Furthermore lyophilized
artificially depleted plasma could be reliably substituted for lyophilized coumarin
plasmas from coumarin-treated patients, which makes lyophilized plasma
calibration a practical procedure. A further ECAA report34 indicated that a
minimum of 20 lyophilized depleted plasmas, with a proportion of samples
from healthy subjects comparable to a full calibration (i.e. seven normals to 20
abnormal plasma samples), is required to give a similar lSI to a full calibration
based on 60 lyophilized plasmas and 20 fresh plasma samples from healthy
individuals.

53
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Ratios equivalent to INR range 2.0 - 3.0
3.0

Vertical lines are ratios equivalent

2.5
to INR range 2.0 - 3.0

1.5 III I I I I I I I I I I
1.0 I I I I I I I I I I I I I I I I I -,---,
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8
lSI
Figure 6.4 Effect of lSI on therapeutic window for warfarin dosage

LOW lSI THROMBOPLASTIN

Reference has been made to the importance of a low lSI thromboplastin in

maintaining good laboratory precision when PR are expressed as INR. A major
clinical advantage of highly responsive reagents, i.e. lSI ranging from 1.0 to
1.2 in terms of oral anticoagulant dosage, is shown in Figure 6.4, which shows
that, as the lSI increases, the width of the therapeutic equivalents expressed
in PR progressively diminishes 23 • Dosage control with a poorly responsive
(high-lSI) preparation spans a very narrow PR range compared with those
observed with low-lSI reagents. PT measurements based on a low-lSI reagent
not only are more precise but give a wider therapeutic interval which is easier
to achieve and should be safer for the patient. The Fourth ACCP Consensus
Conference35 recommended that thromboplastins should have an lSI of less
than 1.5 and preferably less than 1.2.
The lSI calibration procedure is described in detail in the Appendix.

COMPUTERIZED CALIBRATION PROGRAMS

The mathematics of the lSI calculation described in the Appendix have proved
too complex for general laboratories. The calculation has been computerized in
several centres and specialist laboratories have developed their own programs.
'National Control Laboratories' should be able to provide a formatted disk with
a recommended calibration program and its methods of use. *

"'Such a program, a modification of which is being prepared for WHO Biologicals (Geneva), is
available at nominal charge on application to the ECAA Central Facility: Professor L. Poller,
University of Manchester, Oxford Road, Manchester M13 9PT, UK or from Health Technology,
WHO, Geneva.

54
 PROTHROMBIN TIME

SOURCES OF ERROR

The test is relatively simple but subject to a number of pre-test and test variables
which affect the result. Faulty blood collection and haemolysis affect the test.
The use of siliconized or plastic test tubes instead of the recommended
borosilicate glass tubes for the test, length and bore of glass tubes, angle and
speed of the tilting are amongst the variables which affect the PT when
performed manually.

ECATSTUDY

The universal use of all centres of the same reagent in the ECAT study overcame
some of the problems of PT standardization. A single recommended reagent
and technique was therefore selected by the Executive Committee to be employed
at all centres with a manual tilt-tube technique. The manual method was recom-
mended because coagulometers accelerate the test to varying degrees and may
also affect the lSI.
The Manchester Reagent thromboplastin was selected for the ECAT study
because it has a low lSI (1.1 approximately), shows good responsiveness to the
individual extrinsic clotting factors and little inter-batch variation. It has been
shown to detect the acceleration of extrinsic clotting which occurs in the days
following major sUf§ery36. It shows good responsiveness to depression of factor
VII in liver disease 7. The precision of testing with Manchester Reagent in
national external quality assessment has been shown to be greater than with
some other reagents38 .

TECHNIQUE

A standardized manual technique was provided therefore for the ECAT study
as follows:

Prothrombin time test

Reagents
1. Calcium chloride 0.025 moVL. Prepare from an MIl solution. Store at
2-8°C.
2. Thromboplastin, Manchester Reagent. Store at 2-8 dc. Not to be/rozen.
Method
N.B.: A manual (tilt-tube) technique must be used.
Testing should be performed as soon as possible after collection, preferably
within I h from venipuncture. Maintain plasma in a stoppered polystyrene
container at room temperature.
55
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Warm sufficient test tubes for performance of the test in a 37°C water
bath {tolerance limits 37.0 ± 0.2 0q. Clean, new glass test tubes should be
used.
1. Resuspend the thromboplastin extract by gentle inversion and transfer 0.1
vol into glass test tubes in the water bath. Allow to warm for 1-2 min.
Thromboplastin should not be allowed to pre-warm in the test tubes for
> I h.
2. Add 0.1 ml plasma using automatic pipette. New pipetting tips must be used
for each plasma.
3. After approximately 1 min add 0.1 ml pre-warmed calcium chloride (0.025
moUl) and start stopwatch.
Tilt gently, keeping tubes under water as much as possible to maintain
optimal temperature. The speed and angle of tilting the test tube must be
standardized (three tilts through every 90° every 5 s) to control glass activation
and minimize cooling. Record end-point in clotting time (seconds).
Tests should be performed in duplicate. If the discrepancy between duplicate tests
is more than 5% of the mean clotting time the test should be repeated, i.e. if the
mean clotting time is 30 s, duplicates should range between 28.5 and 31.5 s.

APPENDIX
Recommended protocol for thromboplastin calibration
1. Blood samples from healthy individuals and patients stabilized on coumarin
treatment are required for the conventional fresh plasma thromboplastin
calibration. The total number of plasmas recommended is 20 normals and
60 samples from coumarin treated patients. Testing is performed on separate
days but these need not be consecutive.
A recommended plan for a working day is as follows:
PT tests on:
Eight fresh plasmas from long-term-stabilized coumarin patients (single
tests).
Two fresh plasmas from healthy subjects (duplicate tests).
One lyophilized plasma from a healthy subject (duplicate test, tested on
one day only).
The following reagents are to be tested: (a) IRP, (b) local thromboplastin.
Number of tests per day: 50 (approx).
Number of test days: eight (approx).
As a guide, the time schedule of a single exercise could be as follows:
1.1. Collection of blood samples. Blood drawn from healthy subjects and
from patients stabilized on oral anticoagulants should be centrifuged
immediately after collection and the plasma transferred to a non-wettable,
stoppered container. Maintain at room temperature until tested.
1.2. Testing of the plasma samples with the thromboplastins is described in
section 6.

56
 PROTHROMBIN TIME

Notes:
(a) Single tests only should be performed on the samples from coumarin
patients.
(b) Tests on each plasma should be performed with each thromboplastin
in turn before proceeding to the next plasma.
2. Selection of healthy subjects and patients
2.1. Normals. These samples should be obtained from healthy adults of
both sexes with a reasonable age spread. Select different normal subjects
on each day of testing. Test each sample in duplicate.
2.2. Plasmas from coumarin-treated patients. The patients should have been
on oral anticoagulants for at least 6 weeks and have a result within the
therapeutic range. Participant centres should include patient samples
displaying a variety of levels of anticoagulation within the therapeutic
range. Perform single tests only.
3. Sample collection
3.1. The blood should be collected by clean venipuncture. One part of sterile
trisodium citrate 0.109 moVL for nine parts of sample is recommended.
If an evacuated tube is employed it should be siliconized (see section 5.2).
3.2. The blood shall be centrifuged immediately after collection (approx 800g)
for 5 min, at room temperature.
3.3. The plasma shall be transferred by siliconized or plastic pipette to a
stoppered, non-wettable container.
3.4. The stoppered plasma container is to be kept at room temperature until
testing.
4. Reconstitution of thromboplastins
International Reforence Preparation for Thromboplastin (IRP). Extreme care
should be taken to avoid wastage of IRP. Reconstitute each ampoule of
IRP with requisite volume of reconstitution fluid. Leave thromboplastin in
ampoule (or vial). Keep at room temperature. Ensure that the thromboplastin
is completely resuspended before use. Gentle tapping on the side of the
ampoule with a finger will facilitate resuspension. It is not necessary or
desirable to shake the suspension vigorously.
5. Other reagents and equipment
5.1. Calcium chloride 0.025 moVL; for recalcification of plasma/thrombo-
plastin mixture. Store at 4 0c. Use fresh aliquots each day.
5.2. Equipment. Syringes and evacuated tubes must be silicone-coated or
made of good-quality plastic material to prevent contact activation of
coagulation factors.
Plastic or siliconized pipettes for transfer of plasma into glass test
tubes at the time of testing.
Use non-siliconized disposable glass test tubes for the actual testing.
Water bath at 37 °C (tolerance limits: 37.0 °C ± 0.2 0C). Use calibrated
thermometer.
New pipetting tips must be used for each test.
Accurate stopwatch.

57
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

6. Testing procedure
6.1. The WHO recommended procedure for calibration of thromboplastin
is based on the manual tilt-tube (or hook) technique. This method must
be used for all calibrations.
6.2. The sequence of thromboplastin testing must be alternated on each day
of testing. This sequence should be retained for the first day of testing
for all patients' and normal plasmas. On the second day this sequence
of testing should be changed.
The exercise should take no more than 10 days to complete with, e.g.,
two normals and six patients on each day. If it can be completed in a
shorter period this is acceptable as long as the order of testing is changed.
A specimen work plan is given below.

Specimen work sheet

Fresh plasma PT (sec)

Reagents

Number IRP Local Reagent

I Fresh normal 1 I I
Coumarin 1
Coumarin 2
Coumarin 3
Coumarin 4
Coumarin 5
Coumarin 6

I Fresh normal 2 I
Coumarin =plasma from oral anticoagulated treated patient)
Note: Test all 'normal' plasmas in duplicate but perform single tests
only on coumarin plasma-treated samples. The sequence should
be alternated on each day of testing.
6.3. The coagulation end-point should be determined by the recommended
tilt-tube technique. Test tubes should be kept under water at 37 °C
as much as possible in order to maintain optimal temperature. Proceed
as follows: pre-warm glass test tubes in the water bath prior to testing.

58
 PROTHROMBIN TIME
The testing procedure for rabbit and human plain IRP is as follows:
(a) transfer 0.1 ml thromboplastin to a test tube;
(b) incubate for 2 min;
(c) add 0.1 ml plasma;
(d) mix gently with the thromboplastin;
(e) wait 1 min for incubation to reach the optimal reaction tem-
perature;
(f) recalcify with 0.1 ml pre-warmed calcium chloride 0.025 moUL;
(g) tilt test-tube manually three times in 5 seconds through an angle
of 90°; record clotting time (seconds).
For bovine IRP the volumes are different.
7. Prothrombin times must be recorded to one decimal place and must not be
rounded to the nearest second. For example if a PT of 34.1 s is observed on
the stopwatch this time must be recorded; the time should not be rounded
to 34 s.

lSI calculation
The prothrombin times of all plasma specimens (20 healthy subjects and 60
patients) are converted to the corresponding logarithms. Let z be the logarithm
of a prothrombin timels determined with the IRP and x be the logarithm of
prothrombin timels determined with the local reagent to be calibrated. The
relationship z = al + blx is calculated by the following formulae, in which al
and b l are the orthogonal regression line parameters representing the intercept
and the slope respectively.

with

and al =z-blx;
x is the arithmetic mean of x and z the mean of z; Sx and Sz are the standard
deviations of the x and z values and r the correlation coefficient.
This orthogonal regression line estimates the relation between log prothrombin
timels of IRP and local thromboplastin. With the certified parameters a and
b of the calibration line of the IRP against WHO preparation 67/40, the intercept
and slope of the local reagent in relation to WHO first primary IRP 67/40 are
as follows:
intercept: anew =a - b . al
slope: blocal = lSI =b . bl
where a and b are the certified parameters and al and bl are the newly determined
local parameters.

59
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
To transform the logarithmic value into the value that would have been
obtained with the first WHO IRP human combined (67/40) the following
equation is used:
PT67140 =antilog(a + b . x)
where a and b are the certified parameters and x is the log PT.

References
1. Quick 1. The prothrombin time in haemophilia and in obstructive jaundice. J Bioi Chern 1935;
109: 73-4.
2. Owren PA, Aas K. The control of dicoumarol therapy and the quantitative determination of
prothrombin and proconvertin. Scand J Clin Lab Invest 1951; 3: 201-8.
3. Owren PA. Thrombotest. A new method for controlling anticoagulant therapy. Lancet 1959;
2: 754-8.
4. Broze GJ, Leykam JE, Schwartz BD, Miletich JP. Purification of human brain tissue. J Bioi
Chern 1985; 260: 10917-20.
5. Guha A, Bach R, Konigsberg W, Nemerson Y. Affinity purification of human tissue factor:
interaction of factor VII and tissue factor in detergent micelles. Proc Nat! Acad Sci USA
1986;83:299-302.
6. Morrissey JH, Fair DS, Edgington TS. Structure and properties of the human tissue factor
apoprotein. Thromb Haemostas 1987; 58: 257.
7. Fujikawa K, Suzuki K. Cellular coagulant and anticoagulant proteins. In: Poller L, ed. Recent
advances in blood coagulation 5. Edinburgh: Churchill Livingstone, 1991; 35-51.
8. Tripodi A, Arbini, A, Chantarangkul V, Mannucci PM. Recombinant tissue factor as substitute
for conventional thromboplastin in the prothrombin time test. Thromb Haemostas 1992; 67:
42-5.
9. European Concerted Action on Anticoagulation (ECAA). Poller L, van den Besselaar AMHP,
Jespersen J, Tripodi A, Houghton D. The ECAA field studies of coagulometer effects on the
lSI of ECAA thromboplastins. Thromb Haemostas. 1998; in press.
10. Kirkwood TBL. Calibration of reference thromboplastins and standardisation of the
prothrombin time ratio. Thromb Haemostas 1983; 49: 238-44.
11. Hermans J, van den Besselaar AMHP, Loeliger EA, van der Velde EA. A collaborative
calibration study of reference materials for thromboplastins. Thromb Haemostas 1983; 50:
712-17.
12. Thomson, JM, Tomenson JA, Poller L. The calibration of the second primary international
reference preparation for thromboplastin (thromboplastin, human, plain, coded BCT/253).
Thromb Haemostas 1984; 52: 336-42.
13. Poller, L, Taberner D. Dosage and control of oral anticoagulants: an international collaborative
survey. Br J Haematol1982; 51: 479-85.
14. WHO Expert Committee on Biological Standardisation, 28th Report. WHO Technical Report
Series. 1977; 610: 14-15,45-51.
15. WHO Expert Committee on Biological Standardisation, 33rd Report. WHO Technical Report
Series. 1983; 687: 81-105.
16. ICTHlICSH. Prothrombin time standardisation: report of the expert panel on oral anticoagulant
control. Thromb Haemostas 1979; 42: 1073-114.
17. Kendall MG, Stuart A. The advanced theory of statistics, 2. London: Charles Griffin, 1961.
18. Loeliger EA, Lewis SM. Progress in laboratory control of anticoagulants. Lancet 1982; 2:
318-20.
19. van den Besselaar AMHP. Multicentre replacement of the International Reference Preparation
rabbit plain. Thromb Haemostas 1993; 70: 794-9.
20. van den Besselaar AMHP, van der Velde EA. The manufacturer's calibration study. In: van
den Besselaar AMHP, Gralnick HR, Lewis SM, eds. Thromboplastin calibration and oral
anticoagulant control. The Hague: Martinus Nijhoff, 1984; 127-49.
21. Thomson JM, Darby KV, Poller L. Calibration of BCT/441, the ICSH Reference Preparation
for Thromboplastin. Thromb Haemostas 1986; 55: 379-82.

60
 PROTHROMBIN TIME

22. Tomenson IA. A statistician's independent evaluation. In: van den Besselaar AMHP, Gra1nick
HR, Lewis SM, eds. Thromboplastin calibration and oral anticoagulant control. The Hague:
Martinus Nijhoff, 1984; 87-108.
23. Hirsh I, Dalen IE, Deykin D, Poller L. Oral anticoagulants: mechanism of action, clinical
effectiveness and optimal therapeutic range. Chest 1992; 102: 312-26S.
24. Altman DG, Bland 1M. Measurement in medicine: the analysis of method comparison studies.
Statistician 1983; 32: 307-17.
25. D'Angelo A, Seveso MP, D'Angelo SV, Gilardoni F, Macagni A, Manotti C. Comparison of
two automated coagulometers and the manual tilt tube method for the determination of
prothrombin time. Am I Clin Patho11989; 92: 321--8.
26. Poggio M, van den Besselaar AMHP, van der Velde EA, Bertina RM. The effect of some
instruments for prothrombin time testing on the international sensitivity index (lSI) of two
rabbit tissue thromboplastin reagents. Thromb Haemostas 1989; 62: 868-74.
27. Ray MJ, Smith IR. The dependence of the International Sensitivity Index on the caogulometer
used to perform the prothrombin time. Thromb Haemostas 1990; 63: 424-9.
28. Chantarangkul V, Tripodi A, Mannucci PM. The effect of instrumentation on thromboplastin
calibration. Thromb Haemostas 1992; 67: 588-9.
29. Clarke K, Tabemer DA, Thomson 1M, Morris lA, Poller L. Assessment of value of calibrated
lyophilized plasmas to determine International Sensitivity Index for coagulometers. I Clin
Patho11992; 45: 58--60.
30. Poller L, Triplett DA, Hirsh I, Carroll I, Clarke K. The value of plasma calibrants in correcting
coagulometer effects on International Normalized Ratios. Am I Clin Patho11995; 103: 358--65.
31. Houboyan IL, Goguel AF. Procedure of reference calibrated plasmas for prothrombin time
standardisation. (Abstract, XIVth Congress ISTH, New York) Thromb Haemostas 1993; 69:
663.
32. van den Besselaar AMHP. The use of lyophilized calibration plasmas and control blood for
INR calculation on external quality control of the prothrombin time. Am I Clin Pathol 1994;
102: 123-7.
33. European Concerted Action on Anticoagulation. Poller L, van den Besse1aar AMHP, Iespersen
J, Tripodi A, Houghton D. Comparison of artificially deleted, lyophilized coumarin and fresh
coumarin plasmas in thromboplastin calibration. Br I Haematol. 1998; 101: 462-7.
34. European Concerted Action on Anticoagulation. Poller L, Barrowcliffe Tw, van den Besselaar
AMHP, Jespersen I, Tripodi A, Houghton D. Minimum lyophilized plasma requirement for
lSI calibration. Am J Clin Pathol, (In press).
35. Dalen JE, Hirsh 1. ACCP 4th Consensus Conference Introduction. Chest 1995; 108: 225-65.
36. Poller L, McKernan A, Thomson 1M, Elstein M, Hirsch PI, Iones ffi. Fixed minidose warfarin:
a new approach to prophylaxis against venous thrombosis after major surgery, Br Med J 1987;
295: 1309-12.
37. Kitchen S, Malia RG, Greaves M, Preston FE. Comparison of human and rabbit brain
thromboplastin in evaluation of haemostatic defect of liver disease. Lancet 1986; 2: 1463.
38. Taberner DA, Poller L, Thomson JM, Darby KY. Effect of international sensitivity index
(lSI) of thromboplastins on precision of international normalised ratios (INR). J Clin Pathol
1989; 42: 92-100.

61
7
Endogenous thrombin potential
H. C. HEMKER and S. BEGUIN

INTRODUCTION

The one and only function of the clotting system is to generate thrombin in
the amounts and at the site required for haemostasis. The endogenous thrombin
potential (ETP), is a parameter that reflects this function quantitativelyl,2. The
combined action of the procoagulant and anticoagulant systems in the :'lood
is reflected in the course of thrombin concentration in a sample of blood after
the clotting system is triggered, i.e. in the thrombin-generation curve (TGC;
Figure 7.1). Such curves are characterized by an optional lag phase, a rise to
a peak value and a decline to zero in the course of approximately 10 min.
It is the function of the enzyme thrombin to convert substrates. If a thrombin
substrate is present at a constant concentration during the whole of the TGC,
then its velocity of conversion will be proportional to the concentration of
thrombin. The course of the resulting product will therefore be represented by
a curve, the slope of which represents the reaction velocity and is proportional
to the level of thrombin (Figure 7.2, thin dotted line). No more product will

100

l~~. ".1
i'
.s

o 5 10 15 20
Time (min)

Figure 7.1 Thrombin generation curves (TGC) as obtained by the subsampling method
From left to right: PPP-Extrinsic system; PPP-Intrinsic system; PRP. The area under the curve is
defined as the endogenous thrombin potential (ETP)

63
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
150,---------------,

.. -.- ......... -...... .

~ 100
c:
:0
E
e 50
~

o 5 10 15 20 25

Time (min)

Figure 7.2 Substrate consumption during thrombin generation

Circles: Course of free thrombin; Upper (thin) dotted line: course of product from an ideal
substrate; Lower (bold) drawn line: slowly converting substrate added at high concentration (SQ68);
Dotted line: exhaustible, quickly converting substrate (such as fibrinogen in a clotting assay)

be formed after thrombin generation is over. The end-level of the product

attained will directly reflect the 'enzymatic potential' that was developed in the
sample. Its actual value depends on the area under the TGC and the kinetic
properties and concentration of the substrate. We defined the ETP as this area
under the TGC, so that it becomes a substrate-independent parameter. It is an
area in a graph of thrombin concentration versus time and therefore has the
dimension of concentration multiplied by time, i.e. nmoVL x min.
In practice the concentration of the substrate does not remain constant
during the experiment, because substrate is converted into product. So a realistic
product-generation curve will be lower than the ideal one. The difference will
be minimal if the substrate is present in high concentration and converted
slowly by thrombin (Figure 7.2, bold line) and it will be large if a low amount
of a substrate is present that is converted quickly (Figure 7.2, bold dotted line).
Fibrinogen is a good substrate of thrombin, present at a relatively low
concentration (0.5-1 ~moVL); it is converted completely during the very first
moments of thrombin generation. A clotting time therefore indicates primarily
the lag phase of thrombin generation. Under conditions where thrombin
generation starts immediately (such as in thromboplastin time measurements),
the clotting time indicates the velocity of thrombin generation during the fIrst
seconds of the process. If there is a lag phase before the thrombin burst (e.g.
in case of the APTT, the clotting time of platelet-rich plasma (PRP), or the
activated whole-blood clotting time), then the clotting time is determined by
the lag phase plus a minor contribution from the velocity of thrombin appear-
ance during the first seconds of thrombin generation.
There is not necessarily a correlation between the lag time and the ETP.
Increasing concentrations of heparin, for example, cause a similar decrease of
the ETP in the intrinsically and the extrinsically triggered clotting system, but
only influence the lag time in the former (the APTT).
The time-honoured subsampling method3 can always be used to establish a
TGC from which the ETP can be established. Timed samples are taken from
a clotting mixture, and the concentration of thrombin is measured in each
sample. We developed a semi-automated procedure but the procedure remains
64
 ENDOGENOUS THROMBIN POTENTIAL
labour-intensive and unsuitable for routine use. We therefore also developed
the continuous method, an alternative technique in which thrombin-catalysed
product formation from suitable substrates is monitored directly. In such experi-
ments the total amount of product formed by thrombin is proportional to the
ETp4.
A technical problem arises from the circumstance that part of the thrombin
in plasma binds to <X2-macroglobulin, which complex has no known physi-
ological activity but nevertheless can still cleave chromogenic substrates. The
product formation observed is the result of the combined activities of free
thrombin and <X2-macroglobulin-bound thrombin. The relevant data, i.e. the
amount of product formed by free thrombin only, can be extracted from the
experimental product formation by a simple mathematical operation (see
below). However, the operation has to be carried out on the whole course of
the product-generation curve, so we need to monitor product formation continu-
ously. This restricts the application of the continuous method, at this moment,
to the use of chromogenic substrates in an optically clear medium, i.e. in
defibrinated platelet-poor plasma (PPP). The principle is applicable, however,
to all substrates that give a product that can be detected in time, e.g. by
fluorescent, electrochemical or nuclear magnetic resonance signals. If signal-
substrates are found that are cleft by thrombin but not by the <X2-macroglobulin-
thrombin complex, the need for continuous measurement disappears and the
spectrophotometric method reduces to an end-point measurement. Such
substrates are not yet available, however. The most relevant value of the ETP
is evidently obtained in a system that most closely mimics the situation in vivo,
i.e. whole blood in the presence of homologous vessel wall. As long as this is
technically impossible we have to make a number of choices as to how the
reaction is triggered (intrinsic or extrinsic system) and to what extent components
from the vessel wall (such as thrombomodulin or TFPI) are added.
Here we present the two methods for determining the ETP: the subsampling
method, which is suitable for any type of medium (PPP 5 , PRp6 and whole
blood7) and the continuous method, adapted to a centrifugal analyser and
suitable for routine laboratory and epidemiological use, but restricted to PPP,
which needs to be defibrinated 8 •

SUBSAMPLING METHODS
Materials
Buffers
Buffer A: 20 mmollL Tris-HCI, 100 mmollL NaC!, 0.5 giL bovine serum
albumin pH 7.35.
Buffer B: same as buffer A with 20 mmollL EDTA, pH 7.9.

Stopping fluids
For plasma: I mollL citric acid or acetic acid.
For blood: 10 J.1IllollL PPACK in physiological salt solution.
65
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Triggers
Any tissue factor preparation can be used as the trigger for the extrinsic system.
A dilution that in normal plasma gives a clotting time of around 36 s will yield
a suitable Tac. As long as the thromboplastin clots normal plasma in < 120 S,
the ETP is independent of the tissue factor concentration. This corresponds
to IlJlIloVL of procoagulant phospholipid (pSIPC 1/4 weight %) and> 5 pmoVL
of tissue factor. Most commercial tissue factor preparations contain Ca2 +,
which should be taken into account when calculating the final Ca2+ concentration.
Usually polybrene is also added by the manufacturer, so effects of heparin may
go unobserved when using commercial thromboplastins. As a rule we use relipidated
recombinant human tissue factor (Innovin, Dade).
Optically clear APTT reagents, i.e. contact activating solutions containing
phospholipids, can be used as a trigger for the intrinsic system. We used Actin
FS® (Dade), a suspension of ellagic acid and soybean phosphatides.
As procoagulant phospholipid we use a suspension of phosphatidylserine
(PS) and phosphatidylcholine (PC) in 1:4 proportion at a concentration of
20 IlmoVL. The final concentration in the reaction mixture is 1 IlmoVL.

Substrates
The chromogenic substrate used for measuring thrombin in the subsampling
assayisS2238:H-D-Phe-Pip-Arg-pNA.2HCl.Theleaving-groupisparanitro-
aniline (pNA), the specific absorbance of which is 9.93 mOD per IlmollL
per cm light-path at the measuring wavelength of 405 nm. Because of the
high concentration of thrombin relative to the other proteases of the clotting
system (> 20:1), the specificity of the substrate for thrombin is a minor
concern, and cheaper and less specific substrates such as S2165 may be used
with good results.

Plasma preparation
Blood
Blood is collected on 0.13 moVL trisodium citrate; nine parts of blood to one
part of citrate solution. An open (non-vacuum) venipuncture is preferable, in
order to avoid platelet activation. For experiments in PRP an open system is
obligatory. The first millilitre of blood is to be discarded. For the continuous
method, blood is preferentially taken from fasting subjects; a recent meal may
cause marked turbidity impairing the spectrophotometric measurement.

PRP
Samples are centrifuged at 250g, at a temperature not lower than 15°C for 10
min. The supernatant PRP is pipetted off and the platelets are counted. In the
66
 ENDOGENOUS THROMBIN POTENTIAL

meantime PPP is prepared and, if required, the platelet count is adjusted to

300 000/J.1l with the autologous PPP. Temperatures lower than 15°C will damage
the platelet, induce release of platelet factor 4 (Le. heparin neutralization) and
appearance of procoagulant phospholipids.

ppp
PPP is obtained as the supernatant from blood centrifuged at 1000g, at a
temperature not lower than 15°C for 15 min. This plasma is centrifuged a
second time at the same speed and the pellet is discarded. If a pool of normal
plasmas is made (at least 10 donors), the pooled plasma is centrifuged at 4 °C
for 1 hat 23 OOOg. All plasma, if not used immediately, is stored at -80°C.

Measurement of thrombin generation

Extrinsic system, PPP

An aliquot of 240 III of PPP is incubated with 40 III of buffer A (that may
contain substances to be tested) and 20 III of 20 IlmollL procoagulant
phospholipid, until temperature equilibration. Coagulation is initiated at t =
oby addition of 60 III of CaCl2 0.1 mollL (16.7 mmollL final concentration)
containing tissue factor. Samples (10 Ill) of the reaction mixture are taken at
suitable intervals and added to prewarmed (37°C) cuvettes containing 490 III
of 200 IlmollL S2238 in buffer B. The reaction is stopped by the addition of
300 III of 1 mollL citric acid (or acetic acid) after 2 min and the optical density
(OD) is measured at 405 nm.
The sampling interval is dependent upon the velocity of thrombin generation.
In human material the usual sampling frequency is two per minute. Sampling
at 5-10 s intervals is readily possible, so up to five experiments can be carried
out in parallel. In laboratory animals (mouse, rat, rabbit, pig) thrombin
generation may be much faster than in humans and the sampling frequency
must be increased in accordance. An evolving clot is wound out on a plastic
spatula and removed.
The method can be adapted to microtitre plates if the volumes are propor-
tionally reduced. It is essential, however, that the dilution upon subsampling
is > 20 times, otherwise the interaction between thrombin and antithrombin
will continue at a rate high enough to influence the readings.

Intrinsic system, PPP

Essentially the same mixture is used as in the intrinsic system, substituting

tissue factor by a partial thromboplastin (APTT reagent), 1:10 or 1:20 diluted.
Alternatively contact activation can be obtained with kaolin. To 240 III of PPP
is added 20 J.1l of a kaolin (5 mg/ml suspension in buffer A) and 20 J.1l of buffer
A. The mixture is incubated for 4 min at 37°C and vortex-mixed every 30 s.
At 4 min, 20 III of PSIPC (20 !lIDollL) are added and at 5 min coagulation is
triggered by addition of 60 III of CaCl2 (16.7 mmollL final concentration).
67
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

PRP
An aliquot of 240 III of PRP is incubated with 60 III of buffer A (that may
contain substances to be tested) until temperature equilibration. Coagulation
=
is initiated at t 0 by adding 60 III 0.1 mmoVL CaCI2 • A low concentration of
recombinant tissue factor (1 % of the concentration used in the extrinsic system)
may be added to the CaCl2 solution in order to rule out variability caused by
the traces of thromboplastin that may be present in the PRP. This concentration
of tissue factor should give clotting times of around 10 min when added together
with Ca2 + to non-defibrinated PPP. In PRP, all experiments need to be carried
out within 60 min of venipuncture and only experiments carried out in parallel
can be directly compared, as thrombin generation is influenced by platelet
ageing.

Blood
It is possible to execute a thrombin-generation test on non-anticoagulated
blood. If the blood is obtained from an ideal venipuncture and collected in a
plastic tube a time lapse of minimally 7 min will be available before thrombin-
generation starts. Normally, however, 480 III of citrated blood is incubated with
=
120 III of buffer A. Coagulation is initiated at t 0 by adding 120 III 0.05 moVL
CaCl2 with or without recombinant tissue factor as in PRP. With blood,
subsamples of 20 III are usually taken. The stopping fluid is 10 llffioVL PPACK
in an isotonic salt solution. After stopping the tubes are centrifuged (5 min at
1000g) and pNA is determined photometrically in the clear supernatant. In
the course of the clotting process a 10% change in the haematocrit occurs. If
optimal precision is required this can be compensated for by hydrolysing the
sedimented erythrocytes in each tube and calculating the change in erythrocyte
content from the haemoglobin measured spectrophotometrically.

Equipment
A water bath at 37°C is used to contain both the plasma and the tubes with
chromogenic substrate. If experiments with whole blood are carried out it is
preferable to fill the bath with isotonic salt solution, so as to prevent haemolysis
if a drop of bath fluid should accidentally fall into an erythrocyte-containing
tube. In order to avoid the need for accurate manual timed sampling, and the
errors that it may cause, we equipped both the sampling and the stopping
pipette with a pushbutton that is connected to a personal computer, and that
records sampling and stopping times. We established a program in which the
spectrophotometer is also connected to this computer, so that the increase of
OD over time in the cuvette is calculated from the OD of each tube and the
incubation time. Amidolytic activities, expressed as the equivalent amount of
a-thrombin, are calculated via a calibration factor obtained with active site
titrated human a-thrombin.
68
 ENDOGENOUS THROMBIN POTENTIAL

Calculation of thrombin-generation curves

The u2-macroglobulin present in normal human plasma binds part (about

20%) of the thrombin formed and thus inhibits its biological, but not its
amidolytic, activity. So during thrombin generation a persistent enzymatic
activity builds up that cannot be distinguished from thrombin, but that disturbs
direct measurement of the thrombin potential. The concentration of u2-macro-
globulin (2-4 J.lmollL) is much higher than that of thrombin (0.1-0.3 J.lmollL
at the peak of thrombin generation), so the velocity of complex formation is
proportional to the concentration of thrombin. This relation can be used to
dissect the course of amidolytic activity into its two constituent curves: that
of free thrombin and that of the complex4 ,8,9.
Suppose that at a given moment sample number n is taken, and that the
amidolytic activity is A(n). This activity is the sum of the activity of free
thrombin, present at that moment (F(n)) and the activity of the complex (C(n)).
At the next sampling (n + 1) the amidolytic activity measured is A(n + 1). The
increase (or decrease) of amidolytic activity (M) consists of two components:
in the first place the increase (or decrease) of free thrombin (tlF) and in the
second place the increase of the complex (tlC), so that tlA = tlF + tlc.
Because the complex formation is proportional to the concentration of free
thrombin (proportionality constant = k), we know that tlC = k . F(n). So if
we know C(n) and F(n), we can calculate C(n + 1) because C(n + I) = C(n)
+ tlC = C(n) + k . F(n). Knowing C(n + I) allows us to calculate F(n + I)
because the observed amidolytic activity at moment n + I is the sum of C(n
+ I) and F(n + I), i.e. A(n + I) =F(n + 1) + C(n + I) or F(n + 1) = A(n + 1)
- C(n + 1) = A(n + 1) - C(n) - k·F. In this simple way we can calculate the
values of C and F at any moment (n + I) if we know their values at the
previous moment (n), the amidolytic activity at the moment n + I and the
constant k.
At the beginning of the experiment, the values A(O) and F(O) are zero.
Starting from these, the following values can be calculated if k is known. The
'constant' k varies with the u2-macroglobulin concentration in the plasma, the
substrate used and the conditions of the test (substrate concentration, pH etc.).
It therefore should be determined in each (set of) experiment(s). If the experiment
is continued long enough for no free thrombin to be left, that value of k is the
correct one that makes the curve of free thrombin (F(n)) descend smoothly to
zero towards the end of the experiment, without attaining negative values
(other than by experimental scatter around zero). In practice five or 10 last
values of F(n) are summed and k is varied until this sum is as close as possible
to zero. The calculations are easy to program and also readily done in a
spreadsheet. Table 7.1 shows how this should be organized. Advanced
spreadsheets contain a tool (e.g. Solver in Excel) that can find the correct
value of k automatically. For our normal pool the value of k is 0.288 ±
0.012, therefore 0.3 is a good starting estimate. In the program the constant
should be adapted to the sampling interval, i.e. 0.5 x 0.29 when this interval
is 30 s, etc.
If the data are amidolytic activities from a subsampling experiment, expressed
69
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Table 7.1 Calculation of the ETP from an amidolytic activity curve (subsampling method).

A B C D E F

1 N Time Data Thrombin (12M k

2 I 0 0 0 0 =sum(D25:030)
3 2 0.5 I =C3-E2 =E2 + ($F$l *03)
4 3 1 m =C4-E3 =E3 + ($F$l *D4)
5 4 1.5 n =C5-E4 =E4 + ($F$l *05)
etc.
22 21 10 u =C22-E21 =E21+ ($F$I*D22)
etc.
32 31 15 z =C31-E30 =E31 + ($F$I*032)

The sampling times are entered in column B and the corresponding amidolytic activities are
entered in column C. In column E the amount of u2M-thrombin is calculated and in column D
the amount of free thrombin, using an assumed value of k (in cell $F$l). In cell F2 the free
thrombin concentrations, calculated in column D, are summed over a range of sampling points
(here 25:30) where free thrombin can be assumed to have reached the zero level. The correct
value of k (Fl) is found by changing it (Le. $F$l) until F2 is as close as possible to zero. The
ETP is obtained as the sum (D3 ... D32) multiplied by the sampling interval in min ( here 0.5 )

as nmollL thrombin, the sum of all values in column D, multiplied by the

sampling interval in minutes, gives the ETP.
This calculation splits the experimental data into two curves (Figure 7.3).
The procedure lacks mathematical rigour but will work in practice. A more
sophisticated treatment is beyond the scope of this chapter. It should be noted
that the kinetic properties of the <Xz-macroglobulin-thrombin complex are not
the same as those of thrombin. The concentrations of C are expressed, however,
as the equivalent concentrations of thrombin; k contains a factor compensating
for this effect. This is one more reason for k to be dependent on the experimental
circumstances.

150 y-----------------,

~ 100
.s
c:
:c
E
2
('3. 50

5 10 15 20
Time (min)

Figure 7.3 Dissection of a thrombin generation curve as obtained in the subsampling method
Circles: Experimental amidolytic activity; Bold line: Course of free thrombin; Thin line: Course
of (12 -macroglobulin-thrombin complex

70
 ENDOGENOUS THROMBIN POTENTIAL
CONTINUOUS, SPECTROPHOTOMETRIC METHOD

Principle
The concept behind ETP determination is simple. Thrombin being an enzyme,
its function is quantified by the amount of substrate that it converts. To a
volume of defibrinated PPP we add an artificial substrate of thrombin. Then
the clotting system is triggered and the amount of substrate that is converted
by thrombin during its transient presence in the sample is continuously moni-
tored. The concentration and the kinetic properties of the substrate are chosen
such that during the rise and fall of the thrombin concentration the velocity
of product (= colour) formation is always practically proportional to the
concentration of thrombin. In this way the amount of product formed is
proportional to the surface under the TGC, i.e. to the ETJi4, which is a direct
indicator of the amount of natural substrate that thrombin can convert when
it is generated in a volume of clotting blood in vivo.

Materials and methods

Except where mentioned below, the same materials are used as in the subsampling
method.

Substrates
For the continuous method substrates are required that are converted by
thrombin at a suitable rate and that are not readily converted by abundant
proteases such as kallikrein. Those that are split by factor Xa with reaction
velocities similar to those of thrombin (Table 7.3) can be used because the peak
concentration of factor Xa in clotting plasma is < 5% of that of thrombin. Yet
these substrates should not bind with high affinity to factor Xa because that
would make them efficient factor Xa inhibitors. We used commercially available
Malonyl-a-aminoisobutyryl-arginine para-nitroanilide methyl ester hydro-
chloride (CH 3 0-CO-CH z-CO-Aib-Arg-pNA . HCI; SQ 68) from Serbio
Laboratories, France (European Patent 88400304.7). Other suitable substrates
were synthesized in our laboratory (Table 7.3), using a procedure described in
detail elsewhere10-1Z. The chromogenic leaving-group is always para-nitroaniline,
the specific absorbance of which is 9.93 mOD per lJllloVL per cm light-path
at the measuring wavelength of 405 nm. We determined conversion factors
between the 00 end-level and the ETP in nmoVL x min with each of the
substrates 8 (Table 7.4).
Until 10% of the substrate is converted by free thrombin a linear relation
between ETP and 00 end-level can be assumed. The permitted upper limit of
product is therefore 50 lJllloVL, which sets an upper limit of 500 mOD to the
00 end-level. Usually an increase of 10 mOD is the lowest that can be reliably
measured against the yellow background of plasma. Depending upon the
kinetic properties of a substrate the range of ETP values that can be measured
71
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
in the OD window of 10-500 mOD is different (Table 7.4). Substrates 1-5 are
suitable for measurements in a wide range around the normal level, whereas
substrates 6 and 7 can be used to assess with high precision the effect of
inhibitors (e.g. activated Protein C (APC), heparin) or of anticoagulant therapy,
but ETP that are higher than normal will be underestimated.
Caution! Considerable errors can be introduced if substrates are used that have
not been selected and characterized according to the principles outlined here.

Defibrination
Defibrination is required to avoid disturbance of the OD recording by the
turbidity of emerging fibrin. If fibrin polymerization can be avoided in a way
that does not interfere with thrombin generation, defibrination is no longer
required. For defibrination we use snake venom enzymes (the only action of
which, in the coagulation system, is on fibrinogen), such as Reptilase and
Ancrod. A Reptilase solution (Boehringer Mannheim, Germany) is prepared
so that 20 ~, added to 1 m1 of plasma, causes a clot in 4 min. Higher concentra-
tions should be avoided because they may activate factors V and/or VIII and
thereby increase the thrombin potential.
Ancrod, the fibrinogen clotting enzyme of the Malayan pit viper, was obtained
as the commercial preparation Arvin® (Knoll AG, Ludwigshafen, Germany).
It is used at a final concentration of 1 Ulml, causing a clotting time of around
3 min in normal plasma. It does not activate clotting factors or platelets. It
also does not cause haemolysis and therefore can be added directly to the tube
in w4ich blood is taken. In this way defibrinated plasma can be obtained directly
upon centrifugation of the blood.
Defibrination is carried out as follows: plasma samples are mixed with 1:50
volume of the Reptilase or Ancrod solution and incubated for 10 min at 37°C,
then kept on ice for 10 min. The fibrin is wound out on a small plastic spatula.
Unlike Reptilase, Ancrod does not cause haemolysis and can be added to whole
blood, rendering defibrinated plasma directly upon centrifugation. In large
series of samples, defibrination is the rate-limiting step of the procedure.

Measurement
The reactions can be carried out in any laboratory automaton capable of
measuring with sufficient precision the course of the optical density at 405 nm
at 30 s intervals during 12-15 min. We used the Cobas Bio and Cobas Fara
centrifugal analysers (Hoffmann-La Roche; a listing of the required settings
of the instruments is available from the authors).
The reaction mixture consists of four parts of defibrinated plasma, that are
mixed with one part of (partial) thromboplastin solution and one part of start
solution. In the Cobas machines the volumes are 80 ~ of defibrinated plasma
(or 75 ~ defibrinated plasma to which may be added 5,.11 of a substance to be
tested) and 20 ~ of the thromboplastin, without Ca2+. Mter 30 s of incubation,
72
 ENDOGENOUS THROMBIN POTENTIAL

which is sufficient for temperature equilibration, thrombin generation is started

by adding 20 ~l of a prewarmed start solution containing 0.1 mollL of CaCl2
and 3 mmollL of substrate so that the final concentration becomes 500 ~01lL.
In each run of 28 samples, three normal pool plasmas are included, usually at
positions I, 14 and 28.
After starting the reaction the optical density at 405 nm is recorded at
intervals of maximally 30 s for at least 15 min. Examples of the curves obtained
are shown in Figure 7.4; this figure also shows the results of the splitting of
the curve in the thrombin part and the a2-macroglobulin-thrombin complex
part.

Data handling
The centrifugal analyser is connected to a personal computer via an installed
option that makes the results available on a communication line following the
RS-232 standard. The data are in the form of a succession of ASCII characters
representing the series of OD readings, the time of the reading is given by the
position in the list, knowing the sampling interval. The data are used as input
for a program (Pascal or Basic), using a function for character extraction from
the RS-232 interface buffer of the personal computer and transformed to a
standard format file which gives the OD as a function of time and from which
the ETP can be calculated off-line. Programs for reception-recording on
IBM-compatible PCs are available from the authors.

Calculation of the ETP from the continuous method

In the continuous method the progress curve of product (i.e. OD) should be
split into the curve of product formed by free a-thrombin and that formed by
the a2-macroglobulin-thrombin complex. For mathematical reasons that will
be obvious to those interested, the splitting of curves discussed above can also

5 10 15
Time (min)

Figure 7.4 Dissection of a product-generation curve as obtained in the continuous method

Circles: Experimental product formation; Bold line: Course of product converted by free thrombin;
Thin line: Course of product converted by o2-macroglobulin-thrombin-complex.

73
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
be applied to such progress curves (Table 7.2). Only the criterion for a zero
end-level of thrombin differs. If thrombin attains the zero level, the amount
of product formed by thrombin will be constant. To test this, column F
(Table 7.2) is added and Fl is changed until cell F2 is as close as possible to
zero. This can again be done by trial and error or via a spreadsheet tool. The
fixed end-level of product formed by free thrombin is proportional to the ETP.
This calculation renders the amount of substrate converted by free thrombin.
Theoretically the relation between the area under the TGC (i.e. the ETP)
and the total amount of substrate converted is given by ETP A . Kmlkcat • =
S5. Application of this formula requires exact knowledge of Km and kcat of
the substrate in plasma. These are difficult to obtain due to the instability of
thrombin in the presence of natural inhibitors. Data such as those in Table 7.3
are tentative values, only suited to obtain a global impression of the reaction
behaviour that is to be expected. We therefore determined the proportionality
between the 00 end-level and the ETP by assessing the ETP independently
from a TGC obtained by subsampling. This procedure was used to obtain the
conversion factors in Table 7.4. In actual practice it is not necessary to obtain
the absolute value of the ETP as long as it is possible to compare the results
of the test plasma with those of a standard normal (pool) plasma tested in
parallel, and expressed as a percentage of normal. A home-made normal
plasma pool should be used for this purpose as long as we do not have proof
that commercial control plasmas can be used.

Reference values
Under our experimental conditions, plasmas from normal donors are found
=
to give an 00 end-level of 54.8 ± 8.1 mOD (n 118) from SQ68 and 108.1 ±
Table 7.2 Calculation of the ETP from an optical density curve (continuous method).

A B C D E F

Product from Product from

1 N Time Data Thrombin Ct_2 _M - Thrombin k
2 I 0 0 0 0 =sum(F25-F30)
3 2 0.5 I =C3-E2 =E2 + ($F$1 *D3)
4 3 1 m =C4-E3 =E3 + ($F$I*D4)
5 4 1.5 n =C5-E4 =E4 + ($F$1 *D5)
etc.
22 21 10 u =C22-E21 =E21+ ($F$1*D22) =(D22-D21h
etc.
32 31 15 z =C31-E30 =E31+ ($F$1*D32) =(D32-D31h

The sampling times are entered in column B and the corresponding optical densities are entered
in column C. In column E the amount of product due to Ct2M-thrombin activity is calculated
and in column D the amount of product due to the activity of free thrombin, using an assumed
value of k (in cell $F$I). In column F, from line 10 on, the differences between cells in column
E are entered (i.e. the first derivative of the function in column E). In cell F2 the derivative
calculated in column F are summed over a range of sampling points (here 25:30) where free
thrombin can be assumed to have reached the zero level. The correct value of k (Fl) is found by
changing it (i.e. $F$I) until F2 is as close as possible to zero. The ETP is proportional to the
end level in column D. The proportionality constant is dependent upon the substrate used

74
 ENDOGENOUS THROMBIN POTENTIAL
Table 7.3 Kinetic constants of chromogenic substrates used for measuring the ETP.

No. Substrate Thrombin Factor Xa

K.n kcat RV K.n k cat RV

(mmoIIL) (S-I) (mmoIIL) (S-I)

MZ.Aib.Arg.pNA 0.53 0.37 1.59 0.59 0.73

(8Q68)
2 H.Val.Arg.pNA 1.2 0.20 0.33 3.8* 0 0
3 Msc.Val.Arg.pNA 1.28 1.60 2.5 1.24 0.75 1.2
4 H.Glu.Gly.Val.Arg.pNA 3.3 3.50 2.6 1.4 0.9 1.3
5 H.Glu.Val.Arg.pNA 0.90 1.40 2.8 1.5 1.1 1.5
6 DEMZ.Gly.Arg.pNA 0.23 2.57 9.8 0.65 2.9 7.0
7 Boc.Gly.Val.Arg.pNA 0.20 11.2 45 1.0 1.9 3.5
8 HD.Phe.Pro.Lys pNA 0.04 90.0 464 >1 <1 <2

RV: Reaction velocity at 500 jllIloVL substrate concentration, relative to thrombin action on
MZ-Aib-Arg-pNA (8Q68); MZ: malonic acid monomethyl ester; DEMZ: diethyl malonic acid
monomethyl ester; Msc: 2-(methylsulphonyl)ethyloxycarbonyl; Boc: tert-butyl oxycarbonyl; *)
substrate 2 is not hydrolysed by factor Xa, the K.n mentioned is the binding constant measured
from competitive inhibition of 8Q 68. Note that some of these values differ from those
reported earlier because they represent results obtained with more advanced methods (R.
Wagenvoord, work in progress)

=
16.8 mOD (n 84) from Msc-Val-Arg-pNA. With the conversion constants
of Table 7.4, this represents an endogenous thrombin potential of
384.8 ± 51.7nmollL x min and 394.3 ± 62.3 nmoVL x min, respectively. The
normal value of the ETP in the intrinsic system appears not significantly higher:
414 ± 41 nmoVL x min.

Table 7.4 Relation between OD signal and ETP with different chromogenic substrates.

No. Substrate End level Substrate Conversion ETPRange

(mODI converted factor (%)
em) (%)
MZ.Aib.Arg.pNA (8Q68) 55.2± 1.1 6.9 20-1000
2.3
2 H.Val.Arg.pNA 86.9± 1.8 4.4 12-600
6.6
3 Msc.Val.Arg.pNA 108 ± 7.7 2.2 3.5 10-500
4 H.Glu.Gly.Val.Arg.pNA 33 ±4 0.7 12 30->1000
5 H.Glu.Val.Arg.pNA 168 ±4.9 3.4 2.3 10-500
6 DEMZ.Gly.Arg.pNA 587 ± 17 12 0.65 2-100
7 Boc.Gly.Val.Arg.pNA # (1200) 50 0.014 I-50
8 HD.Phe.Pro.Lys.pNA # (13000) 100 0.1-2

The third column gives the end-level of the thrombin dependent OD-signal with a normal pool
plasma (mean ± 8EM; n = 24 or more); the fourth column gives the % of the initial amount of
substrate (500 jllIlollL) that is converted in this process; the conversion factor is the multiplier
to calculate the ETP from the final OD level; the range gives the ETP values (% of normal) that
can be reliably measured with that substrate.
For substrates 7 and 8, the figures for the normal values have been calculated via the catalytic
efficiencies relative to 8Q688

75
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
The within-run imprecision of the determination, i.e. the standard error of
a single value calculated from 20-24 identical samples determined in the same
run on a Cobas instrument, is about 4% (extrinsic 3.5%, intrinsic 4.4% (n =
22». The between-run imprecisions, obtained when the same normal plasma
is tested in different runs on different days, is on the mean 1% higher (extrinsic
=
system 4.7% intrinsic 5.6% (n 58».
Hepes buffer and Tris buffer give the same normal ETP value in the extrinsic
system (403.8 nmoVL x min ± 12.4 versus 398.4 nmoVL x min ± 14.1; n 17), =
in the intrinsic system Hepes buffer may give slightly lower values than Tris
buffer (399.0 ± 10.5 nmoVL x min versus 435.2 nmoVL x min ± 16.6; n 17). =
In normal PRP, measured under our conditions, the normal ETP is 411 ±
53 nmoVL x min (mean ± SEM, n 28). =

Spectrophotometric method using the end-level of

o2-macroglobulin-thrombin complex
The o2-macroglobulin in the plasma can be considered as a slow-reacting
thrombin substrate, present in excess. The amount of o2-macroglobulin-
thrombin present after thrombin generation (in serum) is therefore a direct
indicator of the ETP of the original plasma4 ,13. If the o2-macroglobulin in
plasma of different individuals and at different moments were constant, then
the amidolytic activity of the o2-macroglobulin-thrombin complex would
immediately render the ETP. This is not the case, however, and the correlation
between ETP and end-level of ormacroglobulin-thrombin, although highly
significant, is not close enough to make the methods equivalent4 •
If two samples are to be compared that only differ as to an addition, we can
obtain the ratio of the ETP with and without addition directly from the ratio
of the 02-macroglobulin-thrombin levels in the serum.
This approach has successfully been applied to assess the functioning of the
protein C pathway by measuring the effect of adding thrombomodulin or
APC 13 ,14. The APC sensitivity ratio thus obtained is much more precise than
that from the APTT. It allows, for example, differentiation between the degree
of acquired APC resistance induced by different types of oral contraceptives 14•
Another application of this method is the estimation of the effect of drugs.
The effect of drugs added to plasma in vitro is directly rendered by the ratio
of the o2-macroglobulin-thrombin complex end-levels of the sample with the
inhibitor and the blank. Effects in patients or volunteers can also be measured,
if it can be assumed that the o2-macroglobulin in a patient or volunteer will
be constant over the time-course of action of the drug, so that all samples can
=
be compared to the t 0 sample.

Execution
The execution of the experiment is exceedingly simple. The thrombin-like
activity is determined, as in the samples from a manual thrombin-generation
experiment (see above) but only in one sample, after 15-20 min of incubation.
76
 ENDOGENOUS THROMBIN POTENTIAL
The ratio is determined between the activity in the sample after addition or
administration of the modifier and the blank or zero time plasma.

Pathophysiological aspects
Hyper- or hypocoagulability can be brought about by a large number of different
pathological conditions, and there exists a large spectrum of laboratory tests
to locate the underlying cause. To our knowledge, however, there is not yet a
single laboratory parameter that increases in all forms of hypercoagulability
and similarly decreases in all forms of hypocoagulability. Such an assay would
be useful for the detection of plasma-based thrombotic tendency and for the
surveillance of drug-induced hypocoagulability.
We think that the ETP is a good candidate as a general indicator of
coagulability l-6. The ETP is decreased to 15% and 35% of normal by oral
anticoagulation (international normalized ratio 2.5--4.0) as well as by heparin
administration (APTT 1.5-2.5 x control)s. Contrary to the classical tests, the
ETP also renders the effect of mixed oral anticoagulant and heparin treat-
ments. Preliminary results indicate that the effect of dermatan sulphate and
direct irreversible thrombin inhibitors (hirudin) is also truthfully rendered by
the ETP.
The ETP is increased in untreated subjects with congenital antithrombin
deficiency8, in persons with the prothrombin Leiden mutation (p. Kyrie, personal
communication), in persons with the factor V Leiden mutation 16 , and in women
using oral contraceptives 14 • In deep-vein thrombosis (phlebographically con-
firmed), it is increased by 29.4% (extrinsic) and 53% (intrinsic)s. In (angiographi-
cally assessed) coronary artery disease the increase is 10% and 17% respectivelys.
The APe system is only partly active in isolated plasma because
thrombomodulin (TM) is linked to the vessel wall. TM or APe can be added
to the ETP assay to reveal APe resistance 13- 16 . The ratio of the ETP plus and
minus TM or APe is a sensitive indicator for disturbances of the APe system.

References
1. Hemker HC. Thrombin generation, an essential step in haemostasis and thrombosis. In:
Bloom AL, Forbes CD, Thomas DP, Tuddenham EGD, eds. Haemostasis and thrombosis.
Edinburgh: Churchill Livingstone, 1993; 477-90.
2. Hemker HC, Beguin S. Thrombin generation in plasma: its assessment via the endogenous
thrombin potential. Thromb Haemostas 1995; 74: 134--8.
3. Biggs R, Macfarane RG. Human blood coagulation and its disorders. Oxford: Blackwell
Scientific Publications, 1967; 67.
4. Hemker HC, Wielders SJH, Kessels H, Beguin S. Continuous registration of thrombin
generation in plasma, its use for the determination of the thrombin potential. Thromb
Haemostas 1993; 70: 617-24.
5. Hemker HC, Willems GM, Beguin S. A computer assisted method to obtain the prothrombin
activation velocity in whole plasma independent of thrombin decay processes. Thromb
Haemostas 1986; 56: 9-17.
6. Beguin S, Lindhout T, Hemker HC. The effect of trace amounts of tissue factor on thrombin
generation in platelet rich plasma, its inhibition by heparin. Thromb Haemostas 1989; 61:
25-9.

77
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
7. Kessels H, Beguin S, Andree H, Hemker HC. Measurement of thrombin generation in whole
blood - the effect of heparin and aspirin. Thromb Haemostas 1994; 72: 78-83.
8. Wielders S, MukheIjee M, Michiels J, Rijkers DT, Cambus JP, Knebel RW, Kakkar V, Hemker
HC, Beguin S. The routine determination of the endogenous thrombin potential, first results
in different forms of hyper and hypocoagulability. Thromb Haemostas 1997; 77: 629-36.
9. Kessels H, Willems G, Hemker HC. Analysis of thrombin generation in plasma. Comput
BioI Med 1994; 24: 277-88.
10. Rijkers DTS, Adams HPHM, Hemker HC, Tesser G. A convenient synthesis of amino acid
p-nitroanilides, synthons in the synthesis of protease substrates. Tetrahedron 1995; 51: 11235--50.
11. Rijkers DTS, Wielders SJH, Tesser GJ, Hemker He. Design and synthesis of thrombin
substrates with modified kinetic parameters. Thromb Res 1995; 79: 491-9.
12. Rijkers DT, Hemker HC, Tesser GJ. Synthesis of peptide p-nitroanilides mimicking fibrinogen
and hirudin binding to thrombin. Design of slow reacting thrombin substrates. Int J Pept
Protein Res 1996; 48: 182-93.
13. Duchemin J, Pittet JL, Tartary M, Beguin S, Gaussem P, Aihenc Gelas M, Aiach M. A new
assay based on thrombin generation inhibition to detect both protein C and protein S deficiencies
in plasma. Thromb Haemostas 1994; 71: 331-8.
14. Rosing, Tans G, Nicolaes GA, Thomassen MC, van OerIe R, van der Ploeg PM, Heijnen P,
Hamulyak K, Hemker He. Oral contraceptives and venous thrombosis: different sensitivities
to activated protein C in women using second and third generation oral contraceptives. Br J
Haematol. 1997; 97: 2338-44.
15. Nicolaes GA, Thomassen MC, Tans G, Rosing J, Hemker HC. Effect of activated protein C
on thrombin generation and on the thrombin potential in plasma of normal and APe resistant
individuals. Blood Coag Fibrinolysis 1997; 8: 28--38.
16. Rotteveel RC, Roozendaal KJ, Eijsman L, Hemker He. The influence of oral contraceptives
on the time integral of thrombin generation (thrombin potential). Thromb Haemostas 1993;
70: 959-62.

78
8
Fibrinogen
M. P. M. DE MAAT, G. D. O. LOWE and F. HAVERKATE

INTRODUCTION

Blood coagulation ultimately depends on the thrombin-induced conversion of

fibrinogen to fibrinl,2. Fibrinogen is a dimeric symmetric macromolecule. Each
half consists of three different polypeptide chains, An, B~ and y, linked together
by disulphide bridges. The two sets of chains are connected at the aminoterminal
region by three disulphide bridges, two between the y-chains and one between
the n-chains. Fibrinogen is primarily the substrate of thrombin action resulting
in the release of both fibrinopeptide A and fibrinopeptide B from the
aminoterminal end. This release of the fibrinopeptides converts fibrinogen to
fibrin.

PATHOPHYSIOLOGY

Less specific cleavages of fibrinogen may occur in vivo 3 . For example, during
treatment with streptokinase or urokinase there is extensive fibrinogenolysis.
But even the plasma of healthy individuals shows some evidence of fibrinogen
cleavage4,5. It is assumed that fibrinogen is synthesized as high molecular weight
(HMW) fibrinogen with two intact carboxyl ends of the An-chains and becomes
degraded in the circulation. From the 340 kD HMW fibrinogen, the low
molecular weight (LMW) fibrinogens of about 305 kD and low' molecular
weight (LMW') fibrinogen of about 270 kD result after the removal of a 35
kD carboxyterminal polypeptide from one An-chain and from both An-chains,
respectively6,7. The identity of the proteolytic enzymes responsible for the
in-vivo generation of LMW fibrinogen under physiological conditions is at
present uncertain, but analysis of the carboxyl terminus of the An-chain has
excluded plasmin, gelatinase and trypsin 8,9. According to Holm and Godal6
the physiological distribution of heterogeneous fibrinogens in normal human
79
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
plasma is 69.7%, 26.5% and 3.8% for HMW, LMW and LMW' fibrinogen
respectively6. This distribution has been confirmed by Furlan et al.1O. The
clotting times of LMW and LMW fibrinoren are prolonged compared with
the clotting time of HMW fibrinogen 7,l . The role of HMW and LMW
fibrinogen in the pathophysiology is unknown.
Fibrinogen is necessary for clot formation and appropriate platelet function
in vivo. When measured by clotting-rate assays, its normal plasma level ranges
between 1.7 and 4.0 gIL, corresponding to 5--12 JIDloVL. The Michaelis-Menten
constant Km for the cleavage of the Aa_chain I2- 15 by thrombin is about 7
J.Ull0VL. Therefore, the velocity of clot formation is significantly affected by
different fibrinogen concentrations even within the 'physiological' range, and
contributes to the increased thrombotic risk associated with higher plasma
fibrinogen concentration.
Low fibrinogen is an important diagnostic and pathophysiological variable
in various clinicopathological conditions l - 3, e.g. disseminated intravascular
coagulation (DIC), liver disease, thrombolytic or defibrination therapy. Therefore,
its measurement is often required by the clinician as an emergency test.
In addition, epidemiological studies have been accumulating convincing
evidence l 6--19 that fibrinogen is an independent risk factor for cardiovascular
disease. An increase of the plasma fibrinogen of 1 standard deviation (15%)
approximately doubles the cardiovascular risk. Fibrinogen levels are associated
with many environmental, lifestyle and physiological risk factors, such as
smoking, body mass index, exercise, age, gender, race20--22, stress23 . It has been
suggested that the association of some of these factors, such as smoking, with
coronary artery disease may to some extent be mediated by increased fibrinogen
levels. The genetic contribution to the plasma level of fibrinogen is calculated
to be between 20% and 50%24,25 and several polymorphisms in the fibrinogen
genes have been identified that are related to the plasma fibrinogen level and the
regulation thereof4,26--28. Several drugs have been reported to reduce plasma
fibrinogen levels, such as fibric acid derivatives and ticlopidine29--32. The mechanism
by which these drugs lower fibrinogen has not yet been elucidated, but lowering
the fibrinogen levels by drugs is an interesting concept in risk reduction.

FIBRINOGEN ASSAYS

Numerous methods have been published for the determination of fibrinogen

in plasma33 . They are based on different principles: clotting rate assays34,
clottable protein-based assays35--37, assays based on changes in turbidimetz
and light-scattering during clot formation 38--43, immunological techniques
and heat or salt precipitation-based assays45. Clotting rate assays and assays
based on changes in turbidity and light-scattering (prothrombin time-derived
methods) are now the most used in clinical practice. Assays measuring total
protein concentrations are mainly used when a dysfibrinogenaemia is suspected.
Immunological assays of fibrinogen with polyclonal antibodies required multiple
steps, because fibrin(ogen) degradationJroducts (FOP) were also determined,
which resulted in spuriously high levels . With specific monoclonal antibodies
it has now become possible to determine fibrinogen levels in plasma directly47.
80
 FIBRINOGEN
Immunological and heat precipitation methods measure global fibrinogen and
there are i~dications that thes~ methods may be better gredictors of cardio-
vascular dIsease than the clottmg rate (Clauss) assay48-5 .

Reference values
The normal range of plasma fibrinogen by clotting assays is 1.7 to 4.0 giL.
The distribution of these values is slightly skewed to the right. Immunological
and heat precipitation methods give a higher normal range of 2.5-6.0 gIL.

CALIBRATION AND COMMERCIAL FIBRINOGEN STANDARDS

Before an international fibrinogen standard became available, Furlan and

co-workers lO found deviations of up to 80% of the declared values from the
measured clottable fibrinogen content in a careful study of 10 commercially
available fibrinogen standards. Fortunately, the first International Fibrinogen
Standard (89/644) (NIBSC) has recently become available.
A modification of the method for clottable protein, described by lacobson5I ,
was used as the recommended procedure in the calibration of the International
Standard, since it involved the isolation of the biologically active or clottable
fibrinogen from the plasma using thrombin and because it proposed a simple
spectrographic procedure for the quantitation step52.

CLOTTING RATE ASSAY

Principle
The chronometric assay by Von Oauss34 is based on the clotting time of a mixture
of diluted plasma and a standardized thrombin solution. The recorded clotting
times depend on the plasma concentration of functional fibrinogen. The plasma
dilution reduces the influence of inhibitors, i.e. heparin, but not the retarding
effects of FDP. The fibrinogen values in giL are obtained by comparing the
clotting times of unknown plasma samples with serial dilutions of a reference
plasma with known fibrinogen concentration according to bioassay principles.

Materials
Isotonic barbital-acetate buffer pH 7.35 (BAB)
1. Stock solution: 9.71 g sodium acetate. 3H20 and 14.71 g sodium 5,5-diethyl-
barbiturate in 500 ml distilled water.
2. Working solution (barbital-acetate buffer BAB): 250 ml stock solution, 200
ml NaCI4.25% (w/v), 217 ml HCI 0.1 mollL and 683 ml distilled water. The
pH may be corrected with very small amounts of HCI.
Thrombin reagents are now widely available.
81
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Procedure
Citrated plasma (fresh or frozen) is used for the assay. The plasma to be tested
is diluted 1 + 9 with BAB immediately before the assay.
0.1 mIl + 9 diluted citrated plasma
incubate 30 s at 37°C
add 0.1 mI thrombin solution 33 IUlmI
record the clotting time, using a coagulometer.

Calculation of results
The calibration curve of the Clauss assay is obtained from the clotting times
of serial dilutions (1 + 5, 1 + 9, 1 + 19, 1 + 29, 1 + 49) of a standard plasma
with known clottable fibrinogen concentration. Double logarithmic paper is
used to plot times on the ordinate against the concentrations (giL) on the
abscissa. The clotting times of the diluted test plasma samples are then converted
into fibrinogen concentrations. The calibration curve will be approximately
linear in the log-log plot (Figure 8.1).

Precision and reliability

Each test is done in duplicate and must be repeated if the clotting times differ
by more than 1.0 s. Appropriate measurements can be obtained only within
the range of 0.8-4.0 giL of the calibration curve. Greater plasma dilutions
(1 + 19, 1 + 29 or 1 + 39) may be required for high fibrinogen values. Vice-versa,
values below 0.8 gIL should be repeated with a I + 4 plasma dilution. For the
calculation of the results any deviation from the regular 1 + 9 plasma dilution
must be taken into account by the appropriate multiplication factor. In a quality
control study performed for the ECAT Angina Pectoris Study the coefficients
of variation from day to day are 3.3% in the normal fibrinogen range 53 •
Excessive heparin levels (> 0.8 UlmI) may give falsely low fibrinogen values,

50r-----------------~

'6'
!CD 10
..,E
go
'i3
.g

1~~~----~~~~~

0.5 10
flbrlnogen (g/l)

Figure 8.1 Typical example of a reference curve of the clotting rate assay by Clauss

82
 FIBRINOGEN

and heparin will have to be neutralized by polybrene (1,5-dimethyl-1,5-

diazundeca-methylene polymethobromide)54. Clauss fibrinogen is usually much
lower than total clottable or immunologically measured fibrinogen in dys-
fibrinogenaemia with slowly clotting abnormal fibrinogen 3,ss. This also occurs
in conditions with high levels of fibrin degradation products (FDP), such as
disseminated intravascular coagulation and therapeutic fibrinolysis, although
no effect of FDP below 190 J1g1ml was found in samples containing normal
fibrinogen levelss6 .

PROTHROMBIN nME-DERIVED METHODS

Mter the introduction of coagulation automates in the haemostatic laboratory,

the prothrombin-time (PT)-derived fibrinogen method has become popular.
The assay is based on the change in turbidity or light-scattering after complete
clotting of the plasma. In practice, after a PT (see Chapter 6), the turbidity or
light-scattering is measured for some minutes until it is stable, and the fibrinogen
concentration is calculated from the total change in turbidity or light-scattering.
This method is cheap, especially because the PT is usually standard in a routine
haemostasis package. The correlation between the clotting rate method (Clauss
method) and the PT-derived method is usually very goodS7- s9 , but there may
be problems in samples that did not clot completely after the programmed
incubation time, e.g. heparin-containing samples and plasma from patients
with dysfibrinogenaemia.

TOTAL CLOTTABLE PROTEIN DETERMINATION

Principle
Gaffney and WongS2 reported a method for the determination of total clottable
protein, which is used to determine the fibrinogen level in the international
standard. Alternative methods for total clottable fibrinogen have been described
by Blombiick and Blombiick3s , Ratnoff and Menzie36, and Swain and Feders37 •
In the method described by Gaffney and WongS2 the fibrinogen is separated
from other plasma proteins by clotting diluted plasma, eliminating the soluble
proteins by gently synerizing the serum with subsequent washing of the clot.
A long incubation time after addition of thrombin allows a quantitative fibrin
formation. Because of the long incubation time the clot includes, besides fibrin,
the larger degradation products of fibrinogen, which may be present in the
plasma. Some traces of other proteins may also remain trapped in the clot.
Therefore excessive washing is obligatory. However, in abnormal plasmas
(especially when the fibrinolytic system is active) the values found with clottable
protein assays may be higher (additional material included in the clot) than
those found with the Clauss assay (inhibiting effect of FDP).
The fibrinogen is converted to fibrin with thrombin in the absence of Ca2+
ions. Therefore, the clot is not stabilized and remains soluble in concentrated
urea. The fibrinogen concentration can be measured by measuring the light
absorbance at 280 nm and 315 nm in the well-washed clot.
83
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Materials
Phosphate buffer
1. Stock: 2.77 g KH2P0 4, 0.822 g Na2HP04.2H20 and H 20 added up to a
volume of 1000 ml, pH 6.35.
2. Working buffer: 1 volume of stock buffer + 2 volumes of physiological saline
(0.15 moVL NaCI).

Thrombin solution
500 Iu/ml solution of bovine or human thrombin in physiological saline (1 IV
- 1 NIH unit).

Clot solvent
400 g urea were solubilized in distilled water. Following the addition of 200 m1
of 1.0 moVL NaOH the solution was made up to 1000 ml with distilled water.

Procedure
1 ml citrated plasma
2 ml working buffer
50 J.1l thrombin solution
are rapidly and thoroughly mixed and left at room temperature for 2 h. The
tube is placed upside-down on a lint-free cloth placed on absorbent material,
such as filter paper. The clot is allowed to synerize spontaneously and then
blotted to dryness using appropriate absorbent material. The clot is then
harvested and washed for 30 min in 50 ml 0.15 moVL NaCI solution. This
washing procedure is repeated twice, blotting dry between each washing. Do
not lose small pieces of the clot. Washing is the most difficult step of the assay.
Mter washing, transfer the clot to a tube containing 7.5 ml of the clot solvent
until complete solubility is achieved. The resultant solution is mixed and its
optical density (00) is measured at 280 nm and 315 nm in a quartz cuvette, 1
em lightpath, using clot solvent as a blank.

Calculation of results
The fibrinogen concentration (fbg) is calculated according to the following
formula:
(00 of 280 nm - 00 of 315 nm) x 7.5/15.87 =fbg (gIL)
where 7.5 is the dilution factor and 15.87 is the extinction coefficient (E1'i:8) for
fibrin at alkaline pH. The relationship between 00 and fbg is linear up to an
84
 FIBRINOGEN

OD of 1. More concentrated solutions than this should be diluted further to

bring the OD below 1, and compensation should be made for the dilution in
the calculation of the fbg. The absorbance values are stable for 24 h.

Correlation between various assay modifications

The clotting rate assay34 is influenced by large amounts of intermediate and
late FDP. In DIC and under thrombolytic therapy it may underestimate the
'true' level of clottable fibrinogen3,42,4~,55,56,60,6(, but its value is a clinically
appropriate indicator of the clotting defect.
With increasing automation of the clinical coagulation laboratory various
adaptations of functional assays have been pro~osed. Fibrin polymerization
is induced by thrombin41 -43,56 or by batroxobin 9,40 and the clotting time can
be recorded by various optical methods. With normal plasma the agreement
between manually performed clotting rate assays34 and the automated tests is
acceptable. Discordant results were reported for plasma samples of patients
with DIC, thrombolytic therapy, heparin treatment or hyperlipaemia41 -43.
Such systematic differences may depend on the way the analyser recognizes
fibrin polymerization or on opacity of the thromboplastin. The photometric
assessment of increasing turbidity (change of absorbance per minute at 334
nm) after addition of batroxobin has successfully been used in a centrifugal
analyser40 . The test system is somewhat less affected by elevated FDP than the
manual clotting rate assay, but influenced by high plasma turbidity39,40. Photo-
optical clot detection and registration of the clotting time by another analyser
apparently reached excellent agreement with the Clauss assay on the KCIO
coagulometer41 . Discrepancies have been observed with analysers monitoring
thrombin-induced fibrin polymerization by light_scattering42-44,56. Spuriously
high fibrinogen values under thrombolytic therapy or in DIC may be misleading
for the clinician.
Clotting rate assays of fibrinogen performed on automatic analysers should
therefore be systematically compared with manual techniques on both normal
and pathological samples.

References
1. Doolittle RF. Fibrinogen and fibrin. In: Bloom AL, Thomas DP, eds. Haernostasis and
thrombosis. Edinburgh: Churchill Livingstone, 1987; 192-215.
2. Hantgan RR, Francis DW, Scheraga HA, Marder VI Fibrinogen structure and physiology.
In: Colman RW, Hirsh J, Marder VJ, Salzman EW, eds. Hemostasis and thrombosis, 2nd edn.
Philadelphia: Lippincott, 1987; 269-88.
3. Francis CWO Marder VI Physiologic regulation and pathologic disorders of fibrinolysis. In:
Colman RW, Hirsh J, Marder VJ, Salzman EW, eds. Hemostasis and thrombosis, 2nd edn.
Philadelphia: Lippincott, 1987; 358-79.
4. Mosesson MW, Finlayson JS, Umfleet RA, Galanakis D. Human fibrinogen heterogeneities.
I. Structural and related studies of plasma fibrinogens which are high solubility catabolic
intermediates. J Bioi Chern 1972; 247: 5210-19.
5. Lipinska I, Lipinski B, Gurewich V. Fibrinogen heterogeneity in human plasma. Electrophoretic
demonstration and characterization of two major fibrinogen components. J Lab Clin Med
1974; 84: 509-16.

85
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
6. Holm B, Godal He. Quantitation of three normally occurring plasma fibrinogens in health
and during so-called 'acute phase' by SDS e1ectrophoresis of fibrin obtained from EDTA-plasma.
Thromb Res 1984; 35:279-90.
7. Holm B, Nilsen DWT, Kieru1f P, Godal He. Purification and characterization of 3 fibrinogens
with different molecular weights obtained from normal human plasma. Thromb Res 1985;
37: 165-76.
8. Nakashima A, Sasaki S, Miyazaki K, Miyata T, Iwanaga S. Human fibrinogen heterogeneity:
the COOH-terminal residues of defective Aa chains of fibrinogen II. Blood Coag Fibrinol
1992; 3: 361-70.
9. Miiller E, Henschen A. Isolation and characterization of human plasma fibrinogen molecular
size-variants by high-performance liquid chromatography and amino acid sequence analysis.
In: Mosesson MW, Amrani DL, Siebenlist KR, DiOrio JP, eds. Fibrinogen 3. Biochemistry,
biological functions, gene regulation and expression. Amsterdam: Elsevier, 1988; 279-82.
10. Furlan M, Felix R, Escher N, Liimm1e B. How high is the true fibrinogen content of fibrinogen
standards? Thromb Res 1989; 56: 583-92.
II. Holm B, Brosstad F, Kierulf P, Godal HC. Polymerization properties of two normally
circulating fibrinogens, HMW and LMW. Evidence that the COOH-terminal end of the
a-chain is of importance for fibrin polymerization. Thromb Res 1985; 39: 595-606.
12. Bando M, Matsushima A, Hirano J, Imadu Y. Thrombin catalyzed conversion of fibrinogen
to fibrin. J Biochem 1972; 71: 897-9.
13. Martinelli RA, Scheraga HA. Steady-state kinetic study of the bovine thrombin-fibrinogen
interactions. Biochemistry 1980; 19: 2343-50.
14. Higgins DL, Lewis SD, Shafer JA. Steady-state kinetic parameters for the thrombin catalyzed
conversion of human fibrinogen to fibrin. J Bioi Chern 1983; 258: 927fr82.
IS. Lewis SD, Shafer JA. A thrombin assay based upon the release of fibrinopeptide A from
fibrinogen: definition of a new thrombin assay based upon the release of fibrinopeptide A
from fibrinogen; definition of a new thrombin unit. Thromb Res 1984; 35: 111-20.
16. Ernst E, Resch K.L. Fibrinogen as a cardiovascular risk factor: a meta-analysis and review
of the literature. Ann Intern Med 1993; 118: 956--63.
17. Lowe GD. Fibrinogen and cardiovascular disease: historical introduction. Eur Heart J 1995;16
(Supp.l A): 2-5.
18. Thompson SG, Kienast J, Pyke SD, Haverkate F, van de Loo Je. Hemostatic factors and the
risk of myocardial infarction or sudden death in patients with angina pectoris. European
Concerted Action on Thrombosis and Disabilities Angina Pectoris Study Group. N Eng! J
Med 1995; 332: 635-41.
19. Heinrich J, Balleisen L, Schulte H, Assmann G, van de Loo J. Fibrinogen and factor VII in
the prediction of coronary risk: results from the PROCAM study in healthy men. Arterioscler
Thromb 1994; 14: 54-9.
20. Balleisen L, Bailey J, Epping PH, Schulte H, van de Loo J. Epidemiological study on factor
VII, factor VIII and fibrinogen in an industrial population: I. Baseline data on the relation
to age, gender, body weight, smoking, alcohol, pill-using, and menopause. Thromb Haemostas
1985; 54: 475--9.
21. Krobot K, Hense Hw, Cremer P, Eberle E, Keil U. Determinants of plasma fibrinogen: relation
to body weight, waist-to-hip ratio, smoking, alcohol, age, and sex. Results from the second
MONICA Augsburg survey 1989-1990. Arterioscler Thromb 1992; 12: 780-8.
22. Folsom AR, Wu KK, Davis CE, Conlan MG, Sorlie PD, Szklo M, for the Atherosclerosis
Risk in Communities (ARIC) Study Investigators. Population correlates of plasma fibrinogen
and factor VII, putative cardiovascular risk factors. Atherosclerosis 1991; 91: 191-205.
23. Mattiasson I, Lindgarde F. The effect of psychosocial stress and risk factors for ischaemic
heart disease on the plasma fibrinogen concentration. J Intern Med 1993; 234: 45--51.
24. Humphries SE, Cook M, Dubowitz M, Stirling Y, Meade Tw. Role of genetic variation at
the fibrinogen locus in determination of plasma fibrinogen concentrations. Lancet 1987; I:
1452-5.
25. Hamsten A, Iseluis L, de Faire U, Blombiick M. Genetic and cultural inheritance of plasma
fibrinogen concentration. Lancet 1987; 2: 988-92.
26. Green F, Hamsten A, Blombiick M, Humphries S. The role of beta-fibrinogen genotype in
determining plasma fibrinogen levels in young survivors of myocardial infarction and healthy
controls from Sweden. Thromb Haemostas 1993; 70: 915-20.

86
 FIBRINOGEN

27. Ferrer Antunes CA, de Maat MPM, Palmeiro A, Pimentel J, Fernandes V. Association between
polymorphisms in the fibrinogen a- and f}-genes on the post-trauma fibrinogen increase.
Throm. Res. 1998: in press.
28. Montgomery HE, Clarkson P, Nwose OM, Mikailidis DP, Jagroop lA, Dollery G, Moult J,
Benhizia F, Deanfield J, Jubb M, World M, McEwan JR, Winder A, Humphries S. The acute
rise in plasma fibrinogen concentration with exercise is influenced by the G-(453)-A
polymorphism of the beta-fibrinogen gene. Arterioscler Thromb Vasc Bioi 1996; 16: 386--91.
29. de Maat MPM, Arnold AER, van Buuren S, Wilson JHP, KIuft C. Modulation of plasma
fibrinogen levels by ticlopidine in healthy volunteers and patients with stable angina pectoris.
Thromb Haemostas 1996; 76: 166--70.
30. de Maat MPM, Knipscheer HC, Kastelein JJP, KIuft C. Modulation of plasma fibrinogen
levels by ciprofibrate and gemfibrozil in primary hyperiipidaemia. Thromb Haemostas 1997;
77: 75-9.
31. Mazoyer E, Ripoll L, Boisseau MR, Drouet L. How does ticlopidine treatment lower plasma
fibrinogen? Thromb Res 1994; 75: 361-70.
32. Branchi A, Rovellini A, Sommariva D, Gugliandolo AG, Fasoli A. Effect of three fibrate
derivatives and of two HMG.coA reductase inhibitors on plasma fibrinogen level in patients
with primary hypercholesterolemia. Thromb Haemostas 1993; 70: 241-3.
33. De Maat MPM, Haverkate F. Critical evaluation of fibrinogen assays. In: Seghatchian MJ,
Samama MM, Hecker SP, eds. Hypercoagulable states: fundamental aspects, acquired disorders,
and congenital thrombophilia. Boca Raton: CRC Press, 1996; 105-16.
34. Von Clauss A. Gerinnungsphysiologische Schnellmethode zur Bestimmung des Fibrinogens.
Acta Haematol1957; 17: 237-46.
35. Blombiick B, Blomback M. Preparation of human fibrinogen fraction 1-2. Arkiv Kemi 1956;
10: 415-43.
36. Ratnoff OD, Menzie C. A new method for the determination of fibrinogen in small samples
of plasma. J Lab Clin Med 1951; 37: 316--20.
37. Swain W, Feders MB. Fibrinogen assay. Clin Chern 1967; 13: 1026--8.
38. Vermylen C, de Vreker RA, Verstraete M. A rapid enzymatic method for assay of fibrinogen
fibrin polymerization time (FPT test). Clin Chim Acta 1963; 8: 418-24.
39. Becker U, Bartl K, Wahlefeld AW. A functional photometric assay for plasma fibrinogen.
Thromb Res 1984; 35: 475-84.
40. De Metz M, van Wersch JWI Use of a centrifugal analyzer for a chromogenic prothrombin
time, a chromogenic activated partial thromboplastin time and a kinetic fibrinogen assay in
a routine hospital laboratory. Haemostasis 1987; 17: 254-9.
41. Cambas JP, Bierme R. Martinon JC, Dousset B. Evaluation des performances d'un automate
en coagulation: l'Electra 700 Nouv Rev Fr Hemoto11985; 25: 313-20.
42. Bick RZ, Wheeler A, Camosano NA. Comparative study of the DuPont antithrombin III
and fibrinogen assay systems. Am J Clin Patho11985; 83: 541-6.
43. Hoffmann JJML, Verhappen MAL. Automated nephelometry of fibrinogen: analytical
performance and observations during thrombolytic therapy. Clin Chem 1988: 34: 2135-40.
44. Exner T, Burridge J, Power P, Richard KA. An evaluation of currently available methods for
plasma fibrinogen. Am J Clin Patho11979; 71: 521-7.
45. Schulz FH. Eine einfache Bewertungsmethode von Leberparenchymschiiden (volumetrische
Fibrinbestimmung). Acta Hepatol1955; 3: 306--10.
46. Hoffman M, Greenberg CS. The effects of fibrin polymerization inhibitors on quantitative
measurements of plasma fibrinogen. Am J Clin Patho11987; 88: 490-3.
47. Hoegee-de Nobel E, Voskuilen M, Briet E, Brommer EJP, Nieuwenhuizen W. A monoclonal
antibody-based quantitative enzyme immunoassay for the determination of plasma fibrinogen
concentrations. Thromb Haemostas 1988; 60: 415-18.
48. Sweetham PM, Thomas HF, Yarnell JWG, Beswich A, Baker lA, Elwood PC. Fibrinogen,
viscosity and the 10-year incidence of ischaemic heart disease. The Caerphilly and Speedwell
studies. Eur Heart J 1996; 17: 1814-20.
49. Cremer P, Nagel D, Labrot B, Mann H, Muche R, Elster H, Seidel D. Lipoprotein Lp(a) as
predictor of myocardial infarction in comparison to fibrinogen, LDL cholesterol and other
risk factors: results from the prospective Gottingen Risk Incidence and Prevalence Study
(GRIPS). Eur J Clin Invest 1994; 24: 444-53.
50. Sweetnam PM, Yarnell JWG, Lowe GDO, Baker lA, O'Brien JR, Rumley A, Etherington

87
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
MD, Whitehead PJ, Elwood PC. The relative predictive power of heat-precipitation
nephelometric and clottable (Clauss) fibrinogen in the prediction of ischaemic heart disease:
the Caerphilly and Speedwell studies. Br J Haematol1998; 100:582-8.
51. Jacobson K. Studies in the determination of fibrinogen in human blood plasma. Scand J Clin
Lab Invest 1955 (Suppl. 14): 1-54.
52. Gaffney PJ, Wong MY. Collaborative study of a proposed international standard for plasma
fibrinogen measurement. Thromb Haemostas 1992; 68: 428-32.
53. Thompson SG, Duckert F, Haverkate F, Thomson 1M. The measurement of haemostatic
factors in 15 European laboratories: quality assessment for the multicentre ECAT Angina
pectoris study. Thromb Haemostas 1989; 61: 301--6.
54. Siefring GE, Riabov DK, Wehrly JA. Development and analytical performance of a functional
assay for fibrinogen on the Du Pont aca™ analyzer. Clin Chern 1983; 29: 614-17.
55. McDonagh J, Carrell N. Disorders of fibrinogen structure and function. In: Colman RW,
Hirsch J, Marder VJ, Salzman EW, eds. Hemostasis and thrombosis, 2nd edn. Philadelphia:
Lippincott, 1987; 301-17.
56. Jespersen J, Sidelmann JA. A study of the conditions and accuracy of the thrombin time assay
of plasma fibrinogen. Acta Haematol1982; 67: 2-7.
57. Palareti G, Maccaferri M, Manotati C, Tripodi A, Chantaranghul V, Rodeghiero F,
Ruggeri M, Mannucci PM. Fibrinogen assays: a collaborative study of six different methods.
Clin Chem 1991; 37: 714-19.
58. Chitolie A, Mackie 11, Grant D, Hamilton JL, Machin SM. Inaccuracy of the 'derived'
fibrinogen measurement. Blood Coag Fibrin 1994; 5: 955-7.
59. Rossi E, Mondonico P, Lombardi A, Preda L. Method for the determination of functional
(clottable) fibrinogen by the new family of ACL coagulometers. Thromb Res 1988; 52: 453--68.
60. Marder VJ, Shulman NR. High molecular weight derivatives of human fibrinogen produced
by plasmin II. Mechanism of their anticoagulant activity. J Bioi Chern 1969; 244: 2120-4.
61. Saleem A, Fretz K. An improved method for plasma fibrinogen based on thrombin time
measurement. Clin Chim Acta 1975; 62: 131--6.

88
9
Activated factor VII
J. H. MORRISSEY

INTRODUCTION

Factor VIla, a vitamin K -dependent plasma serine protease, is the first enzyme
in the coagulation cascade. The majority of this glycoprotein circulates as the
inert zymogen (factor VII) with a concentration in pooled normal plasma of
about 470 nglml (9.4 nmollL)l. However, low levels of activated factor VII
(factor VIla) are also present in the circulation. Typically, factor VIla makes
up less than I % of the total factor VII in plasma2•
Factor VII zymogen consists of a single polypeptide chain (MW 50 000)
with the following domain structure (N-terminal to C-terminal): a y-carboxy-
glutamate-rich domain (Gla domain), two epidermal growth factor-like domains
(EGF domains), and a serine protease domain homologous to trypsin and
chymotrypsin3 • Factor VII has no measurable enzymatic activity until it is
converted to the active form (factor VIla) by hydrolysis of a single peptide
bond located between the second EGF domain and the protease domain. The
result is a two-chain protein of the same MW, whose heavy and light chains
are held together by a disulphide bond.
Currently, two methods exist for measuring the levels of factor VIla in
plasma: an activity-based method employing a modified form of tissue facto~,
and an ELISA-based method using an antibody that binds to the C-terminus
of the factor VIla light chain4 • These methods yield markedly different values
for the plasma content of factor VIla.

PHYSIOLOGICAL ROLE

Factor VIla is a relatively weak serine protease until it binds to its essential
cofactor, tissue factor (also known as tissue thromboplastin). Tissue factor has
no enzymatic activity itself Instead, it binds to factor VIla and enhances its
89
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
enzymatic activity in an allosteric manner. Thus, tissue factor can be considered
to be the regulatory subunit, and factor VIla the catalytic subunit, of the first
enzyme in the so-called extrinsic pathway of blood coagulation. The factor
VIla/tissue factor complex is considered to be the physiological initiator of
coagulation in normal haemostasis and many, if not all, thrombotic diseases5 .
Tissue factor is an integral membrane glycoprotein (MW 45 000) present on
the surface of a variety of different cell types located outside the vasculature.
Tissue factor is also an inducible protein in monocytes and vascular endothelial
cells5 .
In addition to its ability to bind to and enhance the activity of factor VIla,
tissue factor also binds zymogen factor VII with high affinity. Once bound to
tissue factor, factor VII is more easily activated to factor VIla. A variety of
serine proteases have been shown in vitro to catalyse the activation of factor VII,
including thrombin and factors IXa, Xa, and XIla. In addition, factor VIla can
catalyse the activation of factor VII in the presence of tissue factor ('factor
VII autoactivation,)6. It is presently unclear which proteases are the most
important for activation of factor VII to VIla during clotting in vivo.
Recently it has been demonstrated that plasma contains trace levels of factor
VIIa2, which are proposed to be important in priming the clotting system
following the exposure of tissue factor to the blood7,8. Since plasma has a high
content of protease inhibitors, the active forms of most of the coagulation
serine proteases have extremely short biological half-lives (~l min) and con-
sequently cannot be directly measured in plasma samples. Factor VIla, however,
is extremely resistant to plasma protease inhibitors, allowing it to circulate with
a relatively long half-life (about 2.5 h, compared with a biological half-life of
about 5 h for factor VII)9.

PATHOPHYSIOLOGICAL ASPECTS

Substantial clinical interest in factor VII was generated by the results of the
first N orthwick Park Heart Study, a large prosrective study of the incidence
of heart disease in a cohort of middle-aged men! . This study implicated plasma
factor VII coagulant activity as a significant risk factor for ischaemic heart
disease. A similar, although less strong, correlation was also reported by the
PROCAM study!!. Interestingly, follow-up studies of both populations indicate
that elevated factor VII coagulant activity appears to be most significant as a
predictor of fatal heart attacks although, again, this was more significant in
the Northwick Park Heart Study than in the PROCAM study!2,13.
Not every study has found a correlation between elevated factor VII coagulant
activity and risk of heart disease, leading to some controversy about the utility
of factor VII as a risk factor for thrombotic diseases. A major point of dis-
agreement between laboratories has centred on what is actually measured in
factor VII coagulant assays (Chapter 10). These assays employ tissue factor to
initiate coagulation, and tissue factor is well known to promote the conversion
of factor VII to VIla; thus, they measure an aggregate of both factor VII and
VIla. Furthermore, the sensitivity of the assay to factor VIla versus factor VII
90
 ACTIVATED FACTOR VII

appears to depend upon the assay configuration 7 ,14. For example, assays
employing bovine thromboplastin are more sensitive to the plasma content of
factor VIla than assays employing human or rabbit thromboplastin 15 . In
particular, it was proposed that the assay of factor VII coagulant activity used
by the Northwick Park Heart Study might be more sensitive to plasma levels
of factor VIla than assays used in other epidemiological studies. This was
demonstrated directly when the activity-based assay for plasma factor VIla
was developed 16 .
Consistent with earlier predictions7, trace levels of factor VIla have been
found in the plasma of all normal volunteers tested to date2• The amount of
factor VIla in pooled normal plasma is about 3 ng/ml. Using the activity-
based assay, plasma factor VIla levels have been shown to be elevated during
pregnancy and in diseases associated with thrombotic risk, such as diabetes2,17-19.
Several prospective studies are currently addressing the role of factor VIla as
a risk predictor of thrombotic disease. Plasma factor VIla levels have been
shown to be especially sensitive to warfarin treatment, including low-dose
warfarin which is difficult to monitor otherwise2,20. Interestingly, haemophilia
B patients were found to have substantially lower factor VIla levels than normal,
while haemophilia A patients had normallevels21 .
The recently reported ELISA for factor VIla yields values that are lOO-fold
lower than the activity-based assay4. Unlike the values obtained with the
activity-based assay, plasma factor VIla levels measured by the ELISA are
correlated with prothrombin fragment F I +2 levels (Chapter 23). Some possible
reasons for the differences in measured factor VIla levels using these two assays
are discussed below.

ACTIVITY ASSAY FOR FACTOR Vila

Principle
Recombinant tissue factor lacking the membrane anchor and cytoplasmic
domains (soluble tissue factor, or sTF) retains the ability to bind to factor VIla
and enhance its enzymatic activity. However, unlike wild-type tissue factor,
sTF has selectively lost the ability to promote the conversion of factor VII to
VIIa22 • Therefore, sTF shortens the clotting time of plasma in a manner that
is absolutely dependent upon its content of preformed factor VIla. This unique
property of sTF allowed the development of a clot-based assay for plasma
factor VIla that is free from interference from zymogen factor VIe. In order
to isolate the assay from variability in the content of other clotting factors in
the test plasma, the test plasma sample is typically diluted lO-fold and mixed
with an excess of factor VII-deficient plasma (which is correspondingly deficient
in factor VIla). The diluted plasma sample and factor VII-deficient plasma are
mixed with a reagent composed of sTF and phospholipid vesicles, and the time
to clot formation is measured following addition of a CaCl2 solution. A standard
curve is prepared in which varying amounts of purified factor VIla are added
to factor VII-deficient plasma. Factor VIla levels in the test plasma samples
are then determined by reference to the standard curve.
91
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Materials
HEPES-buffered saline (HBS)
50 mmoVL HEPES-NaOH buffer, pH 7.5 (at 25°C), containing 100 mmoVL
NaCl and 0.02% w/v NaN 3 •

HEPES-BSA buffer
50 mmoVL HEPES-NaOH buffer, pH 7.5 (at 25°C), containing 0.02% w/v
NaN 3 and 1% w/v bovine serum albumin, fatty acid-free (Calbiochem). Store
at 4°C.

Phospholipid vesicles
In our laboratory we prepare phospholipid vesicles using a 20:40:40 mixture
of the following purified phospholipids: bovine brain phosphatidylserine (PS),
bovine liver phosphatidylethanolamine (PE), and egg phosphatidylcholine (PC)
(available from Avanti Polar Lipids). The phospholipids are supplied as 10
mg/ml solutions in chloroform. Dispense 80 J.11 PS (0.8 mg), 160 J.11 PE (1.6 mg)
and 160 J.1l PC (1.6 mg) into an Eppendorf tube and dry down under a stream
of nitrogen or argon. When dry, place under high vacuum for an additional
hour to remove any remaining traces of solvent. Add I ml HBS and vortex
well to resuspend the phospholipids. Disperse the aggregated phospholipids
either by passing the suspension 15 times through a vesicle extruder containing
a filter with 100 nm pores (Liposofast vesicle extruder from Avestin), or by
sonication. The result is a 5 mmollL solution of phospholipid vesicles.
Alternatively, a commercial cephalin preparation that is free from activators
of the contact system may be used.

Soluble tissue factor reagent (sTF reagent)

Recombinant sTF has been produced in a number of research laboratories by
expression and purification from mammalian cells, yeast and bacteria23- 25 . It
is also available commercially (e.g. in the Staclot VIla-rTF assay kit from
Diagnostica Stago). The final concentrations in the sTF reagent are as follows:
26.8 J.1g/ml sTF (l J.1mollL), 500 J.1moVL phospholipid vesicles, 50 mmollL
HEPES-NaOH pH 7.5,100 mmoVL NaCl, 0.02% w/v NaN 3 , and 0.1% w/v
bovine serum albumin. Store the sTF reagent in aliquots at -80°C.

Calcium chloride solution

25 mmoVL CaCl2 (make fresh daily from a 1 moVL stock).

Factor VII-deficient plasma

This can be either immunodepleted or congenitally deficient, citrated plasma
(e.g., from George King Bio-Medical).
92
 ACTIVATED FACTOR VII

Calibrator

Recombinant or plasma-derived factor VIla can be used as the calibrant. Both

are available commercially (e.g. American Diagnostica). To make a concentrated
calibrator stock, add purified factor VIla to normal plasma to give a final
concentration of 300 ng/ml. Freeze in small aliquots at -80°C. When ready to
use, thaw quickly at 37°C, mix well, and hold at room temperature until needed.
To prepare a standard curve, dilute the calibrator with HEPES-BSA buffer to
obtain the following final concentrations of factor VIla: 3.0, 1.0, 0.3, and 0.1
ng/ml. (If necessary to extend the sensitivity of the assay, dilutions as low as 0.03
or 0.01 ng/ml factor VIla can be prepared, although this part of the standard
curve will deviate significantly from linearity. The limiting factor in the sensitivity
of the assay is the factor VIla content of the factor VII-deficient plasma.)

Quality control plasma(s)

Obtain citrated plasma samples from two donors known to have high and low levels
of plasma factor VIla. Freeze in small aliquots at -80°C. When ready to use, thaw
quickly at 37°C, mix well and hold at room temperature until needed. Immediately
before use, dilute the control plasmas lO-fold with HEPES-BSA buffer.

Samples

Use platelet-poor, citrated plasma. Avoid contact with unsiliconized glass, and do
not store at 4 °C or on ice. Samples can be stored frozen at -80°C. When ready
to use, thaw quickly at 37°C, mix well and hold at room temperature until needed.
Immediately before use, dilute the samples lO-fold with HEPES-BSA buffer.

Procedure

The clot-based activity assay can use any method for detection of clot formation
that is suitable for measuring prothrombin times, such as the manual tilt-tube
method or an automated coagulometer. In any case it is essential that the
clotting tests be performed using either plastic containers or glass containers
that have been siliconized. Plasma samples should never come into contact
with untreated glass surfaces.
Prior to performing the assay, thaw plasma samples and sTF reagent rapidly
at 37°C, mix well and hold at room temperature until used. Pre-warm the
calcium chloride solution and the cuvettes or test tubes in which clotting times
will be measured to 37°C. Add, in the following order, to the pre-warmed
cuvette or test tube:
50 III diluted plasma sample or diluted calibrator
50 III factor VII-deficient plasma
50 III sTF reagent.
93
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Wait 180 s (to ensure the materials reach 37°C), then add:
50 J.1l calcium chloride solution.
Mix well and measure the time to clot formation from the moment the calcium
chloride solution is added. Typically, the clotting times are determined in
duplicate.

Calibration
The clotting times for the diluted calibrators are measured in duplicate and a
standard curve is prepared by plotting the data on a double-logarithmic scale,
to which a line or second-order polynomial is fitted by the least-squares method
(occasionally a third-order polynomial must be fitted, especially if the standard
curve is extended to include very low concentrations of factor VIla). This
method allows the measurement of plasma factor VIla levels in ng/ml with
reference to the purified factor VIla used in preparing the calibrators. An
example of a typical standard curve obtained in this way is given in Figure
9.lA. However, as the potency of purified factor VIla may vary from laboratory
to laboratory, it has been proposed that the level of factor VIla in the calibrators
itself be calibrated a~ainst the First International Standard Factor VIla
concentrate (89/688)2 . Consistent with this suggestion, the commercially
available kit (Staclot VIla-rTF; Diagnostica Stago) is calibrated against this
standard and yields factor VIla values in mVlml. A typical calibration curve
obtained with the Staclot VIla-rTF assay kit is given in Figure 9.lB.

100 r - - - - - - - - - - ,

3 10

Factor Vila, ngJmL Factor Vila, mU/mL

Figure 9.1 Typical calibration curves for the activity assay for plasma factor VIla (clotting time
versus factor VIla concentration). Panel A: calibration curve using the method described in this
chapter, expressed in nglmL purified factor VIla. If necessary (Le. for measuring very low levels
of factor VIla), the standard curve can readily be extended to 0.03 nglmL. Panel B: calibration
curve using the commercially available kit (Staclot VIla-rTF; Diagnostica Stago), expressed in
mU/ml factor VIla (relative to the First International Standard Factor VIla concentrate). Data
for both curves were collected on a model ST4 coagulometer (Diagnostica Stago). In both cases
the indicated factor VIla concentrations are those obtained after dilution with HBS-BSA buffer.
Plasma factor VIla levels determined for test samples must therefore be multiplied by the dilution
factor of the sample (typically, lO-fold).

94
 ACTIVATED FACTOR VII
Reference values
Using the sTF-based clotting assay, the levels of factor VIla in the plasma of
188 normal volunteers was 0.5-8.4 ng/ml, with a mean value of 3.6 ± 1.4 ng/ml.

Pitfalls
1. Potency of the factor VIla standard. As mentioned above, the potency of
different preparations of purified factor VIla can vary, making it very
difficult to compare the absolute levels of plasma factor VIla measured in
different laboratories. Hubbard and Barrowcliffe26 recommended that the
assay be calibrated against the First International Standard Factor VIla
concentrate (89/688), allowing the results to be presented in mU/ml factor
VIla rather than ng/ml. This is an excellent suggestion, and the process need
only be performed once per batch of factor VIla calibrators. An alternative
is to use (or calibrate against) the commercially available activity assay for
factor VIla, which is itself already calibrated against this standard.
2. Cold activation of plasma samples. Artifactual generation of factor VIla in
citrated plasma samples can occur if liquid plasma samples are held at
reduced temperature (such as 4 0c) in the presence of untreated glass or
any other activator of the contact system. Cold activation can be avoided
by collecting, processing and storing plasma samples in containers made of
plastic or siliconized glass. Furthermore, blood should be handled and
processed at room temperature, and platelet-poor plasma should be frozen
at -80°C. When needed, the plasma samples should be thawed rapidly at
37°C, mixed thoroughly, and held at room temperature until assayed.
Samples can be refrozen and rethawed, although the levels of factor VIla
will be diminished somewhat with each freeze-thaw cycle.
3. Quality of factor VII-deficient plasma. Factor VII-deficient plasma must
have very low levels of factor VIla, otherwise the standard curve will plateau
at lower levels of added factor VIla. In this case, factor VII-deficient plasma
with lower endogenous levels of factor VIla must be obtained.

ALTERNATIVE METHODS FOR THE MEASUREMENT OF PLASMA

FACTOR Vila
Fluorogenlc activity assay
The sTF-based activity assay has been adapted so that thrombin generation
(measured with a fluorogenic thrombin substrate) is measured as the end-point
instead of clot formation 17. The basic principle of the assay is the same, and
the results are comparable to the clot-based assay.

Factor Vila-specific ELISA

Typically, antibodies raised against factor VII or VIla recognize both the
zymogen and activated forms equally well. Recently, however, a factor
95
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

VIla-specific ELISA has been reported, the basis for which is a c~ture antibody
that binds to an activation-specific neoantigen on factor VIla . This capture
antibody is a polyclonal rabbit antibody raised against a synthetic peptide that
comprises the last 12 amino acids at the C-terminus of the factor VIla light
chain. This peptide region is highly likely to differ in structure in factor VII
versus factor VIla, because in factor VII the peptide sequence is embedded in
a continuous polypeptide chain, whereas in factor VIla it has a free C-terminus
and is likely to be highly solvent-exposed.
In practice, ELISA plates are coated with the capture antibody and, after
suitable blocking and washing steps, the wells are incubated with plasma
samples to allow capture of factor VIla. After washing, the bound factor VIla
is detected with a biotinylated second antibody against factor VIINIla (either
a monoclonal or polyclonal antibody), which is subsequently detected using
streptavidin-alkaline phosphatase and a suitable alkaline phosphate substrate.
The assay is calibrated using purified factor VIla.
The activity-based assay and the ELISA give excellent agreement when
measuring purified factor VIla added to plasma4 • However, results from the
two assays differ dramatically when used to measure the endogenous levels of
factor VIla in plasma. Thus, according to the factor VIla-specific ELISA,
normal plasma contains 0.025 ± 0.01 nglml factor VIla, which is some 100-fold
lower than that measured by the activity assays.
The reason for the discrepancy between the factor VIla-specific ELISA and
the activity-based assays for plasma factor VIla is currently not known. One
potential explanation is that the C-terminus of the factor VIla light chain may
undergo proteolysis during its relatively long lifetime in the circulation. If so,
this would render factor VIla unable to bind to the anti-peptide antibody while
potentially having no effect on its activity. Indeed, plasma is known to contain
high levels of arginine-specific carboxypeptidase activity27, which could remove
the last amino acid of the factor VIla light chain. If the C-terminus of the
factor VIla light chain does undergo proteolysis during circulation, this may
explain why the two assays agree when measuring purified factor VIla that is
added to plasma, but do not agree when measuring the endogenous factor VIla
in plasma. Clearly, further research is needed to clarify these questions.

References
1. Fair DS. Quantitation of factor vn in the plasma of normal and warfarin-treated individuals
by radioimmunoassay. Blood 1983; 62: 784--91.
2. Morrissey JH, Macik BG, Neuenschwander PF, Comp PC. Quantitation of activated factor
VII levels in plasma using a tissue factor mutant selectively deficient in promoting factor VII
activation. Blood 1993; 81: 734-44.
3. Hagen FS, Gray CL, O'Hara P, Grant FJ, Saari GC, Woodbury RG, Hart CE, Insley M,
Kisiel W, Kurachi K et al. Characterization of a cDNA coding for human factor VII. Proc
Nat! Acad Sci USA 1986; 83: 2412-16.
4. Philippou H, Adami A, Amersey RA, Stubbs PJ, Lane DA. A novel specific immunoassay
for plasma two-chain factor VIla: Investigation of FVlIa levels in normal individuals and in
patients with acute coronary syndromes. Blood 1997; 89: 767-75.
5. Carson SD, Brozna JP. The role of tissue factor in the production of thrombin. Blood Coag
Fibrinol1993; 4: 281-92.

96
 ACTIVATED FACTOR VII
6. Nakagaki T, Foster DC, Berkner KL, Kisiel W. Initiation of the extrinsic pathway of blood
coagulation: Evidence for the tissue factor dependent autoactivation of human coagulation
factor VII. Biochemistry 1991; 30: 10819-24.
7. Hultin MB, Jesty J. Factor VIla levels in patients with hemophilia. Blood 1992; 80: 3255-6.
8. Rapaport SI. Regulation of the tissue factor pathway. Ann NY Acad Sci 1991; 614: 51-62.
9. Seligsohn U, Kasper CK, Osterud B, Rapaport SI. Activated factor VII: presence in factor
IX concentrate and persistence in the circulation after infusion. Blood 1978; 53: 828-37.
10. Meade TW, Mellows S, Brozovic M, Miller GJ, Chakrabarti RR, North WRS, Haines AP,
Stirling y, Imeson JD, Thompson SG. Haemostatic function and ischaemic heart disease:
principal results of the Northwick Park heart study. Lancet 1986; 2: 533-7.
11. Balleisen L, Schulte H, Assmann G, Epping P-H, Van de Loo J. Coagulation factors and the
progress of coronary heart disease. Lancet 1987; 2: 46l.
12. Ruddock V, Meade Tw. Factor VII activity and ischaemic heart disease: fatal and non-fatal
events. Q J Med 1994; 87: 403-6.
13. Heinrich J, Balleisen L, Schulte H, Assmann G, Van de Loo J. Fibrinogen and factor VII in
the prediction of coronary risk: results from the PROCAM study in healthy men. Arterioscler
Thromb 1994; 14: 54-9.
14. Mann KG. Factor VII assays, plasma triglyceride levels, and cardiovascular disease risk.
Arteriosclerosis 1989; 9: 783-4.
15. Hemker HC, Muller AD, Gonggrup R. The estimation of activated human blood coagulation
factor VII. J Mol Med 1976; 1: 127-34.
16. Miller GJ, Stirling Y, Esnouf MP, Heinrich J, Van de Loo J, Kienast J, Wu KK, Morrissey
JH, Meade Tw, Martin JC et al. Factor VII-deficient substrate plasmas depleted of protein
C raise the sensitivity of the factor VII bio-assay to activated factor VII: an international
study. Thromb Haemostas 1994; 71: 38-48.
17. Kario K, Miyata T, Sakata T, Matsuo T, Kato H. Fluorogenic assay of activated factor VII:
Plasma factor VIla levels in relation to arterial cardiovascular diseases in Japanese. Arterioscler
Thromb 1994; 14: 265-74.
18. Kario K, Matsuo T, Matsuo M, Kolde M, Yamada T, Nakamura S, Sakata T, Kato H, Miyata
T. Marked increase of activated factor VII in uremic patients. Thromb Haemostas 1995; 73:
763-7.
19. Kario K, Matsuo T, Kobayashi H, Matsuo M, Sakata T, Miyata T. Activation of tissue factor-
induced coagulation and endothelial cell dysfunction in non-insulin-dependent diabetic patients
with microalbuminuria. Arterioscler Thromb Vase Bioi 1995; 15: 1114-20.
20. Raskob GE, Durica SS, Morrissey JH, Owen WL, Comp PC. Effect of treatment with low-dose
warfarin-aspirin on activated factor VII. Blood 1995; 85: 3034-9.
2l. Wildgoose P, Nemerson Y, Hansen LL, Nielsen FE, Glazer S, Hedner U. Measurement of
basal levels of factor VIla in hemophilia A and B patients. Blood 1992; 80: 25-8.
22. Neuenschwander PF, Morrissey JR. Deletion of the membrane anchoring region of tissue
factor abolishes autoactivation of factor VII but not cofactor function. Analysis of a mutant
with a selective deficiency in activity. J BioI Chem 1992; 267: 14477-82.
23. Rezaie AR, Fiore MM, Neuenschwander PF, Esmon CT, Morrissey JH. Expression and
purification of a soluble tissue factor fusion protein with an epitope for an unusual calcium-
dependent antibody. Protein Express Purif 1992; 3: 453-60.
24. Shigematsu Y, Miyata T, Higashi S, Miki T, Sadler JE, Iwanaga S. Expression of human
soluble tissue factor in yeast and enzymatic properties of its complex with factor VIla. J BioI
Chem 1992; 267: 21329-37.
25. Ruf W, Rehemtulla A, Morrissey JH, Edgington TS. Phospholipid-independent and -dependent
interactions required for tissue factor receptor and cofactor function. J BioI Chem 1991; 266:
2158-66.
26. Hubbard AR, Barrowcliffe Tw. Measurement of activated factor VII using soluble mutant
tissue factor - Proposal for standardization. Thromb Haemostas 1994; 72: 649-50.
27. Plummer T.H, Husmann M. Human plasma carboxypeptidase N: isolation and characterization.
J BioI Chern 1978; 253: 3907-12.

97
10
Factor VII activity and antigen
G. MARIANI, G. LIBERTI, T. D'ANGELO and L. LO COCO

INTRODUCTION

Factor VII (FVIn is an important element in the initiation of blood coagulation.

When tissue factor (TF) is exposed to the blood, upon vascular injury or
following monocyte activation, a one-to-one stoichiometric complex with FVII
is formed; this constitutes the primary event of blood coagulation and will
lead to the activation of FVII, through the cleavage of a single peptide bond
located at Arg152_Ile 153 • The TF-FVIIa serves to activate factors IX (FIX), X
(FX) and also FVII (autocatalytically). The TF-FVII pathway has the following
features:

1. FVIIa is the only known ligand of TF;

2. the presence of FVIIa in the blood seems to be the procofactor for the
production of the first thrombin molecules which, in turn, will activate FVIII
and FV, thus making the enzyme system more and more productive;
3. FVIIa itself displays very weak enzymatic activity and is cleared from the
blood stream with a half-life of 2-3 h I , which is equal to or somewhat shorter
than that of the zymogen;
4. very low levels of FVIIa are present in the blood under normal conditions;
5. the pathway is modulated by a specific inhibitor, the tissue factor pathway
inhibitor (TFPn, which forms a quaternary complex involving the TF-FVIIa
complex and factor Xa (Chapter 18).

Elevated FVII levels have been reported to be associated with an increased risk
of cardiovascular disease2, whilst FVII deficiencies are associated with a variable
bleeding tendency. As to the congenital deficiencies, there has been a considerable
degree of uncertainty on the relation between the clinical presentation of the
deficiency on one hand and FVII levels and the underlying genetic defect on
the other3 . Other elements may playa role in the interpretation of this
99
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
relationship, such as FVII plasma assays and other variations (gene poly-
morphisms) within the FVII gene.
An accurate laboratory evaluation of FVII levels thus appears the basic
requirement for the evaluation of any change in blood coagulation in which
FVII is involved, and the first assays to be carried out are FVII coagulant
activity (FVIIc) and FYII antigen (FVIIAg). FYIIAg expresses the total amount
of the protein in plasma; FVllc reflects the functional activity of the protein;
in fact, FVIIc is the result of the combined effects of FVIIa and FVII zymogen
in a one-stage clotting time. Unfortunately the results are dependent on the
type of thromboplastin selected4 .

FACTOR VII COAGULANT ACTIVITY ASSAY (FVllc)

FVllc activity is the basic assay in evaluating the level or function of the
extrinsic pathway of blood coagulation. FVllc has been developed to study
the congenital deficiencies of the factor and thromboplastin preparations from
different species were used to identify some deficiency variants 3,5. Bovine
thromboFlastin has also been used to evaluate the state of activation of the
molecule. Furthermore, FVIIc and/or FVIIAg assars are also currently used
in liver disease to evaluate the hepatocyte function 7 ••
FVllc is a sensitive assay allowing the detection of a few nanograms of the
protein. The reproducibility of the assay is also good, with very low intra- and
inter-assay coefficients of variation 9 •

Method outlines
The specific assay of clotting factors of the extrinsic pathway (factors II, V,
VII, X) in general consists of performing the prothrombin time (PT) (Chapter
6) of a mixture made up of a dilution of normal or patient's plasma and a
plasma specifically lacking the factor to be assayed (substrate/plasma). The
substrate plasma can be obtained from patients with a congenital, severe
deficiency « 1%) or from normal plasma artificially depleted by specific
monoclonal or polyclonal antibodies. The percentage of factor activity present
in plasma can be determined by the degree of correction of the clotting time
when the plasma to be tested is added to the substrate plasma. The clotting
time (expressed in seconds) obtained for this mixture is compared to the clotting
time obtained when dilutions of normal plasma (reference plasma) are added
to the same substrate plasma. The reference plasma contains, by definition,
100% FVII activity.

Reagents
FVII-deficient substrate plasma

Given the rarity of patients with a congenital deficiency in the extrinsic pathway
factors, substrate plasmas obtained by immunodepletion are currently used to
100
 FACTOR VII ACTIVITY AND ANTIGEN

diagnose congenital clotting factor deficiencies or to measure the levels of a

specific factor. The choice of an artificially depleted substrate plasma is dictated
by economic, safety and ethical decisions. Commercially available FVII-deficient
immunodepleted plasmas are particularly suitable because these plasmas mimic
the 'true' FYII deficiency as they are lacking FVII antigen and contain physi-
ologicallevels of the other factors/inhibitors. Commercially available substrate
plasmas are provided in the lyophilized state and must be stored at 2-8 °C for
the time indicated by the manufacturer.

Tissue factor reagent

The best thromboplastin reagent must have an International Sensitivity Index
(lSI) close to unity. Particularly suitable for this purpose are recombinant
thromboplastin preparations, derived from human TF, that are highly stand-
ardized and have an lSI of around 1. These preparations contain sufficient
calcium in 'standard' amounts to support coagulation reactions in a PT assay.

Buffer for the dilution of test or normal plasma

Veronal (Owren's) (2.84 x 10-2 moVL 5.5 sodium barbiturate in 0.125 moVL
sodium chloride, pH 7.35) or imidazole buffers (3.4 giL imidazole 5.85 gIL
NaCl, pH 7.4) are both suitable for this purpose.

Specimen collection, preparation and storage

Mix nine parts of freshly collected patient blood with one part of 0.11 or 0.13
mollL (3.2% or 3.8%) sodium citrate. Evacuated, siliconized or plastic tubes
containing the desired anticoagulant are commercially available and can be
used without difficulty. Citrated blood must be immediately centrifuged at
room temperature at 2500g for at least 10 min. Plasma is harvested with a
plastic pipette, placed in plastic (polycarbonate) test tubes and kept at room
temperature for no more than I h. Alternatively, it can be snap-frozen and
stored at -80°C, until assayed.

'Reference plasma'
A 'home-made' pooled normal plasma is preferable as a rule. This can be
prepared taking into account specific needs (only males or females, both,
specific age ranges, etc.), and can be stored for up to 6 months at -80 0c. By
definition it will contain 100% of the normal activity of 1 unitlml; alternatively
it can be calibrated against primary (WHO) and secondary standard (SSC)
(IU/ml).

Assay procedure
Before starting the assay, test plasma(s) and reference plasma must be thawed
at 37°C for no more than 5 min and vortexed. During the assay, plasmas must
101
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
be kept at room temperature. Reconstitution of the thromboplastin reagent is
performed according to the manufacturer's instructions. Pre-warm a suitable
amount of thromboplastin at 37°C.
1. Make serial dilutions of the reference plasma (l: 10 to 1:320) and of the test
plasma (1: 10 and 1:20) in dilution buffer. Keep these at room temperature
and assay within 30 min.
2. Add 0.1 m1 of each dilution to 0.1 m1 of substrate plasma (undiluted), vortex
the mixture and allow to incubate 1-2 min at 37°C.
3. Add 0.2 ml of the pre-warmed Ca-thromboplastin reagent (see reagent
section: tissue factor reagent) and start the stopwatch. When a clearly visible
clot forms, stop the watch. Carry out the assay duplicate (if duplicates agree
within 5% calculate the mean). Start the run from the last dilution and, for
replicates, from the first.

Results
Plot the results of the reference curve on two-cycle log-log graph paper with
the percentage of dilution on the abscissa and the time in seconds on the
ordinate (Figure 10.1). Connected points should yield a straight line. Devia-
tions from linearity may occur for the highest dilutions (1:160 or 1:320). If
deviation from linearity is relevant, repeat the assays. Assign 100% of FVII
activity to the 1: 10 dilution of the reference plasma. Read percentage factor
activity from the abscissa of the graph by finding the point where the clotting

100

"[

'.
r~ .
::!
3 50
CD
1ii' 40 0
1! c
30

20

1.0 10 100
% activity
Figure 10.1 FYII activity calibration curve. For example, if the clotting time of the plasma is
20 S, connect a point from the ordinate to the abscissa with a straight line and the factor VII will
correspond to 100% on the abscissa.

102
 FACTOR VII ACTIVITY AND ANTIGEN
time obtained for the 1: 10 patient's plasma dilution intercepts the reference
curve. Evaluate only those values which fall within the straight part of the
reference curve. If the patient's clotting times do not fit with the straight part
of the reference curve, plot clotting times from other dilutions and take into
account the dilution factor in the calculation. If FVII levels are very Iowa 1:5
dilution can be tested.

Assay procedure with automatic coagulometers

Since volumes of reagents may be different, and some instruments perform
the sample dilutions automatically, standardize your assay by carefully following
the instrument's manual.

Quality control
The use of control plasmas IS recommended. Control samples should be included
in each run and tested in the same manner as the test plasma samples. Each
laboratory should establish a 'normal reference range'. Compare the reference
curves obtained in different runs: if there are differences exceeding 10%, check
controls, reagents and instrument. New reference ranges should be established
for each new batch of substrate plasma, pooled normal plasma and instrument.
Under optimal conditions intra-assay coefficient of variation (CV) is 2-4%;
inter-assay CV should also be < 5%. Normal subjects are expected to have
FVlIc levels ranging from 55% to 170%, the amplitude of this range being
related to well-known genetic and environmental determinants9 •10 .

Specificity and sensitivity

Specificity depends on the substrate plasma and the dilution of the plasma
samples. Values below 2% have to be considered with caution. The suspicion
of undetectable FVII levels in a given specimen may be confirmed by a very
prolonged PT.

FACTOR VII ANTIGEN (FVIIAg) ASSAY

Factor VII antigen (FYIIAg) assay was originally set up to discriminate between
the 'true' deficiencies that are due to the lack of synthesis of the factor (FYIIAg-
negative) and those variants (FYIIAg-positive ) in which the gene defect allowed
the synthesis of normal or reduced levels of a dysfunctional factor 3•5•9 • Indeed,
the assay of FVIIAg is important in characterizing FYII congenital deficiencies,
but this assay, compared to the FVlIc assay may also give us a rough idea of
the activation state of the molecule (higher levels of FVlIc) and of the 'total
FVII mass' as an index of an increased factor synthesis.
103
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
One of the greatest limits of the available FVIIAg assays consists in its
sensitivity, for it is still difficult to discriminate between the absolute lack of
FVII protein and the presence of low levels of FVII protein « 4-5%).
We will here describe the most frequentlx used method for FVIIAg assay,
which was originally described by Boyer et al. I and subsequently commercialized
as an ELISA kit by Diagnostica Stago (Asserachrome® VII:Ag).

Method outlines
The F(abh fragment of an antibody to FVII immobilized on a plastic support
captures the FVII from a plasma sample. Subsequently, another antibody
coupled with peroxidase binds to the free epitopes of FYII, forming a 'sandwich'.
The presence of the peroxidase is then revealed by the addition of a substrate,
orthophenylenediamine (OPD), in the presence of hydrogen peroxide. The
intensity of the colour (optical density, 00) appearing in the mixture has a
direct relation with the FVII concentration initially present in the plasma
sample. Results obtained with the plasma samples to be tested are compared
to the 00 values obtained with dilutions of normal plasma (reference plasma)
that were tested under the same conditions. An antigen concentration of 100%
is assigned to the reference plasma.

Reagents
1. Plastic plates coated with an antibody to FVII: are ready to use in a sealed
envelope to be opened just before use.
2. Phosphate buffer: ready in a concentrated solution, to be diluted with distilled
water before use with a dilution factor 1:10.
3. Washing solution: ready in a concentrated solution, to be diluted with distilled
water before use 1:20.
4. Specific rabbit anti-FVII antiserum coupled with peroxidase: this is stored in
a lyophilized state with stabilizers; the complex is reconstituted with phosphate
buffer; in this state it is stable for up to 24 h at 2-8 cc.
5. Orthophenylenediamine substrate: present in tablets to be dissolved just before
use; after dissolution H 20 2 must be added. The colourless OPD/H 2 0 2
solution is stable for only I h at room temperature.
6. Acid solution: to stop the reaction between peroxidase and OPD/H 20 2 , 3
moVL sulphuric acid or I moVL HCI are used.
7. Specimen collection, preparation and storage: as for FVlIc assay, citrated
plasma has to be used; it can be assayed in fresh plasma or after storage at
-80 cC.
8. 'Reference plasma': a 'home made' pooled normal plasma is preferable as a
rule. It can be prepared taking into account the specific needs (males, females,
specific age ranges, etc.), and can be stored for up to 6 months at - 80 cC.
Alternatively it can be calibrated against WHO standard.
9. Needed equipment: multichannel pipettes, plate-washing equipment and
plate 00 reader set at 492 nm are needed.
104
 FACTOR VII ACTIVITY AND ANTIGEN

Assay procedure
Before starting the assay, test plasma(s) and reference plasma, if frozen, must
be thawed at 37°C and vortexed.
Patient's and reference plasma samples have to be pre-diluted 1:21 in
phosphate buffer. The diluted reference plasma is further diluted from 1:2 to
1:16 in the same buffer to prepare the reference curve.
1. FVII antigen immobilization: 200 III of patient's or reference sample are
pipetted into the precoated wells. The strips are covered and the test mixtures
left to incubate for 2 h at room temperature. The phosphate buffer is used
as a blank for the assay.
2. The wells are washed five times with washing solution.
3. Immobilization of the immunoconjugate takes place by adding 200 III of
the anti-FVII antibody linked to peroxidase. This step lasts 2 h.
4. The wells are washed five times with washing solution. This is followed
immediately by step 5.
5. The enzymatic reaction resulting in colour development. This is started by
the addition of the OPOIH2 0 2 substrate (200 Ill). This reaction must last
exactly 3 min.
6. The stopping acid is added (50 III of 3 mollL H 2 S04 or 100 III of 1 mollL
HCl) and the final well content is left to equilibrate for 10 min, then 00 is
read at 492 nm against the blank.

Results

Plot results of the reference curve on two-cycle log-log graph paper with the
percentage of dilution of reference plasma on the abscissa and the 00 at 492
nm on the ordinate. Connected points should yield a straight line; 100% of the
antigen is assigned to the 1:21 dilution of the reference plasma. Read percentage
factor antigen from the abscissa of the graph by finding the point where the
00 obtained for the I :21 patient's plasma dilution intercepts the reference
curve. Evaluate only those values which fall within the straight part of the
reference curve.

Quality control
The use of control plasmas is recommended. They should be run periodically
and tested in the same manner as the test plasma samples. Each laboratory
should establish a 'normal reference range' for FVIIAg. Compare the slopes
of the standard curves obtained in previous runs: if there are differences
exceeding 10%, check controls, reagents and instruments. New normal ranges
should be established for each new batch of the kit and pooled normal plasma.
Under optimal conditions the intra-assay CV is 5-7%; inter-assay CV should
be < 10%. Normal subjects are expected to have FVIIAg levels ranging from
50% to 150%, the amplitude of this range being related to well-known genetic
and environmental determinants9 ,1O.
105
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Specificity and sensitivity
The specificity of the assay is good as it is insensitive to the severe deficiencies
of the other vitamin K-dependent clotting factors. Values below 2-3% have to
be considered with caution.

Acknowledgements
This work was carried out within the framework of the European Union
Concerted Action BMHI-CT94-1202 'The role of the FVII-tissue factor
pathway in ischaemic heart disease (Clotart),.

References
I. Seligsohn U, Kasper CK, Osterud B, Rapaport SI. Activated factor VII: presence in factor
IX concentrates and persistence in the circulation after infusion. Blood 1978; 58: 828-37.
2. Meade TW, Mellows S, Brozovich M, Miller GJ, Chakrabarti RR, North RR, Haines AP,
Stirling y, Imeson JD, Thompson So. Hemostatic function and ischemic heart disease:
principal results of the Northwick Park Heart Study. Lancet 1986; 2: 533-7.
3. Mariani G, LoCoco L, Bernardi F, Pinotti M. Molecular and clinical aspects of factor VII
deficiency. Blood Coag Fibrinol1998; 9: S83-8.
4. Poggio M, Tripodi A, Mariani G, Mannucci PM. Factor VII clotting assay: influence of
different thromboplastins and factor VII-deficient plasmas. Thromb Haemostas 1991; 65:
160-4.
5. Mariani G, Ghirardini A, Iacopino G. Congenital deficiencies of the vitamin K dependent
clotting factors. In: Shearer MJ, Seghatchian MJ, eds. Vitamin K and vitamin K -dependent
proteins: analytical, physiological, and clinical aspects. Boca Raton: CRC Press; 1993; 163-93.
6. Kario K, Matsuo T, Asada R, Sakata T, Kato H, Miyata T. The strong positive correlation
between factor VII clotting activity using bovine thromboplastin and the activated factor VII
level. Thromb Haemostas 1995; 73: 429-34.
7. Pereira SP, Langley PG, Williams R. The management of abnormalities of hemostasis in
acute liver failure. Semin Liver Dis 1996; 16: 403-14.
8. Patrassi GM, Sartori MT, Viero M, Boeri G, Simioni P, Bassi N, Piccinni P, Girolami A.
Protein C, factor VII and prothrombin time as early markers of liver function recovery or
failure after liver transplantation. Blood Coag Fibrino11993; 4: 863-7.
9. Bernardi F, Marchetti G, Pinotti M, Arcieri P, Baroncini C, Papacchini M, Zepponi E, Ursicino
N, Chiarotti F, Mariani G. Factor VII gene polymorphisms contribute about one third of the
FYII level variation in plasma. Arterieroscl Thromb Vasc Bioi 1996; 16: 72-f1.
10. Mariani G, Bernardi F, Bertina R, Vincente Garcia V, Pridz H, Samama M, Sandset PM, Di
Nucci GD, Testa MG, Bendz B, Chiarotti F, Ciarla MY, Strom R. for the European Community
Concerted Action 'Clotart'. Serum phospholipids are prominent determinants of Factor VII
levels in the most common FVII genotype. Thromb Haemostas (In press).
II. Boyer C, Wolf M, Rothschild C, Migaud M, Amiral J, Mannucci PM, Meyer D, Larrieu MJ.
An enzyme immunoassay (ELISA) for the quantitation of human factor VII. Thromb
Haemostas 1986; 56: 250--5.

106
11
Factor VIII clotting activity
P. M. MANNUCCI and A. TRIPODI

INTRODUCTION

Factor VIII activity has been historically measured with two types of methods:
the one-stage assay, based on the activated partial thromboplastin time (APTf)
(Chapter 5) and the two-stage assay, based on the thromboplastin generation
test. The relative merits of the two assays have been debated 1•2 . In general,
results obtained on the same samples are similar; however, discrepancies
can be seen in plasmas which contain activated factors of the coagulation
cascade. These factors make the clotting times obtained with the one-stage
assay shorter than expected on the basis of the factor VIII content, resulting
in a significant overestimation of the factor VIII activity. For the same reason,
the one-stage assay is often unsuitable to determine factor VIII in concentrates.
Even though the two-stage assay does not seem to suffer from the same
drawback, it is difficult to perform because the reagents employed need a high
degree of standardization difficult to achieve in less specialized clinical
laboratories.
More recently, new two-stage methods which employ chromogenic substrate
have been proposed for the measurement of factor VIU 3•4 ; commercial kits are
available from various manufacturers (Chromogenix; Dade Behring; Diagnostica
Stago) and some of them have been evaluated and found suitable for clinical
application 5•6 • However, to our knowledge, their use is still limited and more
extensive clinical validation is probably needed. For this reason we have chosen
to discuss here the one-stage assay. More information on the other assays can
be found in the cited references. Before giving details of the method, we shall
provide a brief account of the general principles of bioassays on which factor
VIII and other clotting factor assays are based.

107
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
ASSAY CHARACTERISTICS AND PRINCIPLES

Most clotting factors can be assayed by estimating the degree of shortening

of the clotting time brought about by suitable dilutions of the test plasma
adding to a clotting system specifically deficient in the factor to be measured.
Clotting times will then be compared with those obtained when a dilution of
reference plasma is added to the same system, and the activity of the unknown
samples is expressed as potency relative to the reference plasma. Hence, the
essential components of the bioassay are the test system (i.e. the clotting system)
specifically deficient in the factor to be assayed and the reference plasma. The
following paragraphs outline the main components of the factor VIII clotting
bioassay.

Test system
The theory of the bioassay implies that when two samples at different factor
VIII activities (test and reference) are processed at different doses (dilutions)
with the specific assay system (APTT), this will give different responses (clotting
times). If those responses are plotted against the corresponding log-transformed
factor VIII activity one should obtain a pair of dose-response curves which
are parallel throughout their length. The horizontal distance between the two
curves is a measure of the potency ratio between the two samples (Figure 11.1).
It is important to emphasize that parallelism of the two curves is a prerequisite
for correct estimation of factor VIII activity. The occurrence of substantial
deviation from parallelism means that the two samples are not treated in
the same way by the system. In most cases deviation from parallelism is
likely to be due to the presence of an inhibitory substance in patient plasma
or to the deficiency in the substrate plasma of factor(s) other than that to be
measured.
This situation may occur, for instance, when substrate plasma used for the
assay has lost the activity of the labile factor V during prolonged and/or poor
storage conditions. To establish parallelism, patient and reference plasma must
be run at several dilutions. Hence, the simplified and widely adopted procedure
of testing a complete dose-response curve for reference plasma and only one
dilution of the test plasma must be regarded as incorrect. As mentioned above,
the system must be deficient in the factor to be measured. The most obvious
and cheap solution to this problem is the plasma from a patient with severe
haemophilia A (factor VIII clotting activity < 1 U/dl) with no antibody to
factor VIII. This solution is, however, less and less practicable because the
increasing risk of transmission of virus-related infections makes the majority
of severe haemophilic patients unsuitable for the purpose. A valid alternative
is to rely on factor VIII-deficient plasmas obtained by artificial depletion of
normal plasma by means of monoclonal antibody against factor VIle.
Commercial immunodepleted plasmas are also available (e.g. Ortho, Diagnostica
Stago).
108
 FACTOR VIII CLOniNG ACTIVITY

80

70
u
CD
<II

CD 60
E
:;:;

C>
c:
:;:; 50
....0
0
Reference
40

4 10 20 50 100
Log factor VIII activity (U/dl)

Figure 11.1 Graphical calculation of factor VllI clotting activity. The dilutions for the reference
and patient plasmas were 1:10,1:50 and 1:250. According to the graphical calculation (see text),
the factor VIII activity for the patient plasma is 57 U/dl of the reference; if the potency of the
reference calibrated against the international standard is 0.85 then the final result will be 57 x 0.85
= 48 IU/di. Should the patient plasma be tested at higher or lower dilution than the reference, the
results must be multiplied or divided by the appropriate dilution factor.

Reference plasma
A convenient source of reference plasma is plasma pooled from a suitable
number of healthy donors. It is important to note that plasma pooled from an
insufficient number of donors might be not representative of the average value
in the normal population. Hence, the assignment of an arbitrary potency of
100 Uldl may lead to important under- or overestimations of factor VIII levels
in the test sample.
Due to the large biological variability in a normal population (factor VIII
in normals varies from 50 to 150 U/dl), the adequacy of a pooled plasma in
reflecting the average normal content of factor VIII is largely dependent on
the number of healthy donors selected for the preparation. Fifty or more donors
seem to be an adequate number. However, collecting such a large number of
donors might be impracticable and not strictly necessary. At least in theory
plasma from a single donor is suitable as reference plasma, provided it is
calibrated against the international standard for factor VIII/von Willebrand
factor, which is available upon request from the National Institute for Biological
Standards and Control (NIBSC). The calibration is made by assaying the local
109
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
reference plasma against the international standard. To minimize error it is
important to carry out at least four independent assays which involve a complete
fresh set of the international standard and test dilutions each time. Since the
supply of standard is often limited, a suitable alternative might be to perform
two assays from the same vial of the reconstituted standard. The potency of
the local plasma is eventually obtained by multiplying the calculated potency
by that of the International Standard.

METHOD

Due to the large variation in sensitivity to factor VIII among commercial AP1T
reagents, it is impossible to give technical details which apply to all. As a general
rule each laboratory should establish the best conditions with the chosen re-
agent. For instance, it is advisable to make a complete dose-response curve by
testing several dilutions of reference plasma. The first dilution should be
sufficiently high to avoid interference with other plasma components; usually,
a 1:5 or 1: 10 dilution is suitable to start with and the other dilutions should be
spaced by the same factor (i.e. for a factor of five, 1:10, 1:50, 1:250, 1:1250,
etc.). It is advisable to use the same dilution interval for standard and patient
plasma alike. The last dilution to be tested depends on the reagent and substrate
plasma used; usually it will be the one whose clotting time is close to, but yet
shorter than, that of the 'blank' (Le. sample replaced by buffer). For choosing
the most suitable standard dilutions it is advisable to inspect the dose-response
curve and choose at least three dilutions which lie on the steeper and linear
part of the curve.
The dilutions of patient plasma should be chosen in accordance with the
expected factor VIn activity. In general the patient clotting times should be as
close as possible to, but yet longer than, those of the reference plasma. Therefore
if the standard is run at 1:10, 1:50, and 1:250, the patient plasma will be tested
either at 1:5,1:25 and 1:125 or 1:20, 1:100 and 1:500 when lower or higher-
than-normal factor VII1 activity is expected. To avoid the time-trend deterioration
of samples and reagents, it is advisable to take replicate readings of each
dilution in a balanced order (i.e. starting from the first dilution of the standard
to the last of the patient samples and the replicate in the opposite order).
The estimation of the potency of a patient sample relative to the standard
can be done by using a graphical or mathematical solution. The latter is
preferable because it avoids the bias in fitting the dose-response curves and
gives objective criteria to assess for linearity and parallelism. Simple calcula-
tions can be performed by pocket calculator as described by Ingram8 • Com-
prehensive statistical advice on writing computer programs for bioassay are
beyond the scope of this chapter, and can be found in the published literature9--11 .
For the graphical calculation the mean values of the replicate readings for each
dilution are plotted (vertical axis) against the log-transformed percentage
activity (horizontal axis), the first dilution of the reference and patient plasmas
being arbitrarily taken as 100 Uldl activity (Figure 11.1). After drawing the
best-fit lines through the data points and checking for parallelism, the activity
of factor VIII in the patient plasma is derived by drawing a horizontal line
110
 FACTOR VIII CLOTTING ACTIVITY

from the point where the patient curve intercepts the 100 U/dl activity to the
point where it crosses the reference curve. From that point a vertical line is
then dropped to intercept the horizontal axis. This value represents the
percentage activity of factor VIII in the patient plasma relative to the reference
(Figure 11.1). When patient plasma is tested at a higher or lower dilution than
the standard, the result must be multiplied or divided by the appropriate dilution
factor.
The procedure outlined below is intended to fit the characteristics of the
commercial APTT reagent and coagulometer currently used in our laboratory.
Other preparations can be successfully used provided the optimal conditions
are chosen as recommended above.

Reagents
1. Automated APTT (Organon Teknika). This reagent contains micronized
silica as activator and rabbit brain phospholipids as platelet substitute.
2. Imidazole buffer is prepared by dissolving 3.4 g of imidazole and 5.85 g of
NaCI in approximately 70 ml distilled water; 18.6 ml of 1 moVL HCI is then
added, the pH adjusted to 7.3-7.4 and the final volume adjusted to 100 ml.
This stock solution is stored at 4 DC and used to prepare the working solution
by diluting it 1:10 with distilled water.
3. CaCl2 0.025 moVL.
4. Substrate plasma with factor VIII content < 1 U/dl.
5. Reference plasma calibrated against the international standard for factor
VIII/von Willebrand factor.

Equipment
1. Plastic tubes to make dilutions.
2. Automated micropipettes equipped with plastic tips.
3. Photo-optical or mechanical coagulometer (Coag A Mate X 2, Organon
Teknika, or other commercial brand). The procedure described can also be
carried out with the manual tilt-tube technique.

Procedure
1. Three dilutions (1:10, 1:50 and 1:250) of the reference and patient plasma
should be prepared immediately before starting the assay. As mentioned
above. the actual dilution of the patient plasma will depend on the expected
factor VIII activity, a rough idea thereof being obtained by the APTT value
of the test plasma.
2. 100 ~l of each dilution is pipetted into the wells of the coagulometer in
balanced order (see above).
3. Add 100 ~ of substrate plasma.
4. Pipette (or program the coagulometer to do so) 100 ~l of APTT reagent.
111
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
5. Incubate the mixture for exactly 5 min (if other reagent is used, follow the
instructions of the manufacturer).
6. At the end of the incubation add 100 ~ of CaC12 0.025 moVL.
7. Record the clotting time which will be used for the calculation of factor
VIII activity (see above).

NORMAL VALUES

Factor VIII clotting activity in the normal population varies between 50 and
150 U/dl, it is recommended that each laboratory establish its own normal
range.

CLINICAL RELEVANCE OF FACTOR VIII MEASUREMENT

The measurement of factor VIII clotting activity in plasma is essential in

diagnosing patients with haemophilia A and monitoring their treatment with
factor VIII concentrates. Low levels of factor VIII are also found in patients
with acquired antibodies to factor VIII and in disseminated intravascular
coagulation. Factor VIII activity is increased transiently after beta-adrenoreceptor
stimulation such as physical exercise and emotional stress, and after infusion of
pharmacological agents such as desmopressin (DDAVP). Since factor VIII is an
acute-phase-reactant, long term increases can be expected during inflammatory
diseases and in post-surgical patients. More recently, factor VIII emerged as a
risk factor for venous thrombosis l2. A study by Koster et al. 12 showed that high
factor VIII activity (> 150 U/dl) is common among consecutive unselected
thrombosis patients and is associated with an increased risk of venous thrombosis.
If this finding is confirmed in subsequent studies, factor VIII measurement should
be included in the laboratory profile to assess the risk of thrombosis in
thrombophilic patients.

References
I. Kirkwood TBL, Rizza CR, Snape TJ, Rhymes IL, Austen DEG. Identification of sources of
interlaboratory variation in factor VIII assay. Br J Haematol1981; 37: 559-68.
2. Nilsson 1M, Kirkwood TBL, Barrowcliffe Tw. In vivo recovery of factor VIII: comparison
of one-stage and two-stage assay methods. Thromb Haemostas 1979; 42: 1230--9.
3. Segatchian MJ, Miller-Andersson M. A colorimetric evaluation of factor VIII: C. Med Lab
Sci 1978; 35: 347-54.
4. Rosen S, Andersson M, Mikaelsson M, Oswaldsson U. Determination of factor VIII activity
with chromogenic substrate kit method. I. Basic performance. Thromb Haemostas 1984; 40:
139-45.
5. Rosen S, Andersson M, Blombiick M et al. Clinical application of a chromogenic substrate
method for the determination of factor VIII activity. Thromb Haemostas 1985; 54: 818-23.
6. Tripodi A, Mannucci PM. Factor VIII activity as measured by an amidolytic assay compared
with a one-stage clotting assay. Am J Clin Patho11986; 86: 341-4.
7. Takase T, Rotblat F, Goodall AH, KemotT PBA, Middleton S, Chand S, Denson KW, Austen
DEG, Tuddenham EGD. Production of factor VIII deficient plasma by immunodepletion
using three monoclonal antibodies. Br J Haematol1987; 66: 497-502.

112
 FACTOR VIII CLOniNG ACTIVITY
8. Ingram GIC. Blood coagulation factor VIII: genetics, physiological control and bioassay. In:
Sobotka H, Stewart CP, eds. Advances in clinical chemistry, vol 8. New York: Academic Press,
1965; 189-236.
9. Counts RB, Hayes JE. A computer program for analysis of clotting factor assays and other
parallel-line bioassay. Am J Clin Patho11979; 71: 167-7l.
10. Williams KN, Davidson JMF, Ingram GIe. A computer program for the analysis of para11el-line
bioassays of clotting factors. Br J Haematol1975; 31: 13-23.
11. Kirkwood TBL, Snape TJ. Biometric principles in clotting and clot lysis assays. Clin Lab
Haemato11980; 2: 155-67.
12. Koster T, Blann AD, Briet E, Vandenbroucke JP, Rosendaal FR. Role of clotting factor VIII
in effect of von Willebrand factor on occurrence of deep-vein thrombosis. Lancet 1995; 345:
152-5.

113
12
Von Willebrand factor
P. M. MANNUCCI and R. COPPOLA

INTRODUCTION

Human von Willebrand factor (vWF) can be measured by functional and

immunological assays. Functional assays explore the ability of plasma vWF to
agglutinate platelets in the presence of the antibiotic ristocetin; hence, this
method is known as ristocetin cofactor activity assayt. The measurement of
ristocetin cofactor activity is complicated by the need to prepare a standardized
suspension of formaldehyde-fixed normal platelets, which makes the assay
unsuitable for less specialized laboratories. The immunological assays are based
on the electroimmunoassay (EIA) of patient and standard samples in agarose-
containing rabbit antibodies to human vWF 2,3. At optimal conditions of pH
and ionic strength the antigen-antibody complexes precipitate as narrow rocket-
shaped peaks which allow the quantitation of vWF antigen by comparing the
peak heights of the patient samples with those of standards containing known
amounts of vWF. More recently, methods based on enzyme-linked immuno-
sorbent assay (ELISA) have been proposed for the measurement of vWF4,s,
with commercial kits available from several manufacturers (e.g. Diagnostica
Stago; American Diagnostica). In general, results obtained with the EIA are
in close agreement with those obtained with ELISA. However, better precision,
together with higher sensitivity, makes ELISA the method of choice for vWF
antigen measurement.

PRINCIPLES AND ASSAY CHARACTERISTICS

The method is used on the sandwich principle. In the first step, plastic wells of
rnicrotitre plates are coated with antibodies against vWF (capture antibody).
In the second step, suitable dilutions of test and standard plasmas are incubated
to allow vWF antigen to bind the capture antibody. After washing the wells
115
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
to remove the unbound antigen, a second antibody to vWF labelled with
horseradish peroxidase (detecting antibody) is added. This results in the
formation of a sandwich complex in a quantity proportional to the vWF
content in patient and standard plasmas. After a second washing to remove
excess detecting antibody, the peroxidase activity is measured photometrically
by addition of hydrogen peroxide and o-phenylenediamine. The optical density
readings are then converted into U/dl of vWF antigen by comparison with a
calibration curve obtained by processing dilutions of a pooled normal plasma
or standard plasma.

METHOD

Recently we produced and characterized five murine monoclonal antibodies

(MoAb) to human vWF. All MoAb were positive in a direct ELISA and in
multimetric analysis of purified vWF; competition experiments set up after
conjugation of peroxidase showed that two of them (llB6.1S and 7G IO.S) were
directed against different vWF epitopes and were therefore used as capture
and detecting antibodies to set up an ELISA assay for measuring vWF antigen6•
This method may be used as a guideline to set up methods with home-made
or commercial antibodies to vWF.

Reagents
1. Coating buffer: 50 mmoVL Tris-HC1, pH 9.00.
2. Over-coating buffer: as for coating buffer, added with 3% bovine serum
albumin.
3. Washing buffer: 20 mmollL Tris-HC1, 100 mmollL NaCl, 0.1% Tween, 0.1%
bovine serum albumin, pH 7.4.
4. Dilution buffer: as for washing buffer, with added 0.5% bovine serum
albumin.
5. Substrate buffer: 22 mmollL trisodium citrate, 50 mmollL Na2HP04' Adjust
pH to 5.5 with H 3P04.
6. 400 ~glml o-phenylenediamine (Sigma).
7.30%H20 2·
S. 3 moVL H 2S04,
9. Two different monoclonal antibodies were used as capture and detecting
antibodies.
Immediately before use, prepare a solution of o-phenylenediamine 40 mg and
H 20 2 40 ~ and make the volume to 100 ml with substrate buffer.

Procedure
The detecting antibody was conjugated with horseradish peroxidase (Sigma~
(final concentration 250 ~glml) according to the method of Nakane and Kawaoi
116
 VON WILLEBRAND FACTOR
and stored in small aliquots at -20°C in the presence of glycerol (50% final
concentration). Polystyrene plates (Nunc) are coated overnight at 4 °C with
125 p,]Jwell of a 2.5 p,g1ml solution of capture antibody in coating buffer. At
the end of the incubation the solution is discarded and the wells are overcoated
with 200 p,l of overcoating buffer for 1 or 2 h. The wells are then washed three
times with washing buffer and added with 100 p.]Jwell of standard plasma (in
duplicate) diluted from 1:50 to 1:3200 in dilution buffer. Patient samples are
tested at two different dilutions chosen according to the expected concentration
of vWF in the sample (usually, 1:100 and 1:400). After incubation for 2 h (or
overnight) at room temperature, the wells are washed three times and added
with 100 ~]Jwell of detecting antibody (5 p,g1ml) in dilution buffer. After
incubation for 1 h at room temperature, the wells are washed three times and
added with 100 p,]Jwell of o-phenylenediamine and H 2 0 2 solution. After
incubation for 10 min at room temperature (dark), the reaction is stopped by
adding 50 p,]Jwell H 2S04 , The optical density is then measured at 486 nm with
a microplate photometer. vWF antigen in patient samples is derived by
extrapolation from the calibration curve (Figure 12.1).
Results obtained with the above method in plasma samples from healthy
subjects and patients with von Willebrand disease compared favourably with
those obtained with the commercial kit Asserachrom vWF (Diagnostica Stago),
which can be used as a suitable alternative.

REFERENCE PLASMA

As mentioned for factor VIII clotting assay, plasmas from a large number of
healthy donors must be pooled in order to get a standard truly representative

,.
III
en
o
2.0 ::a
,.z
III

~
1.5
,.....
1.0 :
ell
:::I
3
0.5

3200 1600 800 400 200 100 50

RECIPROCAL OF PLASMA DILUTIONS

Figure 12.1 Plasma dilution curve of vWF antigen in pooled normal plasma

117
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
of the average normal plasma and hence suitable to construct the calibration
curve for vWF antigen assay. Fifty or more healthy donors, equally divided
between males and females and covering different age classes, are adequate. As
an alternative a pooled plasma from a limited number of donors (even from a
single donor) can be used, provided such plasma is first calibrated against the
International Standard for factor VIII and vWF (National Institute for Bio-
logical Standards and Control). For details on the calibration procedure, see
Chapter 11.

REFERENCE RANGE

vWF antigen in the normal population varies between 50 and 150 D/dI. However,
age group- and blood group-related differences have been reported within the
same population8 ,9. In general, vWF antigen increases with increasing age,
and blood group 0 values are on average lower than those of non-group O.
Accordingly, it is strongly recommended that each laboratory establish its own
normal range.

CLINICAL SIGNIFICANCE OF vWF ANTIGEN MEASUREMENT

The measurement of plasma vWF antigen is essential to differentiate between

haemophilia A and von Willebrand's disease. In haemophilia vWF antigen is
normal, whereas factor VIII is low. In von Willebrand's disease vWF antigen
is usually low, being moderately reduced in type I and severely reduced in type
III. However, a normal level of vWF antigen does not exclude the diagnosis
of von Willebrand's disease, because molecular variants of the disease have
normal or near-normal vWF antigen and reduced ristocetin cofactor activity
(type II) (for a comprehensive review of vWF classification see ref. 10). Being
an acute-phase reactant, vWF is increasing during inflammatory diseases and
in postsurgical patients. In addition, a number of other clinical conditions such
as diabetes mellitus, atherosclerosis, liver cirrhosis, tumours, uraemia, pregnancy-
induced hypertension and thrombocytopenic purpura have been described in
which plasma vWF antigen levels are increased.

References
1. MacFarlane DE, Stibbe J, Kirby EP, Zucker MB, Grant RA, McPherson 1. A method for
assaying von Willebrand factor (ristocetin cofactor). Thromb Diath Haemorrh 1975; 34:
306--8.
2. Laurel CB. Quantitative estimation of proteins by electrophoresis in agarose gel containing
antibodies. Anal Biochem 1966; 15: 45-52.
3. Zimmerman TS, Hoyer LW, Dickson L, Edgington TS. Determination of the von Willebrand's
disease antigen (factor VIII related antigen) in plasma by quantitative immunoelectrophoresis.
J Lab Clin Med 1975; 86: 152-9.
4. Bartlett A, Dormandy KM, Hawkey CM, Stableforth P, Voller A. Factor VIII-related antigen:
measurement by enzyme immunoassay. Br Med J 1976; I: 994-6.
5. Silveira AMV, Yamamoto T, Adamson L, Hessel B, Blomback B. Application of an enzyme-
linked immunosorbent assay (ELISA) to von Willebrand factor (vWF) and its derivatives.
Tbromb Res 1986; 43: 91-102.

118
 VON WILLEBRAND FACTOR
6. Federici AB, Tombesi S, Coppola R, Colibretti ML, Gobbi A, Albertini A, Mannucci PM.
Preliminary results on the characterization of monoclonal antibodies to von Willebrand factor
(vWF). Abstract of the International Symposium on Biotechnology of plasma proteins:
Haemostasis, thrombosis and iron proteins. Florence, 9-11 April 1990.
7. Nakane PK, Kawaoi PA. Peroxidase-labeled antibody: a new method of conjugation. J
Histochem Cytochem 1974; 22: 1084-91.
8. Cox Gill J, Endres-Brooks J, Bauer PJ, Marks WJ, Montgomery RR. The effect of ABO blood
group on the diagnosis of von Willebrand disease. Blood 1987; 69: 1691-5.
9. Rodeghiero F, Castaman G, Dini E. Epidemiological investigation of the prevalence of von
Willebrand's disease. Blood 1987; 69: 454-9.
10. Ruggeri ZM, Zimmerman TS. von Willebrand factor and von Willebrand disease. Blood 1987;
74: 895-904.

119
13
Antithrombin activity and antigen
J. CONARD

INTRODUCTION

Antithrombin (AT, previously named AT III) is the major physiological inhibitor

of blood coagulation. It inactivates mainly thrombin, but also other activated
factors, such as activated factor X (factor Xa). Heparin accelerates the reaction
between AT and thrombin and AT is also called 'heparin cofactor'1.2.
AT is important in maintaining the haemostatic balance, as demonstrated
by the increased risk of venous thrombosis observed in patients with congenital
deficiency. It can be measured by different methods: immunological or
functional, allowing the detection of quantitative or qualitative deficiencies
(type I and type II respectively), possibly associated with a different risk of
thrombosis.

SAMPLING AND STORAGE

Citrated blood is used: nine volumes of blood are collected into tubes containing
one volume of sodium citrate 0.11 or 0.129 moUL. After centrifugation at
about 2000g for 15-20 min, aliquots of 0.5 ml of plasma are transferred to
plastic tubes that are stoppered and frozen at -20°C or lower. They can be
stored up to 4 months before assay.
At the time of the test, samples are rapidly thawed at 37°C. Lipaemic
plasmas may not be suitable for nephelometric assays.

STANDARD AND CONTROLS

I. AT standard: a freeze-dried AT standard can be obtained from the National
Institute for Biological Standards and Control.
121
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
2. Local reference plasma: the local reference plasma is prepared from a mixture
of equal amounts of citrated plasmas from more than 20 healthy individuals
of both sexes, who are taking no oral contraceptive or other medication.
Aliquots are frozen at -20°C or lower.
3. AT control plasmas: different control plasmas (normal or pathological range)
are commercially available, for instance from Chromogenix or Diagnostica
Stago.

ACTIVITY ASSAVS3 ,4

AT is measured by chromogenic methods in the presence (heparin cofactor

activity) or in the absence of heparin (progressive antithrombin activity). In
routine practice, heparin cofactor activity assays are preferred because they
allow the detection of all types of deficiencies, type I as well as type II. Kits
are commercially available and assays are based on thrombin or factor Xa
inhibition.

Antithrombin Inhibition
Principle
Plasma AT is determined as the capacity for inactivating thrombin in the
presence of heparin which has a catalytic effect on the reaction AT-thrombin.
An excess of purified thrombin is added to the plasma. The amount of thrombin
inhibited by the complex AT-heparin depends on the amount of AT in the
sample. Residual thrombin hydrolyses a thrombin-sensitive chromogenic
substrate and para-nitroaniline (PNA) released is measured at 405 nm. The
method can be summarized as follows:
1. AT + heparin in excess ~ [AT-heparin]
2. [AT-heparin] + thrombin ~ [AT-heparin-thrombin] + residual
in excess thrombin
3. Substrate + residual thrombin ~ pNA (405 nm).

Assay procedure
Bovine thrombin is more appropriate than human thrombin so that the test is
specific for AT, by avoiding the interference of heparin cofactor 115,6.
An end-point assay has been used in manual methods but kinetic automated
assays are preferred for better reliability7,8. Many instruments are suitable and
most of the automates used in coagulation laboratories are adapted for this
test: Diagnostica Stago, Dade-Behring, Instrumentation Laboratory, Bio-
Merieux, for instance.
Different substrates are used: S-2238 Chromogenix, CBS 3447 Stago.
The AT determination can also be performed in some fully automated instru-
ments such as the ACA Du Pont. This is a 'closed' system requiring both the
122
 ANTITHROMBIN ACTIVITY AND ANTIGEN

instrument and special packs. The principle is the same but the substrate is different.
The instrument and reagents are remarkably stable, allowing the performance of
the calibration only when changing pack batches. Pathological and normal
lyophilized control plasmas are checked regularly, about once per week.

Anti-Xa inhibition
Factor Xa, like bovine (but not human) thrombin, has the advantage of being
more specific for AT, since it is not inhibited by heparin cofactor II. For this
reason an AT chromogenic assay based on factor Xa inhibition has been
developed9 .
The principle of the test is similar to the antithrombin inhibition test but
the substrate is different: 8-2765 Chromogenix.

ANTIGEN ASSAYS

AT may be measured by immunological methods: radial immunodiffusion,

electroimmunodiffusion (Laure1l method) or nephelometrylO. The LaurelI method
is time-consuming and not appropriate for routine use. Immunodiffusion had a
coefficient of variation which is relatively high and requires 48 h for completion.
Nephelometry, a fast and automated method, will be described now.

Principle
Immunoprecipitation is performed in a liquid phase by adsorption on latex
particles. A laser nephelometer measures light scattered in a forward direction
by immunocomplexes in an incident beam of monochromatic light. The diffused
light is measured in an angle between 13° and 24°. The reading is performed
at 840 nm.

Instrument
The Behring Nephelometer Analyser (BNA) is an automated instrument, con-
venient to perform large series of AT determinations (45 cuvettes divided into
nine groups of five cuvettes each).

Reagents and standard

AT antiserum and a normal protein standard are provided by Dade Behring.
Results of the standard are expressed in mg/dl and are usually 30 mg/dl, but
we prefer to give results as a percentage of the frozen local normal pool.

Assay procedure
After the plasmas have been introduced into the cuvettes, all steps are automatic.
In lipaemic or lyophilized plasmas, AT determination may not be possible due
123
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
to the turbidity; in this situation the result is marked with a cross. Haemolysis
does not interfere with the result.

REFERENCE VALUES

Normal values are in the range 80-120%.

AT levels are reduced in some physiological conditions and treatments. Low
levels are observed in the neonatal period ll ,12 and in pregnancy (decrease of
about 10%)13. In men AT tends to decrease after the age of 50, while in women
an increase has been associated with the menopausal status 14,15.

CONGENITAL DEFICIENCY IN AT

Congenital AT deficiency is classically associated with a high risk of venous

thromboembolism: deep vein thrombosis, pulmonary embolism, or other
locations such as cerebral vein thrombosis, mesenteric vein thrombosis1 6,17.
Arterial thromboses are extremely rare. The deficiency is transmitted as an
autosomal dominant trait. It is a relatively infrequent finding in the normal
population (0.02%) and in patients with history of venous thromboses (1-4%
depending on the selection of patients).
The first episode is usually observed between the ages of 20 and 30, and
about 85% of patients older than 50 have had at least one thrombosis. A
triggering factor (pregnancy, oral contraceptive, surgical procedure, immobili-
zation) is often present, but no known predisposing factor is found in
approximately half the cases.
Heterozygous affected members have a level of about 50--60% of the normal
value. Comparison of the level of AT heparin cofactor activity and antigen
allows the detection of different types of AT deficiency. The following
classification has been adopted by the Scientific and Standardization Committee
of the International Society on Thrombosis and Haemostasis1 7 :
1. Type I or classical deficiency is a quantitative defect, characterized by a
parallel decrease of AT activity and antigen.
2. Type II is a functional defect, with a low AT heparin cofactor activity but
a normal or subnormal level of AT antigen, due to the production of a
variant protein. Three subtypes have been recognized:
RS, effect on reactive site
HBS, effect on heparin binding site
PE, pleiotropic (or multiple) effect.
Types I and II of AT deficiency have decreased AT heparin cofactor activity,
but antigen is decreased or normal according to the type (Table 13.1). This
explains why the heparin cofactor activity assay is the method of first choice.
The subtype of type II deficiency may be suspected by the determination of
the AT progressive activity (in the absence of heparin), or by the crossed-
immunoelectrophoresis pattern in the presence or absence of heparin. This
can be confirmed by identification of the mutation.
124
 ANTITHROMBIN ACTIVITY AND ANTIGEN

Table 13.1 Classification of congenital AT deficiency

Heparin cofactor Antigen Progressive AT activity

activity

Type!
Type II
RS Normal ~
HBS Normal Normal
PE Limit normal ~

The identification of subtype II HBS (previously named IIc) is of clinical

importance since patients heterozygous for this mutation are at low risk of
thrombosis, in contrast to the high risk observed in all other types of deficiency
(type I, type II RS or PE)18. In addition, homozygous AT deficiency has been
found only for subtype II HBS; it may not be compatible with life for other
subtypes. Patients with HBS subtype have a normal antigen level, a low AT
heparin cofactor activity but a normal progressive AT activity (in the absence
of heparin); the crossed immunoelectrophoresis in the presence of heparin
according to Sas et al. 19 shows two peaks, corresponding to the fraction of AT
that binds to heparin and the other one which does not bind. Mutations involved
in HBS subtype are mostly 24Arg, 47Arg and 129Arg; alterations at these
positions prevent the interaction of the essential pentasaccharide sequence of
heparin with AT.
Many insertions, deletions and other mutations have been identified in type
I and type II RS and PE and reported in a database 17. They are important for
a better understanding of the relations of structure and function of the protein
but, to our knowledge, they have no clinical consequence.
Thrombosis is treated by unfractionated or low molecular weight heparin.
An AT level higher than 35% seems to be required to have a satisfactory
anticoagulant activity of heparin 2o . Antithrombin concentrates are available
but their use should probably be restricted to very few conditions 21 : severe
episodes of thrombosis, AT level below 35%, or possibly at the time of child-
birth.Treatment with oral anticoagulant is effective but its duration is still a
matter of debate. It has been proposed that long-term therapy be administered
only after at least two episodes of thrombosis, especially if they were idiopathic.
Asymptomatic patients are prophylactically treated in high-risk situations:
surgery, immobilization, pregnancy22. Oestrogens for contraception or re-
placement therapy are contraindicated.

ACQUIRED DEFICIENCY

Acquired deficiency may be related to diseases or to treatments, and is

inconstantly associated with thrombosis (Table 13.2)23.

Deficiency related to pathological conditions

1. Liver cirrhosis: AT is decreased in liver cirrhosis, as are other coagulation
125
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Table 13.2 Acquired deficiency in AT

Pathological conditions
Liver cirrhosis
Nephrotic syndrome
Postoperative period
Disseminated intravascular coagulation
Treatments
Heparin
Hormones: (a) oral contraceptives (contain ethinyloestradiol); (b) replacement therapy
with conjugated equine oestrogens or oral oestradiol
L-Asparaginase

factors and inhibitors synthesized in the liver. This deficiency is not usually
associated with thrombosis.
2. Nephrotic syndrome: when the urinary loss of AT exceeds its synthesis, a
plasma deficiency is observed and, together with increased levels of co-
agulation factors, it might be involved in the increased risk of thrombosis
observed in this disease.
3. Postoperative period: a decrease has been reported after knee surgery, and
since this type of surgery is highly thrombogenic, some authors have proposed
AT concentrates in association with heparin24.
4. Disseminated intravascular coagulation (DIe): AT is inconstantly decreased
and it cannot be used as a criterion for the diagnosis of DIe. The decrease
is more frequent in patients with DIC related to cancer with liver metastases.
In severe sepsis, studies have shown an improvement of the biological signs
of DIC after administration of AT concentrates but the effect on mortality
was very moderate or absent.

Deficiency induced by treatments

1. Heparin: unfractionated heparin increases the turnover of AT, resulting in
a decrease in plasma AT when the dose of heparin is high and the admin-
istration prolonged. Low molecular weight heparins do not have such an
effect.
2. Hormones: oral contraception with oestro-progestogens induces a moderate
decrease (about 10%) of AT, related to the dose of ethinyloestradiol. This
may contribute to the increased risk of venous thromboembolism reported
during these treatments. Replacement therapy with conjugated equine
oestrogens, and also with the natural oestrogen, 17~-oestradiol, by oral
route, also decreases AT, in contrast with oestradiol used by cutaneous
route25- 27 . Progestogens do not appear to have any effect on AT, except
lynestrenol 28 .
3. L-Asparaginase: this drug, used in the treatment of leukaemia and other
haematological diseases, is associated with a decrease in AT and may have
a role in the occurrence of the venous thrombosis observed during these
treatments.
126
 ANTITHROMBIN ACTIVITY AND ANTIGEN

CONCLUSION

Congenital AT deficiency, although rare, is an important risk factor for venous

thromboembolism. The method of choice for its detection is the determination
of the heparin cofactor activity. When abnormal, the antigen assay should be
performed as a second step in order to differentiate type I from type II. In
routine practice, investigations can be interrupted if the AT deficiency is of
type I. In contrast, when a type II is detected, it is of clinical importance to do
further investigations in order to discriminate type II HBS at low risk of
thrombosis from other varieties of type II which have a high risk of thrombosis.

References
1. Abildgaard U. Antithrombin and related inhibitors of coagulation. In: Poller L, ed. Recent
advances in blood coagulation. Edinburgh: Churchill Livingstone, 1981; 151.
2. Rosenberg RD, Bauer KA. The heparin-antithrombin system: a natural anticoagulant
mechanism. In: Colman RW, Hirsch J, Marder VJ, Salzman EW, eds. Hemostasis and
Thrombosis: basic principles and clinical practice. Philadelphia: Lippincott, 1994; 837--60.
3. Handeland GF, Abildgaard U, Aasen AO. Simplified assay for antithrombin III activity using
chromogenic peptide substrate. Manual and automated method. Scand J Haematol1983; 31:
427-36.
4. Odegard OR, Abildgaard U. Antithrombin III: critical review of assay methods. Signification
of variations in health and disease. Haemostasis 1978; 7: 127-34.
5. Conard J, Horellou MH, Bara L. Bovine or human thrombin in antithrombin III (AT III)
amidolytic assays. Thromb Haemostas 1985; 54: 25.
6. Friberger P, Egeberg N, Holmer E, Hellgren M, Blombiick M. Antithrombin assay. The use
of human or bovine thrombin and the observation of a second heparin cofactor. Thromb Res
1982; 25: 433--6.
7. Conard J. Automation of antithrombin III. Methods on routinely available instruments. Sem
Thromb Haemostas 1983; 9: 263-7.
8. Odegard OR, Rosenlund B, Ervik E. Automated antithrombin III assay with a centrifugal
analyzer. Haemostasis 1978; 7: 202-9.
9. Demers D, Henderson P, Blajchman MA, Wells MJ, Mitchell L, Johnston M, Ofosu FA,
Fernandez-Rachubinski F, Andrew M, Hirsch J, Ginsberg JS. An antithrombin III assay based
on factor Xa inhibition provides a more reliable test to identify congenital antithrombin III
deficiency than an assay based on thrombin inhibition. Thromb Haemostas 1993; 69: 231-5.
10. Parvez Z, Fareed J, Argawal P, Messmore HI, Moncada R. Laser nephelometric quantitation
of antithrombin III (AT III). Thromb Res 1981; 24: 367-37712.
II. Peters M, Jansen E, Ten Cate.Jw, Kahle LH, Ockelford P, Breederveld C. Neonatal antithrombin
III. Br J Haematol1984; 58: 579-87.
12. Teger-Nilsson AC. Antithrombin in infancy and childhood. Acta Paediatr Scand 1975; 64:
624-8.
13. Weenink GH, Treffers PE, Kahle LH, Ten Cate JW. Antithrombin III in normal pregnancy.
Thromb Res 1982; 26: 281-7.
14. Meade T, Dyer S, Howard DJ, Imeson JD, Stirling Y. Antithrombin III and procoagulant
activity: sex differences and effects of the menopause. Br J Haematol1990; 74: 77-81.
15. Tait RC, Walker 10, Davidson JF, Islam SIA, Mitchell R. Antithrombin III activity in healthy
blood donors: age and sex related changes and the prevalence of asymptomatic deficiency.
Br J Haemato11990; 75: 141-2.
16. Thaler E, Lechner K. Antithrombin deficiency and thromboembolism. In: Prentice CRM, ed.
Qinics in haematology. London: Saunders, 1981; 369-90.
17. Lane DA, Bayston T, 01ds RJ, Fitches AC, Cooper DN, Millar DS, Jochmans K, Perry DJ,
Okajima K, Thein SL, Emmerich J. Antithrombin mutation database: 2nd (1997) update.
Thromb Haemostas 1997; 77: 197-211.
127
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
18. Finazzi G, Caccia R, Barbui T. Different prevalence of thromboembolism in the subtypes of
congenital antithrombin III deficiency: review of 404 cases. Thromb Haemostas 1987; 58:
1094.
19. Sas G, Pepper DS, Cash ID. Further investigations on antithrombin III in the plasma of
patients with the abnormality antithrombin 'Budapest'. Thromb Diathes Haemorrh 1975; 33:
564-72.
20. Krudler JW, Strebus AF, Meinders AB, Briet E. Anticoagulant effect of unfractioned heparin
in antithrombin-depleted plasma in vitro. Haemostasis 1996; 26: 85-9.
21. Lechner K, Kyrle PA. Antithrombin III concentrates. Are they clinically useful? Thromb
Haemostas 1995; 73: 340-8.
22. Conard J, Horellou MH, Van Dreden P, Lecompte T, Samama M. Thrombosis and pregnancy
in congenital deficiencies in AT III, protein C or protein S: study of 78 women. Thromb
Haemostas 1990; 63: 319-20.
23. Beresford CH. Antithrombin III deficiency. Blood 1998; 2: 239-50.
24. Francis ChW, Pellegrini VD Jr, Stulberg BN, Miller ML, Marder VJ. Prevention of venous
thrombosis after total knee arthroplasty. J Bone Joint Surg 1990; 72A: 976-82.
25. Conard J, Samama M, Basdevant A, Guy-Grand B, de Lignieres B. Differential AT III
response to oral and parenteral administration of 17~-estradiol. Thromb Haemostas 1983;
49: 252.
26. Caine YG, Bauer KA, Barzegar S, Ten Cate H, Sacks FM, Waish BW, Schiff I, Rosenberg
RD. Coagulation activation following estrogen administration to postmenopausal women.
Thromb Haemostas 1992; 68: 392-5.
27. Scarabin PY, Alhenec-Gelas M, Plu-Bureau G, Taisne P, Agher R, Aiach A. Effects of oral
and transdermal estrogen/progesterone regimens on blood coagulation and fibrinolysis in
postmenopausal women. A randomized controlled trial. Arterioseler Thromb Vase Bioi 1997;
17: 3071-8.
28. Bounameaux H, Duckert F, Walter M, Bounameaux Y. The determination of antithrombin
III. Comparison of six methods. Effect of oral contraceptive therapy. Thromb Haemostas
1978; 39: 607-15.

128
14
Protein C activity and antigen
R. M. BERTINA

INTRODUCTION

Protein C (PC) is a vitamin K-dependent plasma glycoprotein (MW 62 000)1.2.

In the blood it circulates as the inactive zymogen of a serine protease (activated
protein C, APC), mostly in the form of a two-chain molecule consisting of a
light chain and heavy chain linked by one disulphide bond. Protein C is
synthesized in the liver. During post-translational modification nine glutamic
acid residues in the aminoterminus of the light chain are carboxylated by a
vitamin K-dependent carboxylase. The biological half-life of protein C is about
8 h 3• In pooled normal plasma the concentration of protein C is approximately
4 J.1g/ml (63 nmoVL).

Physiological role
Activated protein C is the key enzyme of the protein C anticoagulant pathway.
Its anticoagulant properties reside in its capacity to inactivate the cofactors Va
and VIlla by proteolytic degradation4. Protein S - another vitamin K-dependent
plasma protein, anionic phospholipid membranes and Ca2 + ions are essential
cofactors for the inactivation reactions. By inactivating the cofactors Va and
VllIa6 •7 , activated protein C will dramatically reduce the rate of thrombin
formation, as can be visualized by the increase in the activated partial thrombo-
plastin time (AnT) after addition of APC to pooled normal plasma.
The activation of protein C occurs at the endothelial surface by the thrombin-
thrombomodulin complex8 • Thrombomodulin is a membrane protein which
.serves as a receptor/cofactor for thrombin. Binding of thrombin to throm-
bomodulin changes the enzymatic properties of thrombin from a procoagulant
enzyme (activation of fibrinogen and platelets) to an anticoagulant enzyme
(activation of protein C). During activation an Arg-Leu bond is cleaved, after
129
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

which an aminoterminal peptide of 12 amino acids is released from the heavy

chain.
Under physiological conditions (i.e. in the presence of Ca2+) the activation
reaction is completely dependent on the availability of thrombomodulin9. In
vitro protein C can be activated by thrombin in the absence of Ca2+ ions.
However, rather high enzyme/substrate ratios are required to reach complete
conversion to APC within a reasonable time interval9.
In citrated plasma APC is neutralized by forming complexes with the APC
inhibitor (PCI or PAI-3) and ul-antiproteinase (u1-antitrypsin)JO,II. The reaction
of APC with the APC inhibitor is accelerated to some extent by relatively high
concentrations of heparin. In whole blood - that still contains physiological
Ca2+ concentrations - APC is also neutralized by u2-macroglobulin12.

Pathophysiological aspects
Hereditary defects
The importance of the protein C anticoagulant system for the control of
thrombin formation in vivo is most dramatically demonstrated in neonates who
are homozygotes or double heterozygotes for hereditary protein C deficiencyJ3,14.
These patients develop massive and fatal (if untreated) thromboembolism
shortly after birth. The most prominent clinical manifestations are large purpuric
skin lesions (purpura fulminans syndrome), central nervous system thrombosis
and central ophthalmic thrombosis.
Interestingly the clinical phenotype of heterozygotes for a hereditary protein
C deficiency is not uniform. Two different phenotypes can be recognized. In
the clinically dominant type of protein C deficien? heterozygotes have an
increased risk of developing venous thrombosisl5--1 . Families with this type
of protein C deficiency are identified by screening symptomatic patients with
unexplained familial thrombophilia: about 5% of these patients are protein
C_deficient I8 ,19. The prevalence of this phenotype in the general population
has been estimated to be 1116000 19 .
In the clinically recessive type of protein C deficiency heterozygotes only
very rarely have thrombosis20 . Families with this type of protein C deficiency
are found among the parents of homo~gous protein C-deficient patients but
also among healthy blood donors 13,20-2 . The prevalence of this phenotype in
the general population has been estimated to be 1130021 ,22.
This apparent discrepancy is explained by the fact that familial thrombophilia
is a multigenic disorder. This means that in the clinically dominant type of
protein C deficiency, apart from the protein C gene defect, other genetic defects
are segregating that contribute to the thrombotic risk. This was first demonstrated
by Koeleman et al. 23 for thrombophilic families in which both a protein C gene
mutation and the factor V Leiden mutation were segregating.
The laboratory diagnosis of hereditary protein C deficiency focuses on the
identification of homozygotes/double heterozygotes and heterozygotes of the
clinically dominant form of the disorder. If not on oral anticoagulant treatment
protein C (activity or antigen) levels should be lower than the lower limit of
130
 PROTEIN C ACTIVITY AND ANTIGEN

the nonnal range, while the plasma concentration of other vitamin K-dependent
proteins is within the normal range 24 . The treatment of patients with oral
anticoagulants offers a special problem for laboratory diagnosis because the
treatment in itself causes a decrease in the plasma protein C level (dependent
on the intensity of the treatment). For typical therapeutic doses of coumarins
protein C antigen and activity decrease to about 50% and 25%, respectively24·25.
Therefore, in these patients, protein C levels should be below the lower limit
of the range observed at the relevant intensity of treatment, while the ratios
PC/FII, PC/FX and/or PC/FYII should also be lower than normally observed
in these patients24.

Acquired abnormalities
Abnormal protein C levels have also been reported in a variety of clinical con-
ditions. Increased protein C levels have been reported for users of oral contracep-
tives26, for ~atients with (acute) angina pectoris27, and patients with a nephrotic
syndrome2 ,29. Reduced levels of protein C have been reported for patients in
the postoperative period30 and for patients with liver disease 31 , various forms
of disseminated intravascular coagulation (DIC)32, insulin-dependent
diabetes33,34, essential hypertension 35 and sickle-cell disease 36. Sometimes only
a decrease in specific activity or protein C activity/protein C antigen ratio is
observed30. In general it is not clear whether the reduced protein C levels trigger
or provoke the development of thrombosis in such patients.
Further, it is important to realize that protein C levels increase with age
(about 4% per decade)37 and that treatment protocols might also influence
protein C levels. Danazol and stanozoloes,39, for instance, have been reported
to result in an increase in plasma protein C levels, while treatment with
L-asparaginase40 or oral anticoagulants41 will result in a decrease of plasma
protein C levels.

Protein C assays
Two different types of assay can be used to measure the concentration of
protein C in plasma: immunological assays and functional assays (for reviews
see ref. 421. The immunological assays include electroimmunoassays16,43,
ELISA44,4 and radioimmunoassays46,47. Functional assays include a variety
of methods differing in concept, specificity and complexity48. All these assays
differ in sensitivity, precision, accuracy and costs. The selection of a protein
C assay for use in the daily routine therefore strongly depends on the infra-
structure of the local clinical laboratory.

Protein C antigen
Initially the recommended method for measuring protein C antigen was the
electroimmunoassay (Laurell or rocket assay) using a locally prepared rabbit
131
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
antiserum against human protein CiS. To date anti-protein C serum is com-
mercially available from many different manufacturers; some manufacturers
even provide ready-to-use Laurell plates for the measurement of protein C
antigen (Diagnostica Stago, American Diagnostica).

Principle of electroimmunoassay
A fixed amount of sample is allowed to migrate in an electric field in an agarose
gel containing specific anti-protein C antibodies at alkaline pH (8.8). Negatively
charged proteins such as protein C will migrate towards the anode. The protein
C will interact with the specific rabbit anti-protein C antibodies. These interac-
tions will result in the formation of insoluble immunoprecipitates (precipita-
tion peaks or rockets) after equilibrium has been established. The length of the
precipitation rocket is directly related to the concentration of protein C in the
plasma. The protein C antigen concentration in a test sample is determined by
measuring the length of the precipitation rocket and reading the antigen content
from a calibration curve which relates length of the rocket to the protein C
antigen content.

Buffers and agarose

The same buffer is used as gel buffer, dilution buffer and electrophoresis buffer,
e.g. 31.6 mmoVL Tricine (Sigma), 91.5 mmoVL Tris, 1 mmoVL EDTA (PH
8.6). Addition of EDTA is needed to achieve identical migration rates for
non-carboxylated protein C and fully carboxylated protein C 43 • A suitable
source of agarose can be Seakem LE Agarose (FMC Bioproducts).
For each plate (10 x 10 cm) 120 mg of agarose is mixed with 12 ml of
electrophoresis buffer in a 20 ml glass tube, placed in a boiling water bath or
thermo-bloc (100 DC) until complete solution of the agarose, and transferred
to a thermo-bloc at 54 DC, where it is stored for at least 30 min before pouring
the plates.

Preparation of the plates, electrophoresis

A glass plate (10 x 10 x 0.15 em) or a piece of 0.2 mm Gelbond® film (FMC
Bioproducts) of similar size is placed on a horizontal table. To 12 ml of 1%
agarose solution (54 DC) the required amount of antiserum is added and mixed
thoroughly, after which the agarose solution is poured evenly over the plate.
After 10 min the solidified gel is placed in a refrigerator (4 DC) in a moist
atmosphere. It should harden for at least 1 h before wells can be punched
(usually 19 wells of 2 mm or 10 wells of 4 mm diameter, at about 3 em from
the edge).
Duplicate samples (4 III in 2 mm wells or 13 III in 4 mm wells) can be applied
in balanced order. Electrophoresis time is 20 h at 120 V (7 mNplate; on the
plate 3.75 V/em).
132
 PROTEIN C ACTIVITY AND ANTIGEN

Staining of the plates

After electrophoresis the gels are rinsed in 0.9% NaCI for about 4 h (room
temperature), pressed, dried and stained with a solution of Coomassie Brillant
Blue R 250 (Serva; 2.5 g in 11 methanol/acetic acid/water, 5/1/5). Afterwards
the excess of staining solution is removed with methanol/acetic acidlwater
(5/1/5). This solution is renewed repeatedly until the plate is clean. The plate
is rinsed with distilled water and dried.

Calculations
The assay is calibrated with dilutions of a pooled normal plasma (1/1, 1/2, 1/4,
1/8) in electrophoresis buffer. Usually the potency of this standard is arbitrarily
set to 100% or 1 unitlml (1 unit being the amount of protein C present in 1 ml
of pooled normal plasma). Some of the commercial standards have been
calibrated against the first international standard for protein C in plasma which
contains 0.82 IU per ampoule49 .
The length of the rockets obtained for the dilutions of the standard is plotted
against the concentration of protein C antigen (percentage, U/ml, IU/ml) on
log-log paper for the construction of the calibration curve (usually a straight
line). The lengths of the rockets for test samples are read on this calibration
curve by intrapolation.

Normal ranges
Among healthy volunteers protein C antigen varies between 0.65 and 1.45 U/ml
=
(n 33; see ref. 16). In patients on oral anticoagulant treatment protein C
antigen is a function of the intensity of treatment (see Table 14.1)1 .

General comments
For the e1ectroimmunoassay of protein C antigen it is essential that ethylene-
diamine tetra-acetic acid (EDTA) is present in the gel and electrophoresis buffer.

Table 14.1 Protein C antigen during stable oral anticoagulant treatment

INR* Protein C antigen (Ulml)

Mean Range

2.4-2.8 0.47±0.09 0.30--0.64

2.8-3.2 0.41±0.08 0.30--0.68
3.2-3.6 0.40±0.08 0.30--0.68
3.6-4.0 0.43±0.09 0.25--0.64
4.0-4.4 0.38±0.08 0.21--0.56
>4.4 0.33±0.07 0.21--0.45

*INR =international normalized ratio

133
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Only under these conditions will the fully carboxylated and non-carboxylated
forms of protein C migrate at an identical rate. This is important because
otherwise the higher mobility of the non-carboxylated protein (as present in
plasmas of patients treated with oral anticoagulants) will result in relatively
over-long precipitation ~eaks and therefore in apparently higher protein C
antigen concentrations43 , 0. A disadvantage of the use of EDTA is that in most
cases twice as much antiserum is needed to obtain the same peak height for
undiluted pooled normal plasma. Good-quality rabbit and anti-protein C serum
can be used at concentrations between 0.5% and 2%.
The staining and destaining solutions contain methanol. If the use of methanol
needs to be avoided, methanol can be replaced by ethanol (some loss in destaining
efficiency).

Alternative protein C antigen assays

Many different protein C antigen assays have been described in the literature.
From these only the ELISA systems are commercially available: ELISA based
on the use of rabbit anti-protein C immunoglobulins (Asserachrom Protein C,
Diagnostica Stago), ELISA based on the use of two different murine monoclonal
antibodies (Thrombonostika® Protein C, Organon Teknika, or Technoclone®
Protein C), or isolated (conjugated) anti-protein C immunoglobulins (Dako).
One should realize that slight differences in specificity may exist between the
various protein C antigen assays, especially in clinical samples: in patients on
oral anticoagulants protein C antigen concentrations determined by polyclonal
ELISA are found to be lower than by Laurell assay50,51. Similarly complexes
between activated protein C and its plasma inhibitors are completely recovered
in the Laurell assay, only for 50% in the 'polyclonal' ELISA tests 50 and not at
all in the monoclonal ELISA from Organon Teknika (because one of the
monoclonals is directed against the active site of APC). A general description
of an ELISA procedure can be found in the chapter on Protein S antigen assays
(Chapter 15).

Protein C activity assay

The ideal protein C activity assay would be a test which measures all functional
aspects of the protein C molecule (Ca 2+ binding, activation by thrombin-
thrombomodulin, interaction with protein S, proteolytic inactivation of factors
Va and VIlla). Such an assay is at present not available. However, today most
of the functional protein C assays make use of the snake venom Protac® to
activate protein C directly in plasma. These assays are considered as second-
generation protein C activity assays. First-generation tests used to be rather
complicated in their design, and laborious, and were essentially based on a
three-step procedure.
1. Partial purification of protein C from the test sample52- 56; this is included
in most of the first-generation assays, but omitted in many of the second-
generation assays.
134
 PROTEIN C ACTIVITY AND ANTIGEN
2. Activation of protein C by thrombin alone (in the absence of Ca2 +)52,SS,
by the thrombin-thrombomodulin complex (in the presence of Ca2 +)S3-S6
or by the protein C activator from the venom of Agkistrodon contortrix
contortrix"57-59.
3. Measurement of the activity of the APC formed with a spectrophotometric
method (kinetic or end-point methods) or with a clotting method 52- s6 •

Second-generation tests
As mentioned above, the discovery of a specific protein C activator (Protac®)
in the venom of Agkistrodon contortrix contortrix 60-62 resulted in a second
generation of protein C activity assays. Because the activity of Protac® is not
dependent on ea2+ ions, and is insensitive to plasma protease inhibitors, Protac®
can be used to activate protein C directly in plasma; during a second incubation
the activity of the APC formed can be quantitated with a spectrophotometric
or clotting technique. Several manufacturers distribute these protein C activity
tests (e.g. Spectrolyse@ and Bioclot@, from Biopool, MDA@ Protein C from
Organon Teknika, Stachrom® Protein C and Staclot® Protein C from Diagnostica
Stago and Coamatic® Protein C from Chromogenix). Most of the chromogenic
tests use a different chromogenic substrate for the measurement of APe. Some
of these tests have been used and evaluated in comparative studies49,50,59,63,64.
Most of these tests have both end-point and kinetic modifications. For several
reasons the author prefers kinetic versions of the test. One of these reasons is
that it is sometimes impossible to have a satisfactory blank in the end-point
method (see, for further discussion, ref. 48). In the test of Biopool (Spectrolyse®)
such problems have been solved elegantly by including an inhibitory anti-protein
C antibody in the blank.
A general problem with the protein C anticoagulant activity assays is the
stability of the standards with respect to the actual prolongation of clotting
time. This needs to be checked and confirmed very carefully. An obvious
advantage of these clotting tests is that they will also detect these protein C
variants that have defects interfering with the proper formation of a Ca2 +_
dependent conformation (these protein C variants have normal amidolytic
activity and reduced anticoagulant activity). On the other hand these variants
are rather rare6S . Moreover, many of the protein C anticoagulant activity tests
have been found to be sensitive to some extent to the presence of the factor V
Leiden mutation66,67.

Protein C activity in patients on oral anticoagulant treatment

From international collaborative studies it has become clear that the different
types of protein C activity assays can all be used for the measurement of
functional r.rotein C in the plasma of patients not treated with oral anti-
coagulants4 ,68. However, in plasma of a patient on oral anticoagulant treatment
the actual J'rotein C activity concentration will depend largely on the method
used41 ,48,S • The main reason for this is that the protein C activators used -
135
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Protac® or thrombin-(thrombomodulin) - cleave both the carboxylated and
non-carboxylated forms of protein C. The effect on the test result will then
largely depend on the design of the test. In the first-generation tests non-
carboxylated (nc) protein C is separated from the carboxylated protein C during
the first step (AI(OHkadsorption)55. In this type of assay the concentration
of carboxylated protein C is measured. This parameter will vary largely with
the intensity of the treatment (much more so than the protein C antigen
concentration) and also the ratio carboxylated protein Cltotal protein C will
decrease with increasing intensity of treatment. In the so-called snake venom
assays both the carboxylated and the non-carboxylated forms of protein Care
activated by Protac®. In the spectrophotometric snake-venom assay one would
therefore expect that the protein C activity concentration would be equal to
the protein C antigen concentration. For most assays, however, the actual
measured protein C activity levels lie somewhere between the protein C activity
as measured with a first-generation test (only carboxylated protein C) and the
result of a protein C antigen assay (total protein C), indicating that possibly
the activation of nc protein C is not complete, or that the yield of ncAPC in
the assay is different from that of APC. In the anticoagulant snake venom
assays the amount of APC formed is quantitated by measuring its effect on a
global clotting time like the APTT or PT, which is sensitive to factor V (and
factor VIII). With this type of assay the measured protein C activity levels are
much lower than that found with the first-generation tests. This may be explained
by assuming that ncAPC will behave as a competitive inhibitor of APC in the
APTT test (for instance by still binding to protein S). Altogether, one has to
be very careful in selecting a protein C activity assay for application in a routine
clinical laboratory, especially when many samples of patients on oral anti-
coagulants are offered for analysis. The use of functional protein C assays for
the laboratory diagnosis of hereditary protein C deficiency therefore still needs
further evaluation.

References
1. Esmon CT. The protein C anticoagulant pathway. Arterioscler Thromb 1992; 12: 135-45.
2. Oahlbiick B, Stenfio J. A natural anticoagulant pathway: protein C, S, C4b-binding protein
and thrombomodulin. In: Bloom AL, Forbes CO, Thomas OP, Tuddenham EGO, eds.
Haemostasis and thrombosis. Edinburgh: Churchill Livingstone 1994; 671-98.
3. Sills RH, Marlar RA, Montgomery RR, Oesphande GN, Humbert JR. Severe homozygous
protein C deficiency. J Pediat 1984; 105: 409-13.
4. Marlar RA, Kleiss AJ, Griffin JH. Mechanism of action of human activated protein C, a
thrombin-dependent anticoagulant enzyme. Blood 1982; 59: 1067-72.
5. Walker FJ. Protein S and the regulation of activated protein C. Sern Thromb Hemostas 1984;
10: 131-8.
6. Koedam JA, Meijers JCM, Sixma 11, Bouma BN. Inactivation of human factor VIII by
activated protein C. J Clin Invest 1988; 82: 1236-43.
7. Suzuki K, Stenflo J, Oahlbiick B, Teodorsson B. Inhibition of human coagulation factor V
by activated protein C. J BioI Chern 1983; 258: 1914-20.
8. Esmon CT. The roles of protein C and thrombomodulin in the regulation of blood coagulation.
J Bioi Chern 1989; 264: 4743-6.
9. Esmon NL, De Bault LE, Esmon CT. Proteolytic formation and properties of y-carboxyglutamic
acid - domainless protein C. J Bioi Chern 1983; 258: 5548-53.

136
 PROTEIN C ACTIVITY AND ANTIGEN
10. Suzuki K, Nishioka I, Hashimoto S. Protein C inhibitor. Purification from human plasma
and characterization. I Bioi Chern 1983; 258: 163-8.
II. Van der Meer FIM, van Tilburg NH, van Wijngaarden A, van der Linden IK, Briet E, Bertina
RM. A second plasma inhibitor of activated protein C: ucantitrypsin. Thromb Haernostas
1989; 62: 756-62.
12. Heeb MI, Gruber A, Griffin IH. Metal ion dependent inhibition of activated protein C by
u2-macrog!obulin and u 2-antiplasmin in human blood. Arteriosclerosis 1990; 10: 893a.
13. Marlar RA, Montgomery RR, Broekmans AW. Diagnosis and treatment of homozygous
protein C deficiency. I Pediat 1989; 114: 528-34.
14. Marlar RA, Neumann A. Neonatal purpura fulminans due to homozygous protein C or
protein S deficiencies. Sem. Thromb Hemostas 1990; 16: 299-309.
15. Griffin IH, Evatt B, Zimmerman TS, K.1eiss AI, Wideman C. Deficiency of protein C in
congenital thrombotic disease. I Clin Invest 1981; 68: 1370-3.
16. Broekmans AW, Veltkamp n, Bertina RM. Congenital protein C deficiency and venous
thromboembolism. A study of three Dutch families. N Engl J Med 1983; 309: 340-4.
17. Allaart CF, Poort SR, Rosendaal FR, Reitsma PH, Bertina RM, Briet E. Increased risk of venous
thrombosis in carriers of hereditary protein C deficiency defect. Lancet 1993; 341: 134-8.
18. Gladson CL, Scharrer I, Hack V, Beck KH, Griffin JH. The frequency of type I heterozygous
protein S and protein C deficiency in 141 unrelated young patients with venous thrombosis.
Thromb Haemostas 1988; 59: 18-22.
19. Bertina RM. Prevalence of hereditary thrombophilia and the identification of genetic risk
factors. Fibrinolysis 1988; 2(S2): 7-13.
20. Seligsohn U, Berger A, Abend M, Rubin I, Attias D, Zivelin A, Rapaport SI. Homozygous
protein C deficiency manifested by massive venous thrombosis in the newborn. N Eng! J Med
1984; 310: 559-62.
21. Miletich I, Sherman L, Broze G Ir. Absence of thrombosis in subjects with heterozygous
protein C deficiency. N Eng! I Med 1987; 317: 991-6.
22. Tait RC, Walker 10, Reitsma PH, Islam SIAM, McCan F, Poort SR, Conkie lA, Bertina
RM. Prevalence of protein C deficiency in the healthy population. Thromb Haemostas 1995;
73: 87-93.
23. Koeleman BPC, Reitsma PH, AHaart CF, Bertina RM. Activated protein C resistance as an
additional risk factor for thrombosis in protein C deficient families. Blood 1994; 84: 1031-5.
24. Bertina RM, Broekmans AW, van der Linden IK, Mertens K. Protein C deficiency in a Dutch
family with thrombotic disease. Thromb Haemostas 1982; 48: 1-5.
25. Pabinger I, Kyrle PA, Speiser W, Stoffels U, lung M, Lechner K. Diagnosis of protein C
deficiency in patients on oral anticoagulant treatment: comparison of three different functional
protein C assays. Thromb Haemostas 1990; 63: 407-12.
26. Cohen H, Mackie U, Walshe K, Gil1mer MD, Machin S1. A comparison of the effects of two
triphasic oral contraceptives on haernostasis. Br J Haernatol1988; 69: 259-63.
27. Gensini GF, Rostagno C, Abbat R, Favilla S, Mannucci PM, Semeri GGN, Bonomi AB.
Increased protein C and fibrinopeptide A concentration in patients with angina. Thromb Res
1988; 30: 517-25.
28. Cosio FG, Harker C, Batard MA, Brandt JT, GrifTm IH. Plasma concentrations of the natural
anticoagulants protein C and protein S in patients with proteinuria. J Lab Clin Med 1985;
106: 218-22.
29. Pabinger-Fasching I, Lechner K, Niessner H, Schmidt P, Ba1zar E, Mannhalter CL. High
levels of plasma protein C in nephrotic syndrome. Thromb Haemostas 1985: 53: 5-7.
30. Blarney SL, Lowe GDO, Bertina RM, K.1uft C, Sue Ling HM, Davies lA, Forbes CD. Protein
C antigen levels in major abdornina1 surgery: relationships of deep vein thrombosis, malignancy
and treatment with stanozolol. Thromb Haernostas 1985; 54: 622-5.
31. Vigano S, Mannueei PM, D'Angelo A, Gelfi C, Gensini GF, Rostagno C, Semeri GGN. The
significance of protein C antigen in acute and chronic liver biliary disease. Am I Clin Pathol
1984; 84: 454-8.
32. Marlar RA, Endres-Brooks I, Miller C. Serial studies of protein C and its plasma inhibitor
in patients with disseminated intravascular coagulation. Blood 1985; 66: 59-63.
33. Ceriello A. Quatrano A, Dello Russo P, Marchi E, Barbanti M, Rita Milani M, Giugliano
D. Protein C deficiency in insulin-dependent diabetes: a hyperg!ycemia-related phenomenon.
Thromb Haernostas 1990; 64: 104-7.
137
 u\BORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
34. Vukovich TC, Schernthaner G. Decreased protein C levels in patients with insulin-dependent
type I diabetes mellitus. Diabetes 1986; 35: 617-19.
35. Kloczko J, Wojtukiewicz M, Bielawiec M, Borowska M. Reduced protein C levels in patients
with essential hypertension. Thromb Haemostas 1987; 58: 793.
36. Green D, Scott JP. Is sickle cell crisis a thrombotic event? Am J Hematol1986; 23: 317-21.
37. Miletich JP. Laboratory diagnosis of protein C deficiency. Sem Thromb Haemostas 1990; 16:
169-76.
38. Gonzalez R, Alberca I, Sala N, Vicente V. Protein C deficiency - response to danazol and
DDAVP. Thromb Haemostas 1985; 53: 320--2.
39. Broekmans AW, Conard J, van Weijenberg RG, Horellou MH, Kluft C, Bertina RM. Treatment
of hereditary protein C deficiency with stanozolol. Thromb Haemostas 1987; 57: 20--4.
40. Conard J, Horellou MH, Dreden P, Potevin F, Zittoun R, Samama M. Decrease in protein
C in L-asparaginase-treated patients. Br J Haematol1985; 59: 725-41.
41. Vigano S, Mannucci PM, Solinas S, Botasso B, Mariani G. Decrease in protein C antigen and
formation of an abnormal protein soon after starting oral anticoagulant therapy. Br J Haematol
1984; 57: 213-20.
42. Marlar RA, Adcock DM. Clinical evaluation of protein C: a comparative review of antigenic
and functional assays. Progr Patho11989; 20: 1040--7.
43. Mikami S, Tuddenham EGD. Studies on immunological assay of vitamin K-dependent factors.
II. Comparison of four immunoassay methods with functional activity of protein C in human
plasma. Br J Haematol1986; 62: 183-93.
44. Boyer C, Rothschild C, Wolf M, Amiral J, Meyer D, Larrieu M1. A new method for the
estimation of protein C by ELISA. Thromb Res 1984; 36: 579-89.
45. Suzuki K, Moriguchi A, Nagayoshi A, Mutoh S, Katszuki S, Hashimoto S. Enzyme
immunoassay of human protein C by using monoclonal antibodies. Thromb Res 1985; 38:
611-21.
46. Ikeda K, Stenfio 1. A radioimmunoassay for protein C. Thromb Res 1985; 39: 297-306.
47. Epstein DJ, Bergum PW, Bajaj SP, Rapaport SI. Radioimmunoassay for protein C and factor
X. Plasma antigen levels in abnormal haemostatic states. Am J Clin Patho11984; 82: 573--81.
48. Bertina RM. Specificity of protein C and protein S assays. Res Clin Lab 1990; 20: 127-38.
49. Hubbard AR. Standardization of protein C in plasma: establishment of an International
Standard. Thromb Haemostas 1988; 59: 464-7.
50. Bertina RM. An international collaborative study on the performance of protein C antigen
assays. Thromb Haemostas 1987; 57: 112-17.
51. Mannucci PM, Boyer C, Tripodi A, Vigano d' Angelo S, Wolf M, Valsecchi C, d'Angelo A,
Meyer D and Larrieu MJ. Multicenter comparison of five functional and two immunological
assays for protein C. Thromb Haemostas 1987; 57: 44-8.
52. Francis RB, Patch MJ. A functional assay for protein C in human plasma. Thromb Res 1983;
32: 605-13.
53. Comp PC, Nixon RR, Esmon CT. Determination of functional levels of protein C, an
antithrombotic protein, using thrombin-thrombomodulin complex. Blood 1984; 63: 15-21.
54. Sala N, Owen NG, Collen D. Functional assay of protein C in human plasma. Blood 1984;
63: 671-5.
55. Bertina RM, Broekmans AW, Krommenhoek-van Es C, van Wijngaarden A. The use of a
functional and immunologic assay for plasma protein C in the study of the heterogeneity of
congenital protein C deficiency. Thromb Haemostas 1984; 51: 1-5.
56. Vigano-D'Angelo S, Comp PC, Esmon CT, D'Angelo A. Relationship between protein C
antigen and anticoagulant activity during oral anti-coagulation and in selected disease states,
J Clin Invest 1986; 77: 416-25.
57. Martinoli JL, Stocker K. Fast functional protein C assay using Protac, a novel protein C
activator. Thromb Res 1986; 43: 253-64.
58. Francis RB, Seyfert U. Rapid amidolytic assay of protein C in whole plasma using an activator
from the venom of Agkistrodon contortrix contortrix. Am J Clin Patho11987; 87: 619-25.
59. Vinazzer H, Pangraz U. Protein C: comparison of different assays in normal and abnormal
plasma samples. Thromb Res 1987; 46: 1-8.
60. Stocker K, Fischer H, ¥eier J, Brogli M, Svendsen L. Protein C activators in snake venoms.
Behring Institut Mitteilungen 1986; 79: 37-47.

138
 PROTEIN C ACTIVITY AND ANTIGEN
61. Exner T, Vaasjoki R. Characterization and some properties of the protein C activator from
Agkistrodon contortrix contortrix venom. Thromb Haemostas 1988; 59: 40-4.
62. Orthner CL, Bhattacharya P, Strickland DK. Characterization of a protein C activator from
the venom of Agkistrodon contortrix contortrix. Biochemistry 1988; 27: 2558-64.
63. Sturk A, Morrien-Salomons WM, Huisman MY, Borm JJJ, Biiller HR, ten Cate JW. Analytical
and clinical evaluation of commercial protein C assays. Clin Chim Acta 1987; 165: 263-70.
64. Franchi F, Tripodi A, Valsecchi C, Mannucci PM. Functional assays of protein C: comparison
of two snake-venom assays with two thrombin assays. Thromb Haemostas 1988; 60: 145--7.
65. Reitsma PH, Bernardi F, Doig RG, Gaudrille S, Greengard JS, Ireland H, Krawczak M, Lind
B, Long GL, Poort SR, Saito H, Sala N, Witt I, Cooper D. Protein C deficiency: a database
of mutations, 1995 update. Thromb Haemostas 1995; 73: 876-89.
66. Ireland H, Bayston T, Thompson E, Adami A, Gon~lves C, Lane DA, Finazzi G, Barbui T.
Apparent heterozygous type II protein C deficiency caused by the factor V 506 Arg to Gin
mutation. Thromb Haemostas 1995; 73: 731-2.
67. Girolami A, Simioni P, Tormene D, Vianello F. Pitfall of protein C resistance. Blood Coagul
Fibrinol 1996; 7:712-13.
68. Mannucci PM, Boyer C, Tripodi A, Vigano-D'Angelo S, Wolf M, Valsecchi C, D'Angelo A,
Meyer D, Larrieu MJ. Multicenter Comparison of five functional and two immunologic
assays for protein C. Thromb Haemostas 1987; 57: 44-8.

139
15
Protein S antigen
R. M. BERTINA

INTRODUCTION

Protein S (PS) is a vitamin K-dependent plasma glycoprotein (MW 70000)1,2.

Unlike the other vitamin K-dependent coagulation factors PS is not the zymogen
of a serine protease. Apart from a domain rich in y-carboxy glutamic acid
residues (Gla-domain), a thrombin-sensitive region (TSR), and four epidermal-
growth-factor like domains (EGF domains), PS contains a large carboxy
terminal domain homologous to the rat-androgen-binding protein (ABP) and
the human sex hormone-binding globulin (SHBGl Protein S is synthesized
in the liver, endothelial cells and megakaryocytes . The biological half-life
of PS is about 42.5 h7• In pooled normal plasma the total PS concentration is
about 25 Ilg/ml (0.35 jlffiol/L)8.

Physiological role
PS is the protein cofactor of the anticoagulant enzyme activated protein C
(APC)9 and therefore is an essential component of the PC anticoagulant
pathway lO,ll. It stimulates the inactivation of factor Va and factor VIlla by
APC both in plasma and purified systems I2- 14 . It is also required for the
expression of the fibrinolytic effect of APe in a whole-blood clot lysis systeml5 .
PS forms a I: I complex with APC on phospholipid surfaces and is thus thought
to stimulate the phospholipid-dependent inactivation of factors Va and VIlla.
The APC cofactor activity of PS is regulated by two independent processes.
First, PS is cleaved by thrombin in the TSR region; this cleavage results in the
formation of a molecule consisting of two chains linked by one disulphide
bridge, which no longer has APC cofactor activity I6,17. However, under physi-
ological conditions (Le. in the presence of Ca2 +) this reaction is rather slow
and therefore will require high local thrombin concentrations.
141
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
A second mechanism controlling the APC cofactor activity of PS is its ability
to form a 1:1 complex with the C4b binding protein (C4bBP), an important
regulatory component of the complement systemlO. The C4bBP PS complex
has no APC cofactor activityl8. However, this complex - which still can bind
to phospholipid membranes - may playa role in the regulation of complement
activation 19 . More recently evidence has been obtained that - at least in vitro
- both free and C4bBP-bound PS also have an anticoagulant activity which is
independent of APC20-22• The physiological relevance of this APC-independent
anticoagulant activity is still unknown.
Binding of PS to C4bBP occurs via a separate unique subunit of C4bBP,
the /3-chain23 • The Kd for binding of PS to C4bBP is very low (5 x 10-10 mollL)
in the presence of Ca2+ (see ref. 23). This might explain why in plasma almost
all /3-chain containing C4bBP is found to be bound to PS8. It is only the excess
PS that circulates as free PS8,25. Interestingly, the genes for the C4bBP u- and
/3-chains are under different regulation25 , In citrated human plasma or in serum,
free and C4bBP PS occur in a ratio of 2:3 8,10,24.

Pathophysiological aspects
Hereditary defects
The importance of PS for the control of thrombin formation in vivo is most
dramatically demonstrated in neonates who are homozygotes or double
heterozygotes for hereditary PS deficiency26,27. Such patients develop, shortly
after birth, clinical symptoms (among others, purpura fulminans syndrom~
that are very similar to those reported for homozygous protein C deficiencr .
Heterozygotes for hereditary PS deficiency have an increased risk for the
development of venous thrombosis at relatively young age29-31. More recent
studies indicate that PS deficiency might also be a weak risk factor for arterial
thrombosis 32 •
In plasma PS circulates both as free and C4bBP-complexed PS. This will
complicate the laboratory diagnosis of hereditary PS deficiencies. In fact three
types of deficiencies have been recognized33 • In type I PS deficiency there is a
reduction of total PS antigen till about 50% of normal23 ,30. Type I deficiencies
are caused by alterations in the PS gene that do not permit normal transcription,
translation and/or secretion of the protein34• In type II PS deficiency the total
PS antigen is normal while one of the PS genes makes a molecule with a
strongly reduced biological activity34. In the laboratory this deficiency can be
diagnosed by the reduced ratio PS activity/free PS antigen. Unfortunately most
of the reported cases of type II PS deficiency are patients with an apparent
reduced PS activity due to the presence of the factor V Leiden mutation (see
Chapter 16 on PS activity). The third type of hereditary PS deficiency is the
so-called type III deficiency. In patients with this type of deficiency the free PS
antigen is reduced while total PS is within the normal range. Recent studies
demonstrated that the phenotypes of both type I and ti~e III PS deficiency
can be found in carriers of the same PS gene defece ' 6. This was partly
explained by an age-dependent increase in total PS (mainly due to an increase
142
 PROTEIN S ANTIGEN
in C4bBP-bound PS). From these studies it is clear that free PS measurements
discriminate better between carriers and non-carriers of a PS gene defect than
total PS measurements.

Acquired abnormalities
Abnormal plasma levels of total and/or free PS have been reported for a number
of clinical conditions. Sometimes the interpretation of the reported data is
hampered by the lack of information on the specificity of the assay (especially
true for some of the total PS antigen assays).
Reduced levels of total PS have been reported for patients usin~ oral
anticoagulants7,24 during pregnancy and use of oral contraceptives37- 9 and
in patients with liver disease24• Conflicting data (reduced/increased PS antigen)
have been reported for patients with diabetes mellitus40,41. Increased PS antigen
(total) has been found in patients with nephrotic syndrome42 . The simultaneous
increase in C4bBP, however, might result in an acquired deficiency of free PS43.
No change in total PS antigen has been observed during disseminated intra-
vascular coagulation7,24.
Under conditions in which acute-phase reactions mal occur, elevated C4bBP
levels might cause a reduction in free PS concentration2 . This may occur during
an acute thrombotic episode, although elevated total and free PS antigen have
also been reported for acute deep vein thrombosis44 .

PROTEIN S ASSAYS

The development of suitable assays for the measurement of PS in plasma is

hampered by the fact that PS circulates in the plasma both as free PS and as
the C4bBP-PS complex 16. The distribution of PS over the free and complexed
form is dictated by the. total concentrations of PS and ~-chain containing
C4bBP and the apparent equilibrium constant8 ,25.
Basically three types of assays can be recognized:
1. immunological assays for total PS
2. immunological assays for free PS
3. PS functional assays (see Chapter 16).
Several immunological assays for the measurement of total PS inglasma have
been reported: Laurell-types of assay30, immunoradiometric assays ,45, enzyrne-
linked immunosorbent assays (ELISA)46-49 and radioimmunoassay (RIAio,51.
The selection of a suitable assay is made on the basis of sensitivity and specificity.
The demonstration that free and complexed PS are measured with the same
efficiency (i.e. the PS concentration is independent of the actual C4bBP
concentration) is an especially important criterion for selection of a total
PS-antigen assay. In some of the Laurell assays two different precipitation
peaks are observed for free and complexed PS, which makes it impossible to
read the total PS concentration. Also, when Laurell, ELISA or RIA systems
are used, it is important to check whether the further addition of C4bBP (in
143
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
sufficiently high concentrations) to a PS containing sample (plasma) will affect
the response in the assay system24,38,51.
In the following paragraphs the general procedure of an ELISA for total PS
will be described. It is important to note that conditions were selected (high
plasma dilutions and long incubation times) that favour complete dissociation
of C4bBP-PS complexes. The latter is important especially when unselected
polyclonal antibody populations are used, that may contain classes of antibodies
that differ in affinity for free and bound PS. This is not a problem in the
Asserachrom total PS test (Diagnostica Stago), in which two different
monoclonal antibodies are used that have the same affinity for free PS and
C4bBP-PS complexes.

Principle of the ELISA for total PS

Wells of microtitre plates are coated with rabbit or goat antibodies against
human PS (e.g. from Dako or Enzyme Research Laboratories). During the
first incubation these immobilized antibodies (catching antibodies) will react
with the PS molecules present in the sample (dilutions). Mter the establishment
of an equilibrium, the wells are emptied, washed and incubated with a suitable
dilution of anti-PS antibodies conjugated to horseradish peroxidase (HRP)
(tagging antibody). The amount of conjugate bound to the wells will be directly
related to the amount of immobilized PS and thus to the concentration of PS
in the sample (dilution). The amount of immobilized conjugate is measured
during a third incubation, via its reactivity towards a specific substrate (expressed
in d absorbance/min). In this way it is possible to plot the d absorbance/min
versus the total PS concentration (dilutions of pooled normal plasma), and to
use this calibration curve to read the concentration of total PS for the test
samples by intrapolation.

Materials
ELISA buffer
This buffer is used for making the dilutions of the samples, for washing the
micro titre plates, and as diluent for the anti-PS conjugate. It consists of:
67 mmollL Na2HP04' 2H20, 14 mmollL KH2P04, 0.1% (v/v) Tween 20 (PH
7.5). This solution is freshly prepared for each series of tests.

Coating buffer
This buffer is used for the coating of microtitre plates. It consists of 0.1 mollL
NaHC0 3, 0.5 mollL NaCI (PH 9.0).

Substrate buffer
This buffer is used to dissolve the HRP-substrate OPD (orthophenylene
diamine). It consists of 22.2 mmollL citric acid, 51.4 mmollL NaH2P04 . H 20
(pH 5.0) and can be stored for several weeks at 4 0c.
144
 PROTEIN S ANTIGEN
Substrate solution
20 mg orthophenylene diamine (dihydrochloride) (Sigma) is dissolved in 50 ml
substrate buffer (see above). About I min before addition to the microtitre
plates 20 ~ of a 30% H 20 2 solution is added. This volume of substrate solution
should be sufficient for four or five plates.

Stop reagent
4 N H 2S04 is used to stop the peroxidase-catalysed conversion of OPD.

Catching antibody
As catching antibody rabbit antibodies against human PS are used, routinely
these antibodies are obtained by isolating the non-Ca2+ -dependent fraction of
the anti-PS antibodies from the antiserum by immunoaffmity chromatography
on PS Sepharose (2 mg purified human PS/ml gel)24,46. Stock solutions (>
0.050 mg IgG/ml) are stored in I ml aliquots at -20°C.

Tagging antibody
The same non-Ca2 +-dependent anti-PS antibodies that are used as catching
antibodies are used as tagging antibodies. Horseradish peroxidase (HRP) is
coupled to the anti-PS IgG using the reagent N-succinimidyI3-(2-piridylthio)
propionate (SPDP) as described elsewhere52• The conjugate is stored in suitable
aliquots at -20°C in a buffer containing 0.3 mollL NaC!. Dilutions of the
conjugate (usually about 114000) are prepared about 30 min prior to use. It
should be noted that anti-PS HRP conjugate can be purchased from a number
of different manufacturers.

Titertek assay plates

These were purchased from Titertek (lmmuno-Assay plates). Other plates may
be used only after careful analysis of their suitability for this assay (homogeneity
of plates, coating performance, background absorbance).

Other materials
For the transfer of samples, conjugate, or substrate solution to the micro titre
plate multichannel pipettes are used. Absorbances are read on a multichannel
analyser (Titertek Multiscan Plus) using a filter of 492 nm.
145
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Test procedure

Coating of microtitre plates

This is performed by incubating the wells with 110 ~l/well of the catching
antibody (> 4 ~g/ml in coating buffer) at 4°C. Plates are sealed with paraftlm
and stored at 4 °C for periods up to 1 month. Minimal time required for coating
is overnight at 4°C.
Prior to the actual test, sufficient plates are removed from the stock. The
coating solution is carefully removed, after which the plates are repeatedly
washed with ELISA buffer (five times). Subsequently, wells are incubated with
150 ~l of a solution of 2% bovine serum albumin (Sigma) in ELISA buffer
(30 min at room temperature), and washed five times with ELISA buffer.

Test protocol
Suitable dilutions of test sample and standard (100 ~) are added to the wells
and incubated for 18 h (usually overnight) at room temperature to allow for
equilibrium distribution of PS over solid and fluid phase. Assay plates then are
washed five times with ELISA buffer before 100 ~lIwell of a dilution of
HRP-Iabelled anti-PS immunoglobulin is added (usually a 114000 dilution of
the stock in ELISA buffer). Plates are then incubated for about 45 min at room
temperature. Subsequently plates are emptied and washed five times with
ELISA buffer. Then 100 Ill/well of substrate solution (see above) is added.
Plates are incubated for about 30 min (until sufficient colour has been formed)
before the peroxidase activity is stopped by the addition of 100 ~l/well of 4 N
H 2 S04 , Finally the absorbance at 492 nm is read in a multichannel analyser
(see Other materials, above).

Calibration
The ELISA for total PS antigen is calibrated with dilutions of pooled normal
plasma (usually from 1/500 serially to 1132000). Routinely, samples from
patients not using oral anticoagulants are tested in 112000 and 1/4000 dilutions,
while plasmas of patients using oral anticoagulants are tested in 111000 and
112000 dilutions. Each dilution is tested in duplicate. All dilutions are made in
ELISA buffer. To calculate the amount of total PS antigen in plasmas, the
absorbance values were read from the standard curve which was obtained of
a least-square fit of third-order polynomial to the data plotted on a double-
logarithmic scale. Test results are valid when the SD is less than 10%.
Each laboratory relies on its own pooled plasma. Care should be taken that
sufficient individual donations have been included in the pool (plasmas of
females using oral contraceptives should not be included). Total PS antigen is
expressed as a percentage or as units/ml, where 1 unit reflects the amount of
PS present in 1 ml pooled normal plasma.
146
 PROTEIN S ANTIGEN

Reference ranges
Vsing the aforementioned ELISA, total PS antigen was measured in 40 healthy
volunteers; mean PS antigen was: 1.12 ± 0.16 unit/ml, while individual values
ranged from 0.77 to 1.44 unit/ml.
In 40 patients using oral anticoagulant treatment mean PS antigen was: 0.52
± 0.10 unit/ml, while individual values ranged from 0.34 to 0.69 units/ml. As
mentioned above, it is recommended that the laboratory diagnosis of PS is
based on the measurements of both total and free PS. It is recommended,
however, to use separate normal ranges for males and females 53 ,54; eventually
these can be extended with a normal range for oral contraceptive users. The
influence of age could also be taken into account35 . Calibration of local pooled
normal plasma can be made against the first international standard PS, plasma
(coded 93/590), which has a potency of 0.90 IV/ampoule for total PS, free PS
and PS activity 55.

Alternative methods for the measurement of total PS antigen

As mentioned above, total PS antigen can also be measured by immuno-
radiometric assay or radioimmunoassay. Moreover, several other ELISA
procedures have been described in the literature4 6-49, occasionally without
providing sufficient details on the specificity of the test (and sensitivity to
C4bBP levels).
Many commercial test kits or reagents are now available for the measurement
of total PS; two test kits are based on the use of two different monoclonal
antibodies in a sandwich ELISA: Asserachrom® total protein S from Diagnostica
Stago, which in fact is a one-step ELISA, and Thromboostika® protein S from
Organon Teknika, which is a classical two-step ELISA. Both tests are specific
for total PS. More recently Diagnostica Stago also developed a nephelometric
test for total PS (Liatest® protein S). Suitable polyclonal coating and detecting
antibodies can be obtained, among other sources, from Dako, Enzyme Research
Laboratories or American Diagnostica. When these reagents are used in an
ELISA for total PS it is important to check for the influence of addition of
C4b-binding protein to a normal plasma on the PS levels measured. When the
PS concentration decreases at increasing C4bBP concentration this indicates
that the system is relatively insensitive for C4bBP-bound PS (this is, for instance,
the case for the Asserachrom ®protein S of Diagnostica Stago). For the measure-
ment of total PS by the Laurell method, commercial reagents are also available
(e.g. Biopool protein S EID kit, Assera®-plate protein S from Diagnostica
Stago).
A general problem in using a Laurell assay for the measurement of total PS
is to devise conditions (antiserum, buffer system, temperature, electrophoresis
conditions) where plasma PS will give one single precipitation peak that indeed
reflects the concentration of total PS (i.e. it will not decrease after addition of
excess C4bBP to a test plasma). Presently there are no general guidelines avail-
able 56 .
147
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
MEASUREMENT OF FREE PS ANTIGEN

As discussed above, PS and C4bBP form a complex under physiological condi-

tions. In plasma (citrated, or ethylenediaminetetra-acetic acid - EDTA) about
40% of the total PS is free while the remaining 60% is bound to C4bBP. Because
the latter complex has no APC-cofactor activity, it is of interest to know what
part of the total PS antigen is actually free. Qualitatively this information can
be obtained by using two-dimensional immunoelectrophoresis: because free
PS moves faster in the electric field than the PS-C4bBP complexes, the
distribution of total PS over the free and bound form can be estimated from
the areas under the respective precipitation peakss7. More quantitative infor-
mation can be obtained by separating free PS from C4bBP-PS complexes by
precipitation with polyethylene glycoIS!,s7. Mter centrifugation the free PS is
still in the supernatant, while the complexes can be recovered from the precipitate.
Two different :Rrotocols are most widely used today: in the method described
by Comp et al. S7 PEG 8000 is used at a final concentration of 3.75%, while in
the MaIm protocolS! the final concentration of PEG 6000 is 5%. The
measurement of free PS antigen in a PEG supernatant sometimes creates a
problem when a PS antigen test is used that is not really specific for total PS
(e.g. non-parallel dilution curves in ELISA systems or Laurell systems). There-
fore most authors express the concentration of free PS antigen as a percentage
of the free PS antigen present in pooled normal plasma.
However, when using an ELISA for total PS antigen, as described above, it
is possible to express the concentration of free PS antigen directly in units/ml
by reading the absorbances on the calibration curve obtained with dilutions
of pooled normal plasma (completely parallel dilution curves). In this way we
found in a group of 40 healthy volunteers a mean free PS antigen of 0.42 ±
0.13 units/ml (range 0.23-0.78 units/mI), while in a group of 40 patients on
oral anticoagulant treatment mean free PS antigen was 0.14 ± 0.07 unitlml
(range: 0.06-0.34 units/ml).
Recently, Diagnostica Stago has developed a very elegant one-step ELISA
for the direct measurement of free PS in plasma. In this assay they use two
different monoclonal antibodies against PS that recognize free PS but not
C4bBP-bound PS58. Comparisons of results obtained with this test and with
the classical PEG rsrecipitation methods demonstrated good correlations between
the two methods 9,60. It is our own experience that only at extremely low free
PS levels (e.g. in PS-deficient individuals on oral anticoagulant treatment) the
direct ELISA gives higher results than the PEG precipitation method. The
most recent development in the area of free PS assays is the ligand sorbent
assay reported by Giri et al., who use the PS-binding capacity of the ligand
C4BP to capture free PS from plasma6 !. However, at present there is only
limited experience with this type of assay, although it seems to be a promising
approach.

References
1. Di Scipio RG, Hermodson MA, Yates SG, Davie EW. A comparison of human prothrombin,
factor IX, factor X and protein S. Biochemistry 1977; 16: 696-706.
148
 PROTEIN S ANTIGEN
2. Di Scipio RG, Davie EW Characterization of protein S, a y-carboxyglutarnic acid containing
protein from bovine and human plasma. Biochemistry 1979; 18: 899-904.
3. Lundwall A, Dackowski W, Cohen E, Schaffer M, Mahr A, Dahlbiick B, Stenfio 1. Isolation
and sequence of the c-DNA for human protein S, a regulator of blood coagulation. Proc Natl
Acad Sci USA 1987; 83: 6716-20.
4. Fair DS, Marlar RA. Biosynthesis and secretion of factor VII, protein C, protein S and the
protein C inhibitor from a human hepatoma cell-line. Blood 1986; 67: 64--70.
5. Stern D, Brett J, Harris K, Nawroh P. Participation of endothelial cells in the protein C-protein
S anticoagulant pathway: the synthesis and release of protein S. J Cell Bioi 1986; 102: 1971-8.
6. Ogura M, Tanabe N, Nishioka J, Suzuki K, Saito H. Biosynthesis and secretion of functional
protein S by a human megakaryoblastic ceU line (MEG-Ol). Blood 1987; 70: 301-6.
7. D'Angelo A, Vigano-D'Angelo S, Esmon CT, Comp PC. Acquired deficiencies of protein S.
J Clin Invest 1988; 81: 1445-54.
8. Griffin JH, Gruber A, Fernandez JA. Reevaluation of total, free and bound protein Sand
C4b-binding protein levels in plasma anticoagulated with citrate or hurudin. Blood 1992; 79:
3203--11.
9. Walker F1. The regulation of activated protein C by a new protein; a possible function for
bovine protein S. J BioI Chern 1980; 255: 5521-4.
10. Dahlbiick B, Stenfto 1. A natural anticoagulant pathway: protein C, S, C4b-binding protein
and thrombomodulin. In: Bloom AL, Forbes CD, Thomas DP, Tuddenham EGD, eds.
Haemostasis and thrombosis. Edinburgh: ChurchiU Livingstone, 1994; 671-98.
11. Esmon CT, Schwarz HP. An update on clinical and basic aspects of the protein C anticoagulant
pathway. Trends Cardiovasc Med 1995; 5: 141-8.
12. Walker F1. Regulation of activated protein C by protein S. The role of phospholipid in factor
Va inactivation. J BioI Chem 1981; 256: 11128-31.
13. Walker FJ, Chavin SI, Fay PJ. Inactivation of factor VIII by activated protein C and protein
S. Arch Biochem Biophys 1987; 252: 322-8.
14. Rosing J, Hoekema L, Nicolaes GAF, Christella M, Thomassen LGD, Hemker HC, Varadi
K, Schwarz HP, Tans G. Effects of protein S and factor Xa on peptide bond cleavages during
inactivation of factor Va and factor Va R506Q by activated protein C. J BioI Chem 1995; 270:
27852-8.
15. De Fouw NJ, Haverkate F, Bertina RM, Koopman J, van Wijngaarden A, van Hinsbergh
VWM. The cofactor role of protein S in the acceleration of whole blood clot lysis by activated
protein C in vitro. Blood 1986; 67: 1189-92.
16. Walker FJ. Regulation of vitamin K dependent protein S. Inactivation by thrombin. J BioI
Chem 1984; 259: 10335-9.
17. Sugo T, Dahlbiick B, Holmgren A, Stenfio 1. Calcium binding of bovine protein S. Effect of
thrombin cleavage and removal of the y-carboxyglutarnic acid-containing region. J Bioi Chem
1986;261:5116-20.
18. Dahlbiick B. Inhibition of protein Ca cofactor function of human and bovine protein. J Bioi
Chem 1986; 261: 12022-7.
19. Schwalbe R, Dahlbiick B, Hillarp A, Nelsestuen G. Assembly of protein S and C4b-binding
protein on membranes. J BioI Chem 1990; 265: 16074-81.
20. Heeb MJ, Mesters RM, Tans G, Rosing J, Griffin JH. Binding of protein S to factor Va
associated with inhibition of prothrombinase that is independent of activated protein C. J
Bioi Chern 1993; 268: 2872-7.
21. HackengTM, van't Veer C, Meijers JC, Bouma BN. Human protein S inhibits prothrombinase
complex activity on endothelial ceUs and platelets via direct interactions with factors Va and
Xa. J BioI Chem 1994; 269: 21051-8.
22. van Wijnen M, Stam JG, van't Veer C, Meijers JCM, Reitsma PH, Bertina RM, Bouma BN.
The interaction of protein S with the phospholipid surface is essential for the activated protein
C-independent activity of protein S. Thromb Haemostas 1996; 76: 397-403.
23. Hillarp A, Dahlbiick B. Novel subunit in C4b-binding protein required for protein S binding.
J BioI Chem 1988; 263: 12759--64.
24. Bertina RM, van Wijngaarden A, Reinalda-Poot J, Poort SR, Born VJJ. Determination of
plasma protein S - the protein cofactor of activated protein C. Thromb Haemostas 1985; 33:
268-72.
149
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
25. Garcia de Frutos P, AIim RIM, Hiirdig Y, ZOller B, Dahlbiick B. Differential regulation of
a- and fl-chains of C4b-binding protein during acute-phase response, resulting in stable plasma
levels of free anticoagulant protein S. Blood 1994; 84: 815-22.
26. Mahasandana C, Suvatte V, Chuansumrit A, Marlar RA, Manco Johnson MJ, Jacobson LJ,
Hathaway WE. Homozygous protein S deficiency in an infant with purpura fulrninans. J
Pediatr 1990; 117: 75{}-3.
27. Gomez E, Ledford MR, Pegelow CH, Reitsma PH, Bertina RM. Homozygous protein S
deficiency due to a one base pair deletion that leads to a stop codon in exon III of the protein
S gene. Thromb Haemostas 1994; 71: 72~.
28. Marlar RA, Neumann A. Neonatal purpura fulrninans due to homozygous protein C or
protein S deficiencies. Sem Thromb 1990; 16: 299-309.
29. Comp PC, Esmon CT. Recurrent venous thrombo-embolism in patients with a partial deficiency
of protein S. N Eng! J Med 1984; 311: 1525-8.
30. Schwarz HP, Fischer M, Hopmeyer P, Batard MA, Griffm JH. Plasma protein S deficiency
in familial thrombotic disease. Blood 1984; 64: 1297-300.
31. Engesser L, Broekmans AW, Briet E, Brommer EJP, Bertina RM. Hereditary protein S
deficiency: clinical manifestations. Ann Intern Med 1987; 106: 677-82.
32. Zoller B, Garcia de Frutos P, Dahlbiick B. Evaluation of the relationship between protein S
and C4b binding protein isoforms in hereditary protein S deficiency demonstrating type I and
type III deficiencies to be phenotypic variants of the same genetic disease. Blood 1995; 85:
3524-31.
33. Bertina RM. Hereditary deficiencies of protein C and protein S. In: Aznar J, Espana F, eds.
Protein C pathway. Iberica: Springer Verlag, 1991; 111-20.
34. Gandrille S, Borgel D, Ireland H, Lane DA, Simmonds R, Reitsma PH, Mannhalter C,
Pabinger I, Saito H, Suzuki K, Formstone C, Cooper DN, Espinosa Y, Sala N, Bernardi F,
AiachM. Protein S deficiency: a database of mutations. ThrombHaemostas 1997; 77: 1201-14.
35. Simmonds RE, Zoller B, Ireland H, Thompson E, Garcia de Frutos P, Dahlbiick B, Lane
DA. Genetic and phenotypic analysis of a large (122-member) protein S deficient kindred
provides an explanation for the familial coexistence of type I and type III plasma phenotypes.
Blood 1997; 89: 4364-70.
36. Zoller B, Garcia de Frutos P, Dahlbiick B. Evaluation of the relationship between protein S
and C4b-binding protein isoforms in hereditary protein S deficiency demonstrating type I and
type III deficiencies to be phenotypic variants of the same genetic disease. Blood 1995; 85:
3524-31.
37. Maim J, Laurell M, Dahlback B. Changes in the plasma levels of vitamin K-dependent proteins
C and S and of C4b-binding protein during pregnancy and oral contraception. Br J Haematol
1988; 68: 437--43.
38. Comp PC, Thurnau GR, Webb J, Esmon CT. Functional and immunologic protein S levels
are decreased during pregnancy. Blood 1986; 68: 881-5.
39. Boerger 1M, Morris PC, Thurnau CT, Esmon CT, Comp PC. Oral contraceptives and gender
affect protein S status. Blood 1987; 69: 692--4.
40. Takahashi H, Tatewaki H, Wada K, Shibatu A. Plasma protein S in disseminated intravascular
coagulation, liver disease, collagen disease, diabetes mellitus, and under oral anticoagulant
therapy. Clin Chim Acta 1989; 182: 195-208.
41. Schwarz HP, Schemthaner G, Griffm JH. Decreased plasma levels of protein S in well-controled
type I diabetes mellitus. Thromb Haemostas 1987; 57: 240.
42. Rostoker G, Goualt-Heilmann M, Levent M, Robeva R, Lang P, Lagrue G. High level of
protein C and protein S in nephrotic syndrome. Nephron 1987; 46: 22{}-1.
43. Vigano d' Angelo S, D'Angelo A, Kaufman CE Jr, Sholer C, Esmon CT, Comp PC. Protein
S deficiency occurs in the nephrotic syndrome. Ann Intern Med 1987; 107: 42-7.
44. Toulon P, Gandrille S, Vitoux IF, Fiessinger IN, Sultan Y, Aiach M. High total and free
protein S antigen in patients with acute deep vein thrombosis. Thromb Res 1990; 59: 213-17.
45. Poort SR, Deutz-Terlouw PP, van Wijngaarden A, Bertina RM. Immunoradiometric assay
for the calcium-stabilized conformation of human protein S. Thromb Haemostas 1987; 58:
998-1004.
46. Deutz-Terlouw PP, Ballering L, Van Wijngaarden A, Bertina RM. Two ELISA's for
measurement of protein S, and their use in the laboratory diagnosis of protein S deficiency.
Clin ChimActa 1989; 186: 321-34.

150
 PROTEIN S ANTIGEN
47. Woodhams BJ. The simultaneous measurement of total and free protein S by ELISA. Thromb
Res 1988;50:213-20.
48. Arniral J, Adam M, Plassant V, Minard F, Vissac AM. Immunoassays for the measurement
of protein S. In: Gaffney PJ, ed. Fibrinolysis, current prospects. London: John Libbey, 1988;
125-8.
49. Krachmalnicoff A, Tombesi S, Va!secchi C, Albertini A, Mannucci PM. A monoclonal
antibody to human protein S used as the capture antibody for measuring total protein S by
enzyme immunoassay. Clin Chem 1990; 36: 43--6.
50. Fair DS, Revak DJ. Quantitation of human protein S in the plasma of normal and warfarin
treated individuals by radioimmunoassay. Thromb Res 1984; 36: 527-35.
51. Maim J, Bernhager R, Holmberg L, Dahlbiick B. Plasma concentrations of C4b-binding
protein and vitamin K-dependent protein S in term and preterm infants: low levels of protein
S-C4b-binding complexes. Br J Haematol1988; 68: 445-9.
52. Carlsson J, Orevin H, Axen R. Protein thiolation and reversible protein-protein conjugation.
N-succinimidyI3-(2-pyridyldithio) propionate, a new hetero-bifunctional reagent. Biochem J
1978; 173: 723-37.
53. Faioni EM, Valsecchi C, Palla A, Taioli E, Razzari C, Mannucci PM. Free protein S deficiency
in a risk factor for venous thrombosis. Thromb Haemostas 1997; 78: 1343--6.
54. Gari M, Falkon L, UrrUtia T, Vallve C, Borrell M, Fontcuberta J. The influence of low protein
S levels in young women on the definition of the normal range. Thromb Res 1994; 73: 149-52.
55. Hubbard AR. Standardisation of protein S in plasma: calibration of the lst international
standard. Thromb Haemostas 1997; 78: 1237-41.
56. Edson JR, Vogt JM, Huesman DA. Laboratory diagnosis of inherited protein S deficiency.
Am J Clin Pathol 1990; 94: 176-86.
57. Comp PC, Doray D, Patton D, Esmon CT. An abnormal plasma distribution of protein S
occurs in functional protein S deficiency. Blood 1986; 67: 5~.
58. Amiral J, Grosley B, Boyer-Neumann C, Marfaing-Koka A, Peynaud-Debayle E, Wolf M,
Meijer D. New direct assay of free protein S antigen using two distinct monoclonal antibodies
specific for the free form. Blood Coag FibrinoI1994; 5: 179-86.
59. Wolf M, Boyer-Neumann C, Peynaud-Debayle E, Marfaing-Koka A, Amiral J, Meijer D.
Clinical applications of a direct assay of free protein S antigen using monoclonal antibodies.
A study of 59 cases. Blood Coag Fibrino11994; 5: 187-92.
60. Aillaud MF, Pouymayon K, Brunet D, Parrot G, Alessi MC, Arnira! J, Juhan-Vague I. New
direct assay of free protein S antigen applied to diagnosis of protein S deficiency. Thromb
Haemostas 1996; 75: 283-5.
61. Giri TK, Hillarp A, Hiirdig Y, Zoller B, Dahlbiick B. A new direct, fast and quantitative
enzyme-linked ligand absorbent assay for measurement of free protein S antigen. Thromb
Haemostas 1998; 79: 767-72.

151
16
Protein S activity
E. M. FAIONI

INTRODUCTION
The function of protein S
Protein S (PS) activity assays are based on the property of PS to act as a co-
factor to activated protein C (APC) in the proteolytic inactivation of factors
Va and VIlla (reviewed in ref. I). Evidence from in-vitro studies suggests that
PS expresses its cofactor activity by binding and localizing APC on cellular
membranes (platelets, endothelial cells, leukocytes) where the clotting process
is ongoing (Figure 16.1). Thrombin-inactivated PS, PS bound to C4b-binding
protein (C4b-BP) and non-carboxylated PS cannot function as a cofactor to
APe. In the first case, proteolysis by thrombin of residues in the thrombin-
sensitive region of PS leads to formation of a two-chain disulphide-linked PS
which does not bind phospholipids as efficiently as native PS; in the second
case, possibly due to the presence of the bulky C4b-BP, masking of the
APC-binding region may occur. Finally, when the vitamin K-dependent
carboxylation of glutamic acid residues in the amino terminal region of PS
(Gla-domain) is impaired (as, for example, during oral anticoagulant therapy),
the resulting partially carboxylated molecule loses calcium-mediated Gla-
dependent binding capacity to negatively charged phospholipids2 •
In healthy individuals approximately 40% of circulating PS is free, while the
remainder is bound to C4b-BPI. The stoichiometry of the C4b-BP-PS interaction
is 1:1 1• C4b-BP is a spider-like molecule containing two types of polypeptide
chains, and only the ~ chain, not the a chain, is capable of binding PS (Figure
16.1)1. For a long time the consensus was that increased concentrations of
C4b-BP, which may be found during acute-phase reactions, would shift the
equilibrium between the bound and the free form of PS towards the former,
thus decreasing the anticoagulant potential of the protein C pathway. Recently
it has been observed that differential regulation of (l and ~ chain synthesis
153
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

\ ~P."-PS
~Gla-domajn
8,
~
~
~
@.a
~
Villa
~f:
APC
ca2+
I - - - thrombin

cel/ membrane
Figure 16.1 Schematic representation of APC cofactor function of protein S. Protein S (PS)
anchors activated protein C (APC) to negatively charged membrane surfaces, thus localizing its
proteolytic inhibitory activity to Va and VIlla. PS is in dynamic equilibrium with C4b-binding
protein (C4b-BP) which binds PS through its fI subunit. The C4b-BP-PS complex does not function
as APC cofactor. Thrombin can inactivate PS, and non-carboxylated forms of PS lose high-affinity
interaction with membranes. Though the binding regions of PS for C4b-BP have in fact been
attributed to the amino-terminal portion of PS (see ref. 1), the figure does not aim to represent
precise intermolecular connections.

occurs so that during acute inflammation mainly the a chain synthesis is

increased, leading to stable free PS levels3• The role of complexed PS is unclear.
Possibly PS may anchor the complement inhibitor to membranes where
complement reactions occurl . Besides its participation in complement regulation,
PS has a role in promoting fibrinolysis and perhaps in cellular differentia-
tion4 ,s. Thus, when we measure the cofactor activity to APe, we are in fact
exploring only a part of its functions, the integration and the regulation of
which are largely unknown.

Protein S activity assays

Measurement of PS activity as defined above should be the functional equivalent
of measuring the concentration of free PS. When both tests are performed,
three patterns of results can be defined: (a) normal levels of free PS concentration
154
 PROTEIN S ACTIVITY ASSAYS
and activity; (b) reduced free PS concentration and activity (type I deficiency
according to the proposal of the Scientific and Standardization Committee
(SSC) of the International Society on Thrombosis and Haemostasis (ISTH),
Munich, 1992; recently also the subtype with normal total PS and reduced free
PS has been suggested to be a type I deficiency6.7); or (c) normal free PS
concentration and reduced activity (type II deficiency, dysfunctional PS
according to the proposal of the SSC). In current laboratory practice, however,
matters are not as simple and clear-cut. Most of the proposed activity assays
do not measure PS activity after separation of free from complexed PS, which
is at variance with the assays that measure the concentration of free PS. This
generates a confounding factor: not exactly the same thing is being measured
in functional and immunological assays that measure free PS, and this com-
plicates the interpretation of results, especially in borderline cases.
In PS activity assays the concentration of procoagulant proteins and APC
is set so that the APC-related prolongation of the clotting time is dependent
on the PS concentration. In most proposed methods the source of PS is patient
plasma, at different dilutions. The use of plasma-based assays is theoretically
associated with the possibility of interference in the assay by substances present
in plasma. For plasma-based PS activity assays interference, in fact, has been
shown for anti-phospholipid antibodies and resistance to APC associated with
factor V Leiden. In the first case anti-phospholipid antibodies interfere aspecifi-
cally with the clotting assay, either by shortening or by prolonging the clotting
times, or specifically with PS function, in both cases resulting in falsely decreased
or increased PS activity levels 8,9. In the case of resistance to APC, at low
dilutions, spuriously low levels of PS are measured 10-13 .
Something in the diagnosis of PS deficiency is faulty, as indicated also by
the observation that only in aPfroximately half of the phenotypic diagnoses
can a genetic defect be detected 1 • Though alternative explanations are possible,
the most likely one is that this discrepancy is due, at least in part, to the low
specificity of PS assays. Even though at present an easy-to-perform, inexpensive,
sensitive and specific PS activity method is lacking, and most of those in use
are subject to false-positive results, PS activity assays cannot be totally discarded,
the reason for this being that type II deficiencies would otherwise go undetected.
The prevalence of PS type II deficiency is unknown. In a recently published
database of mutations, eight of 126 (approximately 6%) mutations were clearly
associated with a type II phenotype l . It is difficult, however, to estimate the
prevalence of type II PS deficiency correctly since, due to the above-mentioned
pitfalls, screening for PS deficiency by activity assays is probably not a widespread
procedure. A recent article based upon the World Health Organization report
of a joint WHOIISTH meeting (Geneva, Switzerland, 6-8 November 1995)
suggested that screening for PS deficiency should be performed by free PS
assaysl6. When should a PS activity assay be performed then, and why? The
answer is not, as is often the case, univocal, and it relies a lot on the setting in
which PS assays are performed. If two methods can be afforded, an activity
assay should be performed together with an assay to measure free PS con-
centration when screening thrombophilic patients. When a type II pattern of
results is discovered, acquired causes of functional PS deficiency (see below)
or interferences in plasma-based assays should be ruled out. When only one
155
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
method can be routinely adopted, if activity assays are chosen, awareness of
pitfalls is mandatory in the interpretation of pathological results and in-depth
diagnostic testing in a specialized centre should follow whenever possible. If
an antigenic test is chosen as the only screening procedure, one must be aware
that a number, as yet undefinable, of type II deficiencies will go undetected.

METHOD FOR DETERMINING PS ACTIVITY

The method that is detailed below is a prothrombin time-based PS activity

assay first described in 1990 by Preda et al. 17. Its choice is not motivated by it
being a better test than others; in fact is has the same drawbacks as other
plasma-based PS activity assays as discussed above and therefore it affords no
particular advantage over other commercially available methods. It was chosen
because it is easy to perform, has low coefficients of variation (reported as less
than 5-8% between assays17-19), can be automated and is commercially available.
While its application in detecting type I deficiencies is well established,
misdiagnosis of type II PS deficiency has occurred 10 ,19,20. Other available
commercial kits are listed below in the section 'Alternative methods'.

Reagents and instrumentation

IL Test@> PS assay kit (Instrumentation Laboratory); distilled water; diluent
(saline solution) can be purchased separately (lL Test@> Factor diluent); ACL@>
coagulometer (Automated Coagulation Laboratory, a nephelometric centrifugal
analyser) (Instrumentation Laboratory).
For plasma preparation, blood is drawn from the antecubital vein without
stasis, rapidly mixed with trisodium citrate (nine parts blood and one part
129 mmoVL citrate) and centrifuged within 2 h at 3000g and 4 °C for 20 min.
Plasma is separated and subdivided into aliquots which are snap-frozen (ethanol
and dry ice or liquid nitrogen) and stored at -80°C for up to 1 year. Testing
of PS activity is carried out on rapidly thawed samples (at 37°C in a water
bath) which may be thawed only once (pS activity is rapidly lost upon repeated
freezing and thawing).
The assay kit does not include a reference plasma, which may be purchased
separately (reference plasma, which is calibrated against the international
standard of PS 21 ). The kit includes a PS control plasma with low PS activity.
Both, however, can be prepared in the laboratory, if needed, as briefly described
below.

Preparation of reference plasma

Since an international standard for PS is available, the reference plasma does
not need to be a pool of plasmas, but it can also be a single healthy-donor
plasma. There are two important points in preparing the reference plasma: it
needs to be calibrated against the standard and it must be stored at -80°C in
aliquots.
156
 PROTEIN S ACTIVITY ASSAYS
Preparation of control plasma
Control plasma can be prepared by pooling plasma from patients on stabilized
oral anticoagulant therapy or with chronic liver disease. Generally this type of
control plasma has approximately 50% or lower of normal PS activity values.
Care should be taken to store the plasma as recommended for the reference
plasma.

Principle of the assay

APC cofactor activity of PS is measured as the degree of prolongation of a
prothrombin time-based clotting assay in which diluted test plasma, PS-depleted
plasma containing Protac®-activated protein C, bovine brain tissue thrombo-
plastin factor and calcium ions are mixed. The prolongation is proportional
to the PS content of the test plasma.

Procedure
1. Reconstitute Protac® (protein C activator) with 0.75 ml distilled water/vial.
2. Reconstitute PS-deficient plasma with 1.0 ml distilled water/vial.
3. Reconstitute PS control plasma with 1.0 ml distilled water/vial.
4. Reconstitute bovine tissue thromboplastin factor with 3.0 ml distilled water/
vial.
Note: all reagents can be aliquoted, frozen and stored at -20°C once they have
been reconstituted. See insert accompanying the kit for duration and condi-
tions of storage.
Patient samples, control plasma and PS-deficient plasma must be diluted
1: lOin factor diluent (saline). Protein C in PS-deficient plasma must be activated
by mixing one part of Protac® and two parts of PS-deficient plasma. Incubation
is carried out at 37°C for 20 min. This is the substrate plasma that the ACL
adds to the diluted plasma samples together with bovine thromboplastin and
calcium to obtain clotting. The ACL generates a reference curve made up of
four points (0%, 50%, 75%, 100%) in the older models, and two points (0%,
100%) in the newer ACL 6000 model. In each run a maximum of 13 samples
(at one dilution), including the control plasma, can be analysed.
For instructions on where to place the reagents and the plasma samples
correctly in the rotor and in the coagulometer, see accompanying insert in the
kit.
In the new ACL 6000, a dedicated PS activity program exists which is
user-friendly, prepares the plasma dilutions, and gives percentage PS activity
per sample at the end of the assay. In older models, sample dilutions must be
made by the operator and the factor II program has to be used. Since this is a
procoagulant factor program it cannot correctly construct the reference curve
and calculate the sample values (prolongation of clotting time is obviously
viewed by this program as a 'lack' of factor). Hence, clotting times in seconds
157
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
have to be obtained from the ACL video screen (they are not reported on the
printout) and activity values intrapolated from the calibration curve using the
following linear function:
y =fix); where y =activity (%) and x =clotting time (s).

Notes
We suggest that plasma samples of unknown PS activity be run at two dilutions
instead of one. The 1:20 dilution (and higher) should be made independently
(Le. not by further diluting 1:2 the original 1: 10 dilution, but by making the
1:20 dilution directly) in PS-deficient plasma diluted 1: 10 (experience has shown
this is the best way to achieve sample curves parallel to the reference curve).
The reason for adopting additional dilutions is that dilution mistakes are
avoided or reduced by this practice, and that suspicion of artifacts which alter
PS activity is generated by the finding of non-parallel curves (this practice, for
example, allowed us to conclude that resistance to APC interfered with the PS
activity assay). When this happens the sample should be tested again at three
or four dilutions.
Though the manufacturer suggests repeating the assay when a correlation
coefficient < 0.98 for the reference curve is found, a better practice, especially
when setting up the assay ex novo, is to include a standard curve among the
samples. This is to say that the reference plasma should be diluted appropriately
and run as a sample, in order to verify two things:
1. that the clotting times corresponding to points in the curve generated by
the machine are the same as those generated by the added reference plasma
dilutions;
2. that the reference curve is calculated on a sufficient number of points (0%,
50%,75%,100%) in the newer models.
If machine- and hand-generated curves are similar after a few trial runs, the
curve generated by the machine may be used. A control run as described should,
however, be performed on a regular basis.

Calculation of the reference range

The normal range should be calculated in each laboratory by determining PS
activity in at least 100 healthy men and women who do not have a history of
thrombosis. The distribution of PS values should be analysed and normalized
by appropriate mathematical transformation (for example, log transforma-
tion). Then the mean ± 1.96 SD can be used as the reference range or, alter-
natively, the 5th to 95th percentile. It must be noted that different ranges for
men and women have been suggested for free PS ~ntigen), and recently an
increase of total PS (antigen) by age was reported7,2 ,23. Consequently, in each
laboratory, when determining the normal range, differences by gender and/or
age or other factors should be checked. In our laboratory different ranges for
158
 PROTEIN S ACTIVITY ASSAYS

men and women have been adopted both for antigen and activity of free PS.
At present in our laboratory the reference range for men is 76-166% and for
women is 58-143%. Obviously, carriers of factor V Leiden were excluded from
the calculation.

Pitfalls
Besides the already-mentioned interference of resistance to APC and anti-
phospholipid antibodies in the PS activity assay, elevated factor VIla levels
can also shorten the prothrombin-time-based clotting assay leading to
underestimation of PS activity 17. It is often suggested, in case of interference,
to perform the assay at higher dilutions and to consider as valid the result
obtained at the highest dilution (in the belief that the interfering factor is
'diluted out'). We find this is not good laboratory practice. It is difficult to
know a priori the potency of the interfering factor, and consequently if and
when it will be diluted out. In addition, non-parallel curves may theoretically
be found in the presence of interferences by factors as yet unknown (as was
the case for interference by resistance to APC before the latter was described).
In all these cases the best procedure is simply not to measure PS activity by
plasma-based methods.
Low levels of PS activity may be an inherited or an acquired trait. Acquired
PS activity deficiency is usually associated with low free PS levels as in liver
disease, oral anticoagulant treatment, pregnancy, hormone replacement theragr'
the nephrotic syndrome, cancer and post-infectious anti-PS antibodies',24-= .
In summary, a diagnosis of inherited PS deficiency should never be based
on an activity test performed only once at a single dilution. A pathological
result should be confirmed at multiple dilutions after ruling out known interfer-
ences and acquired deficiencies. If measurement of free PS is not performed
in parallel, or in any case when a dysfunctional defect is suspected, genetic
analysis is recommended.

Alternative methods
Several methods for measuring PS activity have been reported. All are based
on the measurement of the cofactor activity of PS to APC 28-35. What changes
from one method to the other is the type of clotting assay used. Two are
commercialized, by Behring (Protein S Reagent, Dade Behring) and by
Diagnostica Stago (Staclot protein S). The method described by D'Angelo et
al. 2 is the only one that requires extraction of PS from plasma by anti-PS
monoclonal antibodies linked to beads; after elution the PS activity is determined
in a Xa-based clotting assay and its concentration is measured by ELISA. This
method has several advantages over others: first, it is not sensitive to interference
by substances present in plasma; second, a real specific activity of PS can be
measured, allowing the correct detection of type II deficiencies. However, the
drawback of this method is the requirement of the monoclonal antibody and
the difficulty in automation.
159
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
COMMENTS

Besides its cofactor function to APC, PS has an APC-independent anticoagulant

function: it has been reported that PS can directly inhibit the activity of the
tenase and prothrombinase complexes36--39. Recently, a new activity assay based
on this characteristic of PS was described40 . Though preliminary results are
interesting, the role of the APC-independent PS anticoagulant activity in
thrombophilia needs to be evaluated prospectively, as do the sensitivity and
specificity of this new method.

References
1. Dahlbiick B. The protein C anticoagulant system: inherited defects as basis for venous
thrombosis. Thromb Res 1995; 77: 1-43.
2. D'Angelo A, Vigano-D'Angelo S, Esmon CT, Comp Pc. Acquired deficiency of protein S.
Protein S activity during oral anticoagulation, in liver disease, and in disseminated intravascular
coagulation. J Clin Invest 1988; 81: 1445-54.
3. Garcia de Frutos P, Alim RIM, Hiirdig Y, ZOller B, Dahlbiick B. Differential regulation of a
and fl chains of C4b-binding protein during acute-phase response resulting in stable plasma
levels of free anticoagulant protein S. Blood 1994; 84: 815-22.
4. de Fouw NJ, Haverkate F, Bertina RM, Koopman J, van Wijngaarden A, van Hinsbergh
VWM. The cofactor role of PS in the acceleration of whole blood clot lysis by activated
protein C in vivo. Blood 1986; 67: 1189-92.
5. Godowski PJ, Mark MR, Chen J, Sadick MD, Raab H, Hammonds RG. Reevaluation of the
roles of protein S and Gas 6 as ligands for the receptor tyrosine kinase Rserryro 3. Cell 1995;
82: 355-8.
6. Zoller B, Garcia de Frutos P, Dahlbiick B. Evaluation of the relationship between protein S
and C4b-binding protein isoforms in hereditary protein S deficiency demonstrating type I and
type III deficiencies to be phenotypic variants of the same genetic disease. Blood 1995; 85:
3524-31.
7. Simmonds RE, Zoller B, Ireland H, Thompson E, de Frutos PG, Dahlbiick B, Lane DA.
Genetic and phenotypic analysis of a large (122) member protein S-deficient kindred provides
an explanation for the familial coexistence of type I and type III plasma phenotypes. Blood
1997;89:4364-70.
8. Rossi E, Gatti L, Guarneri D, Finotto E, Lombardi A, Preda L. Functional protein S in
women with lupus anticoagulant inhibitor. Thromb Res 1992; 65: 253-62.
9. Lawrie AS, Lloyd ME, Mohamed F, Irons S, Hughes GR, Savidge GF. Assay of protein S in
systemic lupus erythematosus. Blood Coag Fibrinol1995; 6: 322-4.
10. Faioni EM, Franchi F, Asti D, Sacchi E, Bernardi F, Mannucci PM. Resistance to activated
protein C in nine thrombophilic families: interference in a protein S functional assay. Thromb
Haemostas 1993; 70: 1067-71.
II. Faioni EM, Boyer-Neumann C, Franchi F, Wolf M, Meyer D, Mannucci PM. Another protein
S functional assay is sensitive to resistance to activated protein C. Thromb Haemostas 1994;
72: 648.
12. Cooper PC, Hampton KK, Makris M, Abuzenadah A, Paul B, Preston FE. Further evidence
that activated protein C resistance can be misdiagnosed as inherited functional protein S
deficiency. Br J HaematoI1994; 88: 201-3.
13. Simioni P, Gavasso S, Luni S, Invidiato S, Girolami A. A protein S functional assay yields
unsatisfactory results in patients with activated protein C resistance. Blood Coag Fibrinol
1995; 6: 286-7.
14. Lane DA, Mannucci PM, Bauer KA, Bertina RM, Bochkov NP, Boulyjenkov V, Chandy M,
Dahlbiick B, Ginter EK, Miletich JP, Rosendaal FR, Seligsohn U. Inherited thrombophilia:
Part 1. Thromb Haemostas 1996; 76: 651-62.
IS. Gandrille S, Borgel D, Ireland H, Lane DA, Simmonds R, Reitsma PH, Mannhalter C,
Pabinger I, Saito H, Suzuki K., Formstone C, Cooper DN, Espinosa Y, Sala N, Bernardi F,

160
 PROTEIN S ACTIVITY ASSAYS
Aiach M. Protein S deficiency: a database of mutations. For the Plasma Coagulation
Inhibitors Subcommittee of the Scientific and Standardization Committee of the Inter-
national Society on Thrombosis and Haemostasis. Thromb Haemostas 1997; 77: 1201-14.
16. Lane DA, Mannucci PM, Bauer KA, Bertina RM, Bochkov NP, Boulyjenkov V, Chandy M,
Dahlbiick B, Ginter EK, Miletich JP, Rosendaal FR, Seligsohn U. Inherited thrombophilia:
Part 2. Thromb Haemostas 1996; 76: 824-34.
17. Preda L, Tripodi A, Valsecchi C, Lombardi A, Finotto E, Mannucci PM. A prothrombin
time-based functional assay of protein S. Thromb Res 1990; 60: 19-32.
18. Maccaferri M, Legnani C, Preda L, Palareti G. Protein S activity in patients with heredofamilial
protein S deficiency and in patients with juvenile venous thrombosis. Results of a functional
method. Thromb Res 1991; 64: 647-58.
19. Boyer-Neumann C, Bertina RM, Tripodi A, D'Angelo A, Wolf M, Vigano D'Angelo S,
Mannucci PM, Meyer D, Larrieu Ml Comparison of functional assays for protein S: European
collaborative study of patients with congenital and acquired deficiency. Thromb Haemostas
1993; 70: 946-50.
20. Mannucci PM, Valsecchi C, Krachmalnicoff A, Faioni EM, Tripodi A. Familial dysfunction
of protein S. Thromb Haemostas 1989; 62: 763-6.
21. Hubbard AR. Standardisation of protein S in plasma: calibration of the 1st International
Standard. Thromb Haemostas 1997; 78: 1237-41.
22. Tait RC, Wright EM, Walker ID. Influence of age and sex on the diagnosis of protein S
deficiency. Thromb Haemostas 1997; 78 (Suppl.): 161 (abstract no. PD-659).
23. Faioni EM, Valsecchi C, Palla A, Taioli E, Razzari C, Mannucci PM. Free protein S deficiency
is a risk factor for venous thrombosis. Thromb Haemostas 1997; 78: 1343-6.
24. Comp PC, Thurnau GR, Welsh J, Esmon CT. Functional and immunologic protein S levels
are decreased during pregnancy. Blood 1986; 68: 881-5.
25. Vigano-D'Angelo S, D'Angelo A, Kaufman MJ, Sholer C, Esmon CT, Comp PC. Protein S
deficiency occurs in the nephrotic syndrome. Ann Intern Med 1987; 107: 42-7.
26. MaIm J, Laurell M, Dahlbiick B. Changes in the plasma levels of vitamin K-dependent proteins
C and S and of C4b-binding protein during pregnancy and oral contraception. Br J Haematol
1988; 68: 437-43.
27. D'Angelo A, Mazzola G, Bergmann F, Safa 0, Della Valle P, Vigano D'Angelo S. Autoimmune
protein S deficiency: a disorder predisposing to thrombosis. Haematologica 1995; 80 (Suppl.):
114-21.
28. Comp PC, Doray D, Patton D, Esmon CT. An abnormal plasma distribution of protein S
occurs in functional protein S deficiency. Blood 1986; 67: 504-8.
29. Kamiya T, Sugihara T, Ogata K, Saito H, Suzuki K, Nishioka J, Hashimoto S, Yamagata K.
Inherited deficiency of protein S in a Japanese family with recurrent venous thrombosis: a
study of three generations. Blood 1986; 67: 406-10.
30. van de Waart P, Preissner KT, Bechtold JR, Miiller-Berghaus G. A functional test for protein
S activity in plasma. Thromb Res 1987; 48: 427-37.
31. Suzuki K, Nishioka 1 Plasma protein S activity measured using Protac, a snake venom derived
activator of protein C. Thromb Res 1988; 48: 241-51.
32. Schwarz HP, Muntean W, Watzke H, Richter B, Griffin JH. Low total protein S antigen but
high protein S activity due to decreased C4b-binding protein in neonates. Blood 1988; 71:
562-5.
33. Wolf M, Boyer-Neumann C, Martinoli JL, Leroy-Matheron C, Amiral J, Meyer D, Larrieu
MI. A new functional assay for human protein S activity using activated factor V as substrate.
Thromb Haemostas 1989; 62: 1144-5.
34. Kobayashi I, Anemiya N, Endo T, Okuyama K, Tamura K, Kume S. Functional activity of
protein S determined with use of protein C activated by venom activator. Clin Chem 1989;
35: 1644-8.
35. Wiesel M-L, Charmantier J-L, Freyssinet J-M, Grunebaum L, Schuhler S, Cazenave J-P.
Screening of protein S deficiency using a functional assay in patients with venous and arterial
thrombosis. Thromb Res 1990; 58: 461-8.
36. Heeb MJ, Rosing J, Bakker HM, Fernandez JA, Tans G, GrifTm JH. Protein S binds to and
inhibits factor Xa. Proc Nat! Acad Sci USA 1994; 91: 2728-32.

161
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
37. Heeb MJ, Mesters RM, Tans G, Rosing J, Griffin JH. Binding of protein S to factor Va
associated with inhibition of prothrombinase that is independent of activated protein C. J
Bioi Chern 1993; 268: 2872-7.
38. Hackeng TM, van't Veer C, Meijers JCM, Bouma BN. Human protein S inhibits prothrombinase
complex activity on endothelial cells via a direct interaction of protein S with factor Va and
Xa. Evidence for an activated protein C independent anticoagulant function of protein S in
plasma. J Bioi Chern 1994; 269: 21051-8.
39. Koppelman SJ, Hackeng TM, Sixma JJ, Bouma BN. Inhibition of the intrinsic factor X
activating complex by protein S: evidence for a specific binding of protein S to factor VIII.
Blood 1995; 86: 1062-71.
40. van Wijnen M, van't Veer C, Meijers JCM, Bertina RM, Bouma BN. A plasma coagulation
assay to determine the activated protein C independent anticoagulant activity of protein S.
Thromb Haemostas 1997; 78 (Suppl.): 187 (abstr. no. OC-756).

162
17
Activated protein C (APC) resistance
A. TRIPODI

INTRODUCTION

Activated protein C (APe) resistance, defined as a poor anticoagulant response

of plasma to APC l , was first measured with a modified activated partial throm-
boplastin time (APTI) test. This consists of measuring paired clotting times
for test plasmas with and without added APe. The extent of prolongation of
the clotting time after APC addition over the clotting time before APC addition
can be taken as an index of in-vitro APC resistance l . Over the past few years
new methods have been developed which are based on the same basic principle,
but exploit other tests such as the Russell viper venom (RVV) time 2 , the
prothrombin time (PT)3 and the factor Xa-based (FXa)4. All the above are
global clotting tests which explore the coagulation cascade upstream from
factor V (FV). Therefore they are, at least in principle, suitable for detection
of the FV:Q506 mutationS, and possibly of the specific FV haplotype HR26 ,
which are the most frequent genetic abnormalities associated with APC resistance
so far identified. Another method based on the extent of the inactivation of
factor VIII (FVIII) measured with chromogenic substrate has been proposed7 •
More recently an additional method has been described and is based on the
measurement of the effect of APC on thrombin generation in rlasma upon
addition of intrinsic or extrinsic activators of blood coagulation . A common
feature of all these methods is their variable sensitivity and specificity with
respect to the detection of carriers of the FV:Q506 mutation identified by
DNA analysisS, and for some of them it is difficult to interpret results in patients
whose basal plasma clotting times are prolonged because of anticoagulant
treatment, clotting factor deficiencies9 and the presence of antiphospholipid
antibodieslO. Dilution of test plasmas in FV-deficient plasma prior to analysis
has been proposed to improve the specificity of the APIT ll ,12 and other methods4
in the above situations. This modification was proven to be effective for patients
on oral anticoagulants ll ,l2, but diagnosis in patients with antiphospholipid
163
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

antibodies still remains a problem, although some methods have been reported
to prevent interference ,13. An additional advantage of using the above
modification was shown in a recent large collaborative study designed to compare
the diagnostic efficacy of 13 plasma-based clotting methods for their abilit.(
to detect the FV:Q506 mutation in comparison with that of the DNA analysis I .
The results showed that predilution of test plasmas with FV-deficient plasma
considerably improved the ability of APTT-based methods to discriminate
carriers from non-carriers of the mutation l4 . However, to make a decision
about how to organize a diagnostic work-up in the thrombophilia laboratory,
one must also take into consideration that plasma-based methods (namely,
APTT-based methods) when performed on undiluted test plasmas are, at least
in principle, able to detect acquired APC resistance (Le. not due to the FV:Q506
mutation), whereas predilution of test plasma in FV-deficient plasma and DNA
analysis are not. Whether acquired APC resistance, which has been reported
to be fre~uent in pregnancyl5, oral contraceptive intakeS,I5--17, increased FVIII
activityl ,19 and cerebrovascular diseases20 is clinically important, and therefore
deserves to be detected, is still a matter of debate. Recently, van der Born et
al. 2o have shown in a population-based case--control study that a reduced
response to APC, independent of the FV:Q506 mutation, is associated with an
increased risk of cerebrovascular disease (the lower the APC response, the
higher the odds ratio for cerebrovascular disease). Should this finding be
confirmed in subsequent studies and/or extended to venous thromboembolism,
graded APC responses measured by APTT-based methods on undiluted plasma
could be useful to assess more accurately the thrombotic risk in thrombophilic
patients. On the basis of the above considerations, the following sections of
this chapter will be mainly devoted to the description of the APTT-based
method. Additional information for setting up other methods for APC resistance
and for DNA analysis for the FV:Q506 mutation can be found in the refer-
ences cited.

APTT·BASED METHOD

This method measures the in-vitro anticoagulant response of test plasma to APC.

Principle
Test plasmas are incubated with a suitable preparation of commercial or
home-made APTT reagent (activator plus phospholipids) for a standard period
of time (see below) to activate the contact phase. Coagulation is then initiated
by the addition of CaCl2 in the absence and in the presence of APC, and the
time for clot formation is recorded.

Reagents
Most commercially available reagents are suitable. However, reaction condi-
tions described below apply only to the brand of reagents mentioned here.
Optimal conditions should be established for each reagent.
164
 APC RESISTANCE
1. APT[ reagent (Automated APT[ reagent, Organon Teknika).
2. CaCl2 butTer (PH 7.5): 10 mmoVL Tris-HCl, 50 mmoVL NaCl, 30 mmoVL
CaCI2 , 0.1% bovine serum albumin.
3. APC/CaClrbuffer. 15 nmollL APC preparation (Enzymes Research
Laboratories) in CaCl2 butTer.

Procedure
This applies to the manual (tilt-tube) technique used in combination with the
above reagents. Most commercially available coagulometers can be used in the
APTTmode.
1. Place the necessary glass test tubes in a water bath at 37°C.
2. Pipette 0.1 m1 undiluted test plasma and 0.1 m1 APTT reagent. Incubate for
3 min.
3. Pipette 0.1 m1 CaCl2 butTer and simultaneously start a stopwatch. Mix and
tilt the tube back and forth regularly until clot forms, and record the clotting
time.
4. Repeat the measurement for the same plasma with APC/CaCl2 butTer instead
of CaCl2 butTer.
The above procedure can also be carried out on plasma prediluted 1:5 in
FV-deficient plasma (STA Factor V, Diagnostica Stago).
Results are usually expressed as ratio of clotting time with APC to clotting
time without APC [APC sensitivity (APC-S) ratio]. The value of normalization
by a pooled normal plasma will be discussed in the next section. Many labora-
tories use a commercial APTT-based method (Chromogenix). This follows
from the method originally described by Dahlback and colleagues!

Interpretation of results
Lower than normal APTT APC-S ratios are to be expected in the following
situations: heterozygous and homozygous FV:Q506 mutation; FV haplotype
HR2; women on oral contraceptives; subjects with high FVIII activity and
patients with cerebrovascular disease. Very low APTT APC-S ratios mimicking
homozygosity for the FV:Q506 mutation can be found in heterozygous patients
who also carry a quantitative FV defect2 ! . Results are difficult to interpret for
patients on oral anticoagulants, on heparin, or with lupus anticoagulants and
clotting factor deficiencies.

ADDITIONAL CONSIDERATIONS IN SETTING UP THE APTT-BASED

METHOD
Optimization and standardization of the test
After the first description of the original method!, it became clear that the
results obtained with the APTT-based method were influenced by the type of
165
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
reagent. Commercial APTT reagents vary widely in type and concentration of
phospholipids and activators. These are important determinants of the
responsiveness of different reagents to lupus anticoagulants and to some
clotting factors, namely FVIII. Responsiveness to APe depends on the FV:Q506
mutation, but also on the procoagulant drive of test plasmas, and this may
explain the variable results obtained by different methods for plasmas from
patients with high levels of FVIII and lupus anticoagulants and no identifiable
mutations. Other variables which can affect results are the source and
concentration of the APC preparation and of CaCl2 and the use of different
batches of the same APTT reagent9. Residual platelets in plasmas have been
reported to affect results considerably, especially if the samples are frozen for
later testini2 ,23. It can be anticipated that the extent of the influence will depend
on the reagent used. All the above considerations clearly demonstrate that
optimization of the method within the same laboratory and standardization
across laboratories are formidable tasks. As mentioned above, predilution of
test plasmas in FV-deficient plasma greatly improves the specificity of the
APTT-based method for the FV:Q506 mutation in patients on oral anti-
coagulants ll ,12, but also for patients with normal APTT 14• However, the
choice of the deficient plasmas and the optimal dilution should be established
with caution. Home-made artificially FV-depleted plasmas obtained by
incubation at 37 DC of normal plasma collected in oxalate as anticoagulant
would be the cheapest choice, but it should be remembered that these plasmas
are also FVIII-deficient. As a consequence, FVIII in test plasmas will not be
normalized, making the results dependent on this important variable (see
above). On the other hand, immunodepleted deficient plasmas are more suitable,
but expensive. Alternatively, home-made artificially depleted plasmas can be
supplemented with exogenous FVIII. The optimal dilution depends very much
on the type of system used for testing (i.e. AP'IT reagent/instrument combination,
APC preparation and FV-deficient plasma). Guidelines are given by Bertina24,
who suggests plotting the APC-S ratios obtained by testing increasing dilutions
of both a pooled normal plasma and a coumarin plasma in FV-deficient plasma,
and choosing the dilution which gives the same APC-S ratio for both.

Expression of results

Normalization of results can be achieved by dividing the patient's APC-S ratio

by that obtained for a pooled normal plasma (PNP) run in the same assay9.
The advantages of normalization are limited mainly to keeping at bay the
day-to-day methodological variability due to small changes of testing parameters
(i.e. assay temperature, pipettes, etc.) and to different batches of reagents which
can occur over time within the same laboratory9. The potential advantage of
harmonizing results obtained in different laboratories is less evident. A recent
study showed that normalization achieved with the same PNP made results
obtained with different methods more comparable, but only when the analysis
was restricted to results from non-carriers of the FV:Q506 mutation25 • Whatever
they may be, it should be kept in mind that all advantages of normalization
could be obscured by the distinct effect that even small amounts of plasma
166
 APC RESISTANCE
from carriers of the mutation may have on the performance of the PNP. It was
shown that as low as 2.5% of mutated FV (i.e. one heterozygote accidentally
included in a PNP of 20 subjects) was able to considerably decrease the APC-S
ratio of the PNP25. This calls for the laboratory to screen healthy subjects
selected for the preparation of PNP.
Reference range
This should be established in each laboratory for each particular set of reagents
and procedures. Even small changes in the established procedure may introduce
upredictable variations. Normal ranges should also be checked from time to
time even though no changes in the procedure have occurred, particularly when
results are not normalized. In that situation the effects of changing batches of
APTT reagent and APC preparation may be unpredictable. The number of
healthy subjects selected to establish the normal range should be sufficient to
account for the wide variability of the APC-S ratio in the normal population.
Fifty or more subjects might be adequate. APC-S ratios from healthy subjects
are seldom normally distributed. Therefore, the common practice of taking
the normal range as the interval of mean ±2 SO might not be adequate.
Transformation into log values prior to statistical analysis, or the use of per-
centiles, is more appropriate.

DIAGNOSTIC STRATEGY FOR APC RESISTANCE

Different approaches can be used in different laboratories depending on the

amounts of the resources that can be allotted to the investigation of
thrombophilic patients. The one described in Figure 17.1 might be cost-saving,
yet effective. Patients with a history of thrombosis should be screened with the
APTT-based method. Patients with frank normal values will be considered
normal. Those with frank abnormal values should have DNA analysis to
confirm the presence of the mutation. Those with borderline values can be
tested with predilution of test plasma in FV-deficient plasma. However, this
strategy is valid only when there is evidence that the negative predictive value
of the APTT-based method used for screening is sufficiently high to rule out

APTT-based ~ No
~ APC-resistance
Borderline
/
FV-Def. Plasma+INormall-- No
~APc-resistance

IAbn~ ~
DNA analysis
Figure 17.1 Flow chart for APe resistance testing
167
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
false-negative results for the FV:Q506 mutation. If this is not the case, the use
of the modified APTT-based method (i.e. predilution of test plasma in
FV-deficient plasma) is recommended.

References
1. Dahlbiick B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously UIIIeCOgnized
mechanism characterized by poor anticoagulant response to activated protein C: prediction
of a cofactor to activated protein C. Proc Natl Acad Sci USA 1993; 90: 1004-8.
2. Exner T, Murray B, Chong BH, Chesterman CN. Improved APC resistance method based
on a Russell viper venom clotting test. Thromb Haemostas 1995; 73: 119.
3. Le DT, Greengard JS, Mujumdar V, Rapaport SI. Use of a generally applicable tissue factor-
dependent factor V assay to detect activated protein C-resistant factor Va in patients receiving
warfarin and in patients with a lupus anticoagulant. Blood 1995; 85: 1704-11.
4. Ahlenc-Gelas M, Aillaud MF, Bonvarlet MN, Dupuy G, Juhan-Vague I, Aiach M. Specificity
of an assay based on a factor V-depleted plasma in patients carrying the Arg 506 Gn mutation.
Thromb Haemostas 1996; 75: 971-7.
5. Bertina RM, Koeleman BPC, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, van der
Veblen PA, Reitsma PH. Mutation in blood coagulation factor V associated with resistance
to activated protein C. Nature 1994; 369: 64-7.
6. Bernardi F, Faioni EM, Castoldi E, Lunghi B, Castaman G, Sacchi E, Mannucci PM. Factor
V genetic component different from factor V R506Q contributes to the activated protein C
resistance phenotype. Blood 1997; 90: 1552-7.
7. Varadi K, Moritz B, Lang H, Bauer K, Preston E, Peake I, Rivard GE, Keil B, Schwarz HP.
A chromogenic assay for activated protein C resistance. Br J Haematol1995; 90: 884-9.
8. Rosing J, Tans G, Nicolaes GAF, Thomassen MCLGD, van Oerle R, van der Ploeg PMEN,
HeYnen P, Hamulyak K, Hemker He. Oral contraceptives and venous thrombosis: different
sensitivities to activated protein C in women using second- and third-generation oral contracep-
tives. Br J Haemato11997; 97: 233-8.
9. de Ronde H, Bertina RM. Laboratory diagnosis of APe resistance: a critical evaluation of
the test and the development of diagnostic criteria. Thromb Haemostas 1994; 72: 880-6.
10. Halbmayer DM, Hanshofer A, Schon R, Fisher M. Influence of lupus anticoagulant on a
commercially available kit for APC-resistance. Thromb Haemostas 1994; 72: 645-6.
II. Jorquera n, Montoro JM, Angeles Fernandez M, Aznar JA, Aznar J. Modified test for
activated protein C resistance. Lancet 1994; 344: 1162-3.
12. Trossaert M, Conard J, Horellou MH, Samama MM, Ireland H, Bayston FA, Lane DA.
Modified APC-resistance assay for patients on oral anticoagulants. Lancet 1994; 344: 1709.
13. Martorell JR, Munoz-Castillo A, Gil JL. False positive activated protein C resistance test due
to antiphospholipid antibodies is corrected by platelet extract. Thromb Haemostas 1995; 74:
796-7.
14. Tripodi A, Negri B, Bertina RM, Mannucci PM. Screening of the FV:Q506 mutation. Evalu-
ation of thirteen plasma-based methods for their diagnostic efficacy in comparison with DNA
analysis. Thromb Haemostas 1997; 77: 436-9.
IS. Hellgren M, Svensson PJ, Dahlbii.ck B. Resistance to activated protein C as a basis for venous
thromboembolism associated with pregnancy and oral contraceptives. Am J Obstet Gynecol
1995; 173:210-13.
16. Olivieri 0, Friso S, Manzato F, Guella A, Bernardi F, Lunghi B, Girelli D, Azzini M, Brocco
G, Russo C et al. Resistance to activated protein C in healthy women taking oral contracep-
tives. Br J Haematol1995; 91: 465-70.
17. Henkens CMA, Bom VJJ, Seinen AJ, van der Meer J. Sensitivity to activated protein C;
influence of oral contraceptives and sex. Thromb Haemostas 1995; 73: 402-4.
18. Henkens CMA, Bom VJJ, van der Meer J. Lowered APe-sensitivity ratio related to increased
factor VIII clotting activity. Thromb Haemostas 1995; 74: 1198-9.
19. Laffan MA, Manning R. The influence of factor VIII on measurement of activated protein
C resistance. Blood Coagul Fibrinol1996; 7: 761-5.

168
 APC RESISTANCE
20. van der Born JG, Bots ML, Haverkate F, Slagboom PE, Meijer P, de Jong PTVM, Hofman
A. Grobbee PE, Kluff C. Reduced response to activated protein C is associated with increased
risk for cerebrovascular disease. Ann Intern Med 1996; 125: 265-9.
21. Simioni P, Scudeller A, Radossi P, Gavasso S, Girolami B, Tormene D, Girolami A. 'Pseudo
homozygous' activated protein C resistance due to double heterozygous factor V defects
(factor V Leiden mutation and type I quantitative factor V defect) associated with thrombosis:
report of two cases belonging to two unrelated kindreds. Thromb Haemostas 1996; 75: 422-6.
22. Sidelman J, Gram J, Pedersen OD, Jespersen 1. Influence of plasma platelets on activated
protein C resistance assay. Thromb Haemostas 1995; 74: 993-4.
23. Shizuka R, Kauda T, Amagai H, Kobayashi I. False positive activated protein C sensitivity
ratio (APe). Sensitivity ratio caused by freezing and by contamination of plasma with platelets.
Thromb Res 1995; 78: 189-90.
24. Bertina RM. Laboratory diagnosis of resistance to activated protein C. Thromb Haemostas
1997; 78: 478-82.
25. Tripodi A, Chantarangkul V, Negri B, Mannucci PM. Standardization of the APC-resistance
test. Effects of norma1ization of results by means of pooled normal plasma. Thromb Haemostas
1998; 79: 564-6.

169
18
Tissue factor pathway inhibitor
(TFPI)
P. M. SANDSET

INTRODUCTION

Tissue factor pathway inhibitor (TFPI) is a plasma glycoprotein which belongs

to the Kunitz-type family of protease inhibitors. The mature molecule contains
276 amino acids, and consists of an acidic amino terminal end, three tandem
Kunitz-type inhibitory domains, a highly basic carboxyl terminal end, and
three potential N-linked glycosylation sites!. The molecule contains a high-
affinity binding site for heparin located in the basic region of the carboxyl
terminal end2 and a low-affinity binding site located in the third Kunitz domain3•
TFPI is produced in endothelial cells~. The synthesis is constitutive and not
affected by cytokines5 •
Although wild-type TFPI mutants associated with thromboembolic disease
have not yet been detected, there is now considerable experimental evidence
(see below) that TFPI plays a major role in regulating tissue factor-induced
blood coagulation.

Physiological roles
Inhibition of factor Xa and the tissue factor/factor Vila complex
The target proteases of TFPI activity are factor VIla and factor Xa. Inhibition
is mediated by interaction of the protease with the first and second Kunitz
domains, respectively6. TFPI inhibits factor Xa directly7,8, whereas inhibition
of the tissue factor (TF)NIIa complex requires the presence of factor Xa and
calcium ions6,8. Inhibition of the TFNIIa complex is commonly thought to
proceed in two steps, first by the formation of a TFPI/Xa complex and then
171
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
the building of a quaternary TFPIlXafTFNIIa complex7,8. An alternative
hypothesis is that TFPI binds directly to the XafTFNIIa activator complex.
At supra-physiological concentrations, TFPI may also inhibit TFMIa complexes
in the absence of factor Xa9, which may be clinically relevant when recombinant
TFPI is considered for use as a therapeutic agent.
The TFNIIa catalytic complex cleaves both factor IX and factor X10. Gener-
ation of factor Xa produces rapid feedback inhibition of the TFMIa complex,
but amplification of coagulation through the intrinsic cascade (factors VIII,
IX, and XI) normally provides sustained activation and efficient haemostasis.
In the absence of factor VIII and factor IX, TFPI effectively turns off activation
of coagulation, which explains why individuals with factor VIII or factor IX
deficiencies bleed II.

The heterogeneity of TFPI

TFPI is present in vivo in different pools, in different molecular weight forms,

and as a free molecule or in complex with lipoproteins. The physiological
role of the different types of TFPI is yet not fully known or understood, but it is
obvious that the heterogeneity of TFPI is important for assay methodology.
TFPI is present in at least three different intravascular pools. A major pool,
approximately 50--80% of the total intravascular pool, is normally bound to
the vessel wall, but may be released into plasma following injection of hep-
arin l2,13. Accordingly, this pool is often referred to as heparin-releasable TFPI.
A smaller pool, approximatelI 10-50% of total, circulates in plasma at a
concentration of 50--150 ng/mIl . More than 80% of plasma TFPI is in complex
form with lipoproteins, mainly with low-density lipoproteins l 4-17. A small
amount of TFPI is also associated with high-density lipoproteins IS ,16. Less
than 10--20% circulates as carrier-free TFPI molecules. The smallest TFPI pool
is contained in platelets, which may be released by platelet activation and thus
potentially play an important regulatory role in primary haemostasisl 8 .
Heparin-releasable TFPI is full-length TFPI with MW 40 k0 19, whereas
plasma TFPI, i.e. both carrier-free TFPI and TFPI bound to lipoproteins, is
mostly truncated forms of TFPI 16,20. TFPI bound to low-density lipoproteins
is truncated with MW 34 kO I6 ,20,21, whereas TFPI bound to high density
lipoproteins is due to a mixed disulphide complex between TFPI and apoli-
poprotein All with MW 41 kO I6,20.
The anticoagulant effect of TFPI can be determined in a one-stage diluted
prothrombin time assay in which coagulation is triggered by a small amount
of TF. When different forms of TFPI are purified and tested in this assay, the
anticoagulant responses vary considerably. Heparin-releasable TFPI and other
full-length forms of TFPI exert the strongest anticoagulant effects21 ,22-25,
whereas lipoprotein associated TFPI only has weak anticoagulant proper-
ties 23 . The data may imply that full-length TFPI is biologically active as an
anticoagulant.

172
 TISSUE FACTOR PATHWAY INHIBITOR

Pathophysiological aspects
A number of studies have reported on the plasma levels of TFPI in various
clinical conditions. It does not behave as an acute-phase reactant26,27. Plasma
levels are significantly affected in patients with dyslipidaemia, with low levels
in patients with ~-li~oproteinaemia 13 and high levels in patients with
hypercholesterolaemia 8. Occasionally, low TFPI levels have been detected in
patients with severe disseminated intravascular coagulation (DIC), but it is
now evident that DIe may proceed despite normal or elevated plasma TFPI
levels13,27,29-31.
Low levels of TFPI associated with a clear familial pattern have not been
detected, which may either indicate that inherited TFPI deficiency is rare or that
the approach to detect deficiency has not been appropriate 32 . Experimentally,
gene targeting techniques have been used to disrupt exon 4 of the TFPI gene in
mice33 . Homozygous animals suffered severe circulatory insufficiency, consumptive
coagulopathy, and died in utero. Heterozygous animals displayed normal phenotype
with plasma TFPI levels 50% of normal, but further studies are needed to rule
out the possibility that heterozygosity is associated with thrombosis.
TFPI deficiency may be induced experimentally by the infusion of specific
antibodies or Fab fragments against TFPI34-36. Immunodepletion of TFPI in
the rabbit to values less than 20% of normal dramatically reduced the threshold
by which TF34,3S, but not factor Xa36, could produce severe DIe. These studies
indicate that there is a threshold above which TFPI can prevent the coagulation-
triggering effect of TF, and that TFPI is not a significant inhibitor of factor
Xa in vivo.

TFPIASSAYS

The development of assays for TFPI is severely complicated by the fact that
TFPI is contained in different intravascular pools and that TFPI is heterogeneous
both with regard to truncation and association with lipoproteins. The question
of assay is therefore not only a matter of assay methodology but also a matter
of what to assay. One may ask whether the assay of TFPI in a randomly
collected plasma sample, which contains mainly truncated forms of TFPI
associated with lipoproteins, can yield useful information on the functional
status of TFPI in that particular individual. The answer will require further
knowledge on the equilibrium between different pools of TFPI under various
physiological conditions and disease states, and knowledge of the biological
activity and role of the different forms of TFPI. At present a pragmatic approach
is to assay TFPI in plasma collected before, and 5-15 min after, a standardized
dose of unfractionated heparin, e.g. 5000 IV or 50-100 IV/kg, to acquire
information on resting plasma TFPI levels and on heparin-releasable TFPI.
The latter procedure should still be considered an experimental procedure as
part of a research protocol.
Basically, four types of assays can be recognized:
1. TFPI end-point activity assay,
2. TFPI coagulant activity assay,
173
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
3. immunological assay for total TFPI,
4. immunological assay for free TFPL
Activity assays can be constructed to assay the ability of TFPI to inhibit
TFNIIa catalytic activity or to inhibit factor Xa. Since the physiological role
of TFPI as a factor Xa inhibitor is questionable, this chapter will only describe
assays for TFNIIa inhibitory activity. Our in-house chromogenic substrate
assay37, adapted for use with commercial reagents, will be described in detail.
Assay of anticoagulant activity, and free and total antigen, will be described
only briefly.

TFPI END-POINT ACTIVITY ASSAY

Assay of TFPI activity is based on the ability of a sample containing TFPI to

inhibit known amounts of purified factor VII(a) and TF in the presence of
factor Xa. TFPI is allowed to react with TFNIIa complexes during a prolonged
incubation time, varying from several minutes to hours, to achieve near-
equilibrium conditions for the inhibitory reaction. The assays are therefore
functionally independent on reaction kinetics and technically end-point assays
or capacity assays sensitive to the molar concentration of TFPL
The reaction mixture can either be made to contain excess TF 37,38 or excess
factor VIIa 8,37-39. When the assay is standardized with excess TF, factor VII
activity of the test sample must be abolished to avoid the problem that variation
of factor VII(a) activity in the test plasma may influence the assay results.
Heating of plasma at 56°C for 15 min effectively removes factor VII activity
and defibrinates plasma, but with minimal effect on the ability of TFPI to
inhibit TFNIIa activity37,38. When the assay is standardized with excess factor
VII(a), the assay results will be insensitive to variation of factor VII, but may
be influenced by TF present in the samples, e.g. TF produced to cell culture
supernates.
Residual TFNIIa catalytic activity is determined after the addition of a
substrate, e.g., factor X + chromogenic substrate 37- 41 , factor X + factor
X-depleted plasma8, or tritium-labelled factor IX29 ,42. Presence of heparin in
the test sample may greatly influence both the inhibitory reaction, that requires
factor Xa as a cofactor, and the substrate reaction that involves generation of
factor Xa, mediated by the accelerating effect on antithrombin 38 . Heparin also
accelerates the activation of tritium-labelled factor IX by the TFMIa complex43 .
A heparin-neutralizing agent, el f,olybrene, is therefore used to prevent in-
fluence of heparin on the assays 7- 9.

Materials
Buffers
1. Stock buffer: Tris-saline-citrate (TSC) buffer: 50 mmollL Tris . HCl,
100 mmollL NaCl, 10 mmollL Na 3 citrate (pH 7.5). Sodium azide may be
added to 0.02% to keep the buffer stable at 4 °C for several months.
174
 TISSUE FACTOR PATHWAY INHIBITOR

2. Assay buffer (TSCIBSA buffer): TSC buffer containing bovine serum albumin
(BSA) to final concentration 2 mg/ml. The assay butTer is stable for 7 days
at 4 dc. It is used to dilute all reagents used in the assay.
3. Dilution buffer: TSC/BSA buffer containing polybrene (hexadimethrine
bromide, Sigma), final concentration 2 Ilg/ml, to neutralize heparin present
in the samples.

Reaction mixture

The following reagents, and procedures (1-4), are recommended for the
preparation of a combined reaction mixture (5). All reagent should be kept in
ice water during preparation:
1. Factor VIla: purified from human plasma (from Stago) or human,
recombinant material (Novo-Nordisk). Lyophilized material is first diluted
as described by the manufacturer, then further diluted in ice-cold assay
butTer to final concentration 62.5 ng/ml (1.25 nmollL), and stored in suitable
aliquots (e.g. 250 jJl) at -70°C. Before assay, aliquots are thawed and diluted
lO-fold in assay butTer to final concentration 6.25 ng/ml (125 pmollL).
2. Tissue factor: human, recombinant, relipidated TF (Innovin, Ortho).
Lyophilized material is first diluted as described by the manufacturer and
is stable for 1-2 weeks at 4 °C (may not be frozen). The preparation is further
diluted 20-fold in assay buffer immediately before preparation of the reaction
mixture (see below).
3. Factor Xa: bovine material (from Chemogenix). Lyophilized material is first
diluted as described by the manufacturer and further diluted in ice-cold assay
buffer to final concentration 3.3 nkatlml (0.25 U/ml or 40 nmollL) and stored
in suitable aliquots (e.g. 250 jJl) at -70°C. Before assay, aliquots are thawed
and diluted lO-fold in assay butTer to final concentration 0.33 nkatlml
(4 nmollL).
4. CaCl2 : 75 mmollL.
5. Combined reaction mixture: may be prepared immediately before assay by
mixing equal volumes of factor VIla, TF, factor Xa, and CaCl2 prediluted
as described above (1-4). A convenient volume sufficient for one microtitre
plate is 2.5 ml of each, total 10 ml.

Substrates
The following procedure is recommended for preparation of substrate reagents:
1. Factor X: bovine material (Chromogenix). The reagent must not contain
factor Xa activity. Lyophilized material is first diluted as described by the
manufacturer and further diluted in ice-cold assay buffer to final concentration
0.4 U/ml (64 mmollL) and stored in suitable aliquots at -70°C.
2. S-2222: chromogenic substrate (Chromogenix). Lyophilized material should
be dissolved in distilled water to final concentration 2 mg/ml (2.7 mmollL).
Provided it is protected from light, the solution is stable for 6 months at
4°C.
175
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

3. Substrate reagent mixture: a substrate mixture containing factor X and S-2222

may be prepared during the first incubation of the assay by mixing equal
volumes of factor X and S-2222 prediluted as described above. A convenient
volume sufficient for one microtitre plate is 2.5 ml of each, total 5 ml.

Stop reagent
Acetic acid 50%.

Calibrators, controls, and samples

Platelet-poor, citrated plasma is the preferred material, but TFPI may also be
assayed in ethylenediaminetetra-acetic acid (EDTA) plasma37 and serum37,44.
Plasma (or serum) to be used as calibrator, control or sample is heat-inactivated
at 56 °C for 15 min using a standardized procedure. We pursue the following
procedure: 300 III plasma in 2 ml capped tubes is placed in a water bath at
56 °C for 15 min. The samples are then immediately placed in ice water for
5 min and centrifuged at 15 OOOg for 1 min. Centrifuged plasma is stable at
4 °C for at least 24 h. Calibrators, controls, and samples are diluted in dilution
buffer (see Buffers, above).
1. Calibrators: an international standard for the assay of TFPI is not yet
available. At present the optimal calibration of any TFPI assay should
employ pooled citrated normal plasma (PNP), which may be obtained by
mixing platelet-poor citrated plasma from 20 to 40 healthy volunteers and
stored frozen in suitable aliquots (0.5 ml) at -20 °C.
2. Controls: commercial standards are not available and in-house controls
should be established. Each run should employ at least one control within
the normal range and one control with high TFPI activity. We use plasma
from one individual collected before, and 5 min after, intravenous injection
of unfractionated heparin 5000 IV.
3. Samples: the preferred material is platelet-poor, citrated plasma.

Other materials
Flat-bottomed microtitre plates are purchased from Dynatech, but other plates
may also be used after analysis of their suitability for the assay. Absorbances
are read on a conventional multichannel analyser using a filter of 405 nm (e.g.
Thermomax, Molecular Devices, CA, USA).

Test procedure
Test protocol
Suitable dilutions of standards, controls, and test plasma (25 J.Ll) are added to
the wells. The combined reaction mixture (100 J.Ll) containing factors VIla and
Xa, TF and CaCl2 (see Reaction mixture, above) is then added to each well
176
 TISSUE FACTOR PATHWAY INHIBITOR

using a multichannel pipette. The micro titre plate is sealed using Parafilm and
placed in an incubator on gentle shaking at 37 °C for 30 min. Substrate reagent
mixture (50~) containing factor X and S-2222 (see Substrates, above) is then
added to each well using a multichannel pipette and the plate is further incubated
at 37 °C for approximately 30 min (until sufficient colour has developed, i.e.
OD405 > 1.2 in buffer control). The reaction is quenched by the addition of
50% acetic acid (50 ~). Finally, the absorbance at 405 nm is read in a multi-
channel photometer (see Other materials, above).

Calibration

The assay is calibrated with dilutions of PNP in assay buffer (see Buffers,
above). An eight-point calibration curve is made by pipetting 25 ~ in duplicate
wells of 4% (1125),3%,2% (1/50), 1.5%, 1% (11100), 0.5%, 0.25%, and 0%
plasma, which is diluted 5-fold by the addition of the reaction mixture to final
concentrations 0-0.8% plasma. Routinely, plasma samples from patients not
receiving heparin are tested in duplicates in 1150 and/or 11100 dilutions, while
plasmas of patients treated with heparin are tested in duplicates in 11100 and/or
11200 dilutions. Testing of the samples in two dilutions is recommended. Each
run should include one normal and one high-level control (see Calibrators,
controls and samples, above). The intra-assay CV should be kept below 4-6%
and inter-assay CV below 8-10%.
TFPI activity is calculated from the standard curve which is obtained by a
quadratic fit of the data plotted on a linear scale, and is expressed as a percentage
or in units/mI, where I unit is defined as the TFPI activity present in I mI PNP.
Normal reference levels determined with this method include a wide range with
individual values from 0.45 to 1.75 Ulmi. The activity increases with age and
is positively correlated with total cholesterol37 •

TFPI ANTICOAGULANT ACTIVITY (KINETIC ASSAy)

The anticoagulant effect of TFPI may be determined in a modified or diluted

prothrombin time assay, in which a low concentration of TF is used to trigger
coagulation23 ,25,45--48. Unfortunately, all reported coagulant assays are still
semi-quantitative in nature as it is not possible to convert a given prolongation
in clotting times to a percentage or unit activity per millilitre. The anticoagulant
effect of TFPI in such assays is highly dependent on its truncation. Full-length
TFPI exerts a strong anticoagulant activity, whereas truncated TFPe2,23 and
lipoprotein-associated TFPI 21 ,23,48 only have minor ability to prolong
coagulation. The difference in anticoagulant activity indicates a difference in
biological activity, but it has yet to be proven that full-length TFPI is the active
form of TFPI in vivo. The following protocols for estimation of anticoagulant
activity may be used:
I. Anticoagulant activity of purified or semi-purified TFPI preparations: 40 III
citrated PNP is incubated with 10 ~ Tris buffered saline (50 mmollL
177
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Tris· HCl, 150 mmoVL NaCl, pH 7.5) or 10 ~ TFPI preparation and 50 ~
diluted TF (e.g. Innovin) for 3-5 min at 37°C. Coagulation is triggered by
the addition of 50 ~ of 35 mmoVL CaCI2 • The assay is first standardized
with the buffer control by varying the concentration of TF (usually 11300
to 1I500-fold predilution of Innovin) to obtain a clotting time of PNP of
approximately 50 s. At this clotting time TFPI present in PNP will have only
a small backAround effect on the assay, i.e. responsible for 1-3 s of the
clotting time . The anticoagulant effect of a TFPI preparation may then
be determined as the prolongation of clotting times obtained after its
replacement of buffer in the assay.
2. Anticoagulant activity of heparin-releasable TFPI: citrated plasma samples
are obtained before and after injection of heparin using a standardized
procedure (see TFPI assays, above). In principle, the same anticoagulant
assay described for purified TFPI is utilized, except that patient plasma is
used instead of PNP and that hexadimethrine bromide (25 ~g1ml) is used
instead of buffer (to neutralize heparin). The prolongation of clotting times
obtained after the injection of heparin is a semi-quantitative measure of the
anticoagulant effect of heparin-releasable TFPI.

TFPI ANTIGEN

The heterogeneity of TFPI in plasma makes measurement of TFPI antigen a

challenging task. Pending the specificity of the antibodies, immunological
assays may be constructed to measure full-length TFPI or various truncated
forms of TFPI. A special problem is the measurement of lipoprotein-associated
TFPI, since the larger lipid moiety may interfere with the detection of TFPI.
In selecting an antigen assay one should therefore carefully consider what to
assay, but further knowledge of the pathophysiological role of the different
forms of TFPI is also required. Most assays to date are calibrated with various
types of recombinant TFPI, which makes comparisons of test results difficult
or impossible. It is therefore recommended to also calibrate against PNP. Paral-
lelism of standard curves obtained with purified material and PNP is important
for validation of the assays.

TFPI free antigen

TFPI free antigen may be measured in immunoassays employing antibodies
with specificities for either the third Kunitz type inhibitory domain so,sl or the
C-terminal end39 , which are normally hidden in lipoprotein-associated TFPI.
Such assays do not detect lipoprotein-associated TFPIso ,Sl, but are highly
sensitive to heparin-releasable TFPI 39,s2. TFPI free antigen assays are of
potential major importance if free TFPI, but not lipoprotein-associated TFPI,
is biologically the active form in vivo. For the assay of heparin-releasable TFPI,
it is important first to rule out an in-vitro effect of heparin on the assay results,
as heparin may influence the binding affinity of the antibodys3. In one assay,
plasma free TFPI antigen levels varied from 7 to 26 nglml s2, and the level
178
 TISSUE FACTOR PATHWAY INHIBITOR

increased 8-13-fold after the injection of heparin21. An immunoassay with

similar characteristics is now commercially available (lmubind free TFPI ELISA
kit no. 850, American Diagnostica)51.

TFPI total antigen

Both commercial and in-house methods are now available for the assay of total
TFPI antigen 13,41. Unfortunately, the assays have not yet been properly validated
for ability to detect free and lipoprotein-associated forms of TFPI. Standard
curves calibrated with normal plasma and purified recombinant TFPI are not
parallel (own unpublished result), which may be due to problems detecting
lipoprotein-associated TFPI.

References
1. Wun TC, Kretzmer KK, Girard TJ, Miletich JP, Broze GJ, Jr. Cloning and characterization
of a cDNA coding for the lipoprotein-associated coagulation inhibitor shows that it consists
of three tandem Kunitz-type inhibitory domains. J Bioi Chem 1988; 263: 6001-4.
2. Wesselschmidt R, Likert K, Girard T, Wun TC, Broze GJ, Jr. Tissue factor pathway inhibitor:
the carboxy-terminus is required for optimal inhibition of factor Xa. Blood 1992; 79: 2004--10.
3. Wesselschmidt R, Likert K, Huang Z, MacPhail L, Broze GJ, Jr. Structural requirements for
tissue factor pathway inhibitor interactions with factor Xa and heparin. Blood Coagul Fibrinol
1993; 4: 661-9.
4. Bajaj MS, Kuppuswamy MN, Saito H, Spitzer SG, Bajaj SP. Cultured normal human
hepatocytes do not synthesize lipoprotein-associated coagulation inhibitor: evidence that
endothelium is the principal site of its synthesis. Proc Nat! Acad Sci USA 1990; 87: 8869-73.
5. Warn-Cramer BJ, Almus FE, Rapaport SI. Studies of the factor Xa-dependent inhibitor of
factor VILa/tissue factor (extrinsic pathway inhibitor) from cell supemates of cultured human
umbilical vein endothelial cells. Thromb Haemostas 1989; 61: 101-5.
6. Girard TJ, Warren LA, Novotny WF, Likert KM, Brown SG, Miletich JP, Broze GJ, Jr.
Functional significance of the Kunitz-type inhibitory domains of lipoprotein-associated
coagulation inhibitor. Nature 1989; 338: 518-20.
7. Warn-Cramer BJ, Rao LV, Maid SL, Rapaport SI. Modifications of extrinsic pathway inhibitor
(EPI) and factor Xa that affect their ability to interact and to inhibit factor VILa/tissue factor:
evidence for a two-step model of inhibition. Thromb Haemostas 1988; 60: 453--6.
8. Broze GJ, Jr, Warren LA, Novotny WF, Higuchi DA, Girard JJ, Miletich JP. The lipoprotein-
associated coagulation inhibitor that inhibits the factor VII-tissue factor complex also inhibits
factor Xa: insight into its possible mechanism of action. Blood 1988; 71: 335-43.
9. Callander NS, Rao LV, Nordfang 0, Sandset PM, Warn-Cramer B, Rapaport SI. Mechanisms
of binding of recombinant extrinsic pathway inhibitor (rEPI) to cultured cell surfaces. Evidence
that rEPI can bind to and inhibit factor VIla-tissue factor complexes in the absence of factor
Xa. J Bioi Chern 1992; 267: 876-82.
10. 0sterud B, Rapaport SI. Activation of factor IX by the reaction product of tissue factor and
factor VII: additional pathway for initiating blood coagulation. Proc Nat! Acad Sci USA
1977; 74: 5260-4.
11. Repke D, Gemmell CH, Guha A, Turitto VT, Broze GJ, Jr, Nemerson Y. Hemophilia as a
defect of the tissue factor pathway of blood coagulation: effect of factors VIII and IX on
factor X activation in a continuous-flow reactor. Proc Natl Acad Sci USA 1990; 87: 7623-7.
12. Sandset PM, Abildgaard U, Larsen ML. Heparin induces release of extrinsic coagulation
pathway inhibitor (EPI). Thromb Res 1988; 50: 803-13.
13. Novotny WF, Brown SG, Miletich JP, Rader DJ, Broze GJ, Jr. Plasma antigen levels of the
lipoprotein-associated coagulation inhibitor in patient samples. Blood 1991; 78: 387-93.
14. Hubbard AR, Jennings CA. Inhibition of the tissue factor-factor VII complex: involvement
of factor Xa and lipoproteins. Thromb Res 1987; 46: 527-37.
179
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
15. Warn-Cramer BJ, Maid SL, Zivelin A, Rapaport SI. Partial purification and characterization
of extrinsic pathway inhibitor (the factor Xa-dependent plasma inhibitor of factor VIla/tissue
factor). Thromb Res 1987; 48: 11-22.
16. Novotny WF, Girard TJ, Miletich JP, Broze GJ, Jr. Purification and characterization of the
lipoprotein-associated coagulation inhibitor from human plasma. J Bioi Chern 1989; 264:
18832-7.
17. Hansen JB, Huseby NE, Sandset PM, Svensson B, Lyngmo V, Nordoy A. Tissue-factor
pathway inhibitor and lipoproteins. Evidence for association with and regulation by LDL in
human plasma. Arterioscler Thromb 1994; 14: 223-9.
18. Novotny WF, Girard TJ, Miletich JP, Broze GJ, Jr. Platelets secrete a coagulation inhibitor
functionally and antigenically similar to the lipoprotein associated coagulation inhibitor.
Blood 1988; 72: 2020-5.
19. Novotny WF, Palmier M, Wun TC, Broze GJ, Jr, Miletich JP. Purification and properties of
heparin-releasable lipoprotein-associated coagulation inhibitor. Blood 1991; 78: 394-400.
20. Broze GJ, Jr, Lange GW, Duffin KL, MacPhail L. Heterogeneity of plasma tissue factor
pathway inhibitor. Blood Coagul Fibrinol1994; 5: 551-9.
21. Hansen JB, Huseby KR, Huseby NE, Ezban M, Nordoy A. Tissue factor pathway inhibitor
in complex with low density lipoprotein isolated from human plasma does not possess
anticoagulant function in tissue factor-induced coagulation in vitro. Thromb Res 1997; 85:
413-25.
22. Lindahl AK, Abildgaard U, Larsen ML, Staalesen R, Hammer AK, Sandset PM, Nordfang
0, Beck TC. Extrinsic pathway inhibitor (EPI) released to the blood by heparin is a more
powerful coagulation inhibitor than is recombinant EPI. Thromb Res 1991; 62: 607-14.
23. Lindahl AK, Jacobsen PB, Sandset PM, Abildgaard U. Tissue factor pathway inhibitor with
high anticoagulant activity is increased in post-heparin plasma and in plasma from cancer
patients. Blood Coagul Fibrinol1991; 2: 713-21.
24. Nordfang 0, Bjorn SE, Valentin S, Nielsen LS, Wildgoose P, Beck TC, Hedner U. The
C-terminus of tissue factor pathway inhibitor is essential to its anticoagulant activity.
Biochemistry 1991; 30: 10371-6.
25. Wun TC, Kretzmer KK, Palmier MO, Day KC, Huang MD, Welsch DJ, Lewis C, Wolfe RA,
Zobel JF, Lange GW et af. Comparison of recombinant tissue factor pathway inhibitors
expressed in human SK hepatoma, mouse C127, baby hamster kidney, and Chinese hamster
ovary cells. Thromb Haemostas 1992; 68: 54-9.
26. Sandset PM, Hogevold HE, Lyberg T, Andersson TR, Abildgaard U. Extrinsic pathway
inhibitor in elective surgery: a comparison with other coagulation inhibitors. Thromb Haemostas
1989; 62: 856-60.
27. Warr TA, Rao LV, Rapaport SI. Human plasma extrinsic pathway inhibitor activity: II. Plasma
levels in disseminated intravascular coagulation and hepatocellular disease. Blood 1989; 74:
994-8.
28. Sandset PM, Lund H, Norseth J, Abildgaard U, Ose L. Treatment with hydroxyrnethylglutaryl-
coenzyme A reductase inhibitors in hypercholesterolemia induces changes in the components
of the extrinsic coagulation system. Arterioscler Thromb 1991; 11: 138-45.
29. Bajaj MS, Rana SV, Wysolmerski RB, Bajaj SP. Inhibitor of the factor VIla-tissue factor
complex is reduced in patients with disseminated intravascular coagulation but not in patients
with severe hepatocellular disease. J Clin Invest 1987; 79: 1874-8.
30. Brandtzaeg P, Sandset PM, Joe GB, 0vstebo R, Abildgaard U, Kierulf P. The quantitative
association of plasma endotoxin, antithrombin, protein C, extrinsic pathway inhibitor and
fibrinopeptide A in systemic meningococcal disease. Thromb Res 1989; 55: 459-70.
31. Sandset PM, Reise 0, Aasen AO, Abildgaard U. Extrinsic pathway inhibitor in postoperative!
posttraumatic septicemia: increased levels in fatal cases. Haemostasis 1989; 19: 189-95.
32. Sandset PM, Bendz B. Tissue factor pathway inhibitor: clinical deficiency states. Thromb
Haemostas 1997; 78: 467-70.
33. Huang ZF, Higuchi D, Lasky N, Broze GJ, Jr. Tissue factor pathway inhibitor gene disruption
produces intrauterine lethality in mice. Blood 1997; 90: 944-51.
34. Sandset PM, Warn-Cramer BJ, Maki SL, Rapaport SI. Immunodepletion of extrinsic pathway
inhibitor sensitizes rabbits to endotoxin-induced intravascular coagulation and the generalized
Shwartzman reaction. Blood 1991; 78: 1496-502.
35. Sandset PM, Warn-Cramer BJ, Rao LV, Maki SL, Rapaport SI. Depletion of extrinsic pathway

180
 TISSUE FACTOR PATHWAY INHIBITOR
inhibitor (EPI) sensitizes rabbits to disseminated intravascular coagulation induced with tissue
factor: evidence supporting a physiologic role for EPI as a natural anticoagulant. Proc Natl
Acad Sci USA 1991; 88: 708-12.
36. Warn-Cramer BJ, Rapaport SI. Studies of factor XaJphospholipid-induced intravascular
coagulation in rabbits. Effects of immunodepletion of tissue factor pathway inhibitor.
ArteriosclerThromb 1993; 13: 1551-7.
37. Sandset PM, Larsen ML, Abildgaard U, Lindahl AK, Odegaard OR. Chromogenic substrate
assay of extrinsic pathway inhibitor (EPI): levels in the normal population and relation to
cholesterol. Blood Coagul Fibrinol1991; 2: 425-33.
38. Sandset PM, Abildgaard U, Pettersen M. A sensitive assay of extrinsic coagulation pathway
inhibitor (EPI) in plasma and plasma fractions. Thromb Res 1987; 47: 389-400.
39. Hubbard AR, Weller D, Gray E. Measurement of tissue factor pathway inhibitor in normal
and post-heparin plasma. Blood Coagul Fibrinol1994; 5: 819-23.
40. Berrettini M, Malaspina M, Parise P, Lucarelli G, Kisiel W, Nenci 00. A simple chromogenic
substrate assay of tissue factor pathway inhibitor activity in plasma and serum. Am J Clin
Patho11995; 103: 391-5.
41. Boguacki J, Hammelburger J. Functional and immunologic methods for the measurement of
human tissue factor pathway inhibitor. Blood Coagul Fibrinol1995; 6 (Suppl. 1): S65-72.
42. Warr TA, Warn-Cramer BJ, Rao LV, Rapaport SI. Human plasma extrinsic pathway inhibitor
activity: I. Standardization of assay and evaluation of physiologic variables. Blood 1989; 74:
201-6.
43. Nemerson Y. Tissue factor and hemostasis [published erratum appears in Blood 1988; 71:
1178]. Blood 1988; 71: 1-8.
44. Sandset PM, Andersson TR. Coagulation inhibitor levels in pneumonia and stroke: changes
due to consumption and acute phase reaction. J Intern Med 1989; 225: 311-16.
45. Abildgaard U, Lindahl AK, Sandset PM. Heparin requires both antithrombin and extrinsic
pathway inhibitor for its anticoagulant effect in human blood. Haemostasis 1991; 21: 254-7.
46. Nordfang 0, Valentin S, Beck TC, Hedner U. Inhibition of extrinsic pathway inhibitor shortens
the coagulation time of normal plasma and of hemophilia plasma. Thromb Haemostas 1991;
66: 464-7.
47. Goodwin CA, Melissari E, Kakkar VV, Scully ME Plasma levels of tissue factor pathway
inhibitor in thrombophilic patients. Thromb Res 1993; 72: 363-6.
48. Hansen JB, Huseby KR, Huseby NE, Sandset PM, Hanssen TA, Nordoy A. Effect of
cholesterol lowering on intravascular pools of TFPI and its anticoagulant potential in type
II hyperlipoproteinemia. Arterioscler Thromb Vase Bioi 1995; 15: 879-85.
49. Lindahl AK, Abildgaard U, Staalesen R. The anticoagulant effect in heparinized blood and
plasma resulting from interactions with extrinsic pathway inhibitor. Thromb Res 1991; 64:
155-68.
50. Kokawa T, Enjyoji K, Kumeda K, Kamikubo Y, Harada-Shiba M, Koh H, Tsushima M,
Yamamoto A, Kato H. Measurement of the free form of TFPI antigen in hyperlipidemia.
Relationship between free and endothelial cell-associated forms of TFPI. Arterioscler Thromb
Vasc Bioi 1996; 16: 802-8.
51. 0stergaard PB, Beck TC, 0rsted H, Svendsen A, Nordfang 0, Sandset PM, Hansen JB. An
enzyme linked immunosorption assay for tissue factor pathway inhibitor. Thromb Res 1997;
87:447-59.
52. Hansen JB, Sandset PM, Huseby KR, Huseby NE, Nordoy A. Depletion of intravascular
pools of tissue factor pathway inhibitor (TFPI) during repeated or continuous intravenous
infusion of heparin in man. Thromb Haemostas 1996; 76: 703-9.
53. Abumiya T, Enjyoji K, Kokawa T, Kamikubo Y, Kato H. An anti-tissue factor pathway
inhibitor (TFPI) monoclonal antibody recognized the third Kunitz domain (K3) of free-form
TFPI but not lipoprotein-associated forms in plasma. J Biochem (Tokyo) 1995; 118: 178-82.

181
19
Lupus anticoagulant
D. A. TRIPLETT

INTRODUCTION

Lupus anticoagulants (LA) are immunoglobulins (IgG, IgM, IgA or mixtures)

which interfere with one or more of the in-vitro phospholipid-dependent coagu-
lation tests (e.g. activated partial thromboplastin time [APTI] (see Chapter 5),
kaolin clotting time [KCT], dilute prothrombin time (see Chapter 6) [dPT],
dilute Russell viper venom time [dRVVT], and Textarin time [TT]I,2. LA is a
misnomer since the vast majority of individuals with this inhibitor (synonym:
circulating anticoagulant) do not have underlying systemic lupus erythematosus
(SLE). Paradoxically, despite having prolonged in-vitro coagulation tests, the
vast majority of patients do not have clinical bleeding. The presence of LA in
many cases is associated with a predisposition to either venous or arterial
thromboembolic disease 3,4. LA belongs to a family of antibodies which have
collectively been referred to as antiphospholipid antibodies (APA)I,2. However,
this designation is misleading since the true antigenic targets for these antibodies
are plasma proteins which bind to negatively charged phospholipids (PL). The
binding of plasma proteins to injured cellular membranes exposes neoepitope(s).
The immunological response to these neoepitopes results in APA. Anticardiolipin
antibodies (ACA) also belong to the APA family (Table 19.1)1,2,5. Patients with
autoimmune disease (e.g. SLE) often have both LA and ACA 1,2. A number of
plasma proteins have been identified as antigenic targets for APA. Among these
proteins are ~2-glycoprotein I (apolipoprotein H), prothrombin, annexin V,
and proteins C and S.
LA may be encountered in many different clinical conditions. In some cases
they are detected as part of a preoperative coagulation screen utilizing APTT
or KCT testing. Often, requests for LA are generated when patients present
with unexplained venous or arterial thromboembolic events. LA is found in
8-14% of patients with unexplained venous thrombosis, 30% of strokes in

183
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Table 19.1 Antiphospholipid-protein antibodies
(APA)

Lupus anticoagulants (LA)

Anticardiolipin antibodies (ACA)
Reagin
Specific antibodies to anionic PL (?)
Antiphosphatidylserine
Antiphosphatidic acid
Antiphosphatidylinositol
Specific antibodies to neutral PL (?)
Antiphosphatidylethanolamine
Hexagonal phase

patients less than 50 l.ears of age and 5-15% of women with recurrent
spontaneous abortions .7.

LABORATORY APPROACH TO THE DIAGNOSIS OF LA

Unlike many tests in haemostasis, LA is not a discrete entity. LA is a laboratory

phenomenon which may be seen in many different clinical circumstances (e.g.
following infections [viral, bacterial, protozoal, etc.], drug exposure [quinidine,
quinine, procainamide, etc.] or autoimmune disease [SLED. Consequently, the
identification of LA remains a challenge which requires a carefully coordinated
sequential laboratory approach 8,9. The protein targets for the immunoglobulins
expressing LA activity are highly variable from patient to patient. In many
cases there are 'mixtures' of immunoglobulins in a given patient, which recognize
different plasma proteins. The marked heterogeneity both within and between
patients with LA represents a significant challenge to laboratory detection!
diagnosis.
Criteria for the diagnosis of LA have been developed by the Subcommittee
on Lupus AnticoagulantslPhospholipid-dependent Antibodies of the Scientific
and Standardization Committee (SSC) of the ISTH8 . These recommendations
require four sequential steps for LA diagnosis:
1. prolongation of PL-dependent coagulation tests,
2. demonstration of an inhibitor utilizing mixing studies,
3. evidence of PL-dependence of inhibitor, and
4. rule out other confounding coagulopathies (e.g. factor VIII inhibitors, etc.).
Adherence to these essential criteria greatly improves laboratory performance
in LA detection.

STEP 1: SCREENING PROCEDURES

One of the most important aSJ'ects of testing for LA is proper preparation of

platelet-poor plasma (ppp)8.1 . The residual platelet count should be less than
184
 LUPUS ANTICOAGULANT

10 OOO/ml. This is extremely important when plasma is to be frozen and analysed

at a later date 11 . The 'freeze-thaw' rupture of residual platelets in the PPP may
result in masking of LA.
A variety of tests have been used to screen for LA (Table 19.2). Among the
most popular tests are: APTT, dRVVT, KCT, silica clotting time (SCT), dPT.
The SSC Subcommittee originally recommended two different screening tests
(e.g. sensitive APTT and dRVVT). Using this combination approximately
80--85% of patients with LA will be identified 12. By adding a third test (e.g.
dilute PT), the overall sensitivity approaches 100%. It is extremely important
for the laboratory to choose sensitive APTT and dRVVT reagents. There is
wide variability among commercially available reagents I2,13. The abnormal
screening test(s) is used for subsequent steps in evaluation/confirmation of LA.
Thus if an APTT is prolonged, the mixing and confirmatory procedures are
based on the same APTT test system (reagent/instrument combination).

STEP 2: MIXING STUDIES

By definition, circulating anticoagulants are identified by mixing studies utilizing

a source of normal and patient PPP. The source of normal PPP may be either
commercially available lyophilized plasmas or locally prepared fresh plasma II.
Locally prepared PPP is preferred. Variable ratios of patient to normal plasma
have been used. Most commonly, laboratories employ a I: 1 mix of patient and
normal plasma. In cases where the screening test is only minimally prolonged
a 4 part patient, I part normal plasma mix may be preferable. Although the
classical LA is an immediate inhibitor (inhibitory effect is evident following
initial mixing of patient and normal plasmas), approximately one-third of LA
may exhibit time dependencyl4,15. Optimally, a patient/normal mixture should
be incubated over at least 60 min with testing at 30 and 60 min for inhibitory
effect. Conversely, factor VIII inhibitors are typically time-dependent; however,
they may also present with initial immediate inhibitory effect. The differentiation
of LA and factor VIII inhibitor is of critical importance. Most commonly this
arises in the setting of an acquired autoimmune factor VIII inhibitor l6

STEP 3: CONFIRMATORY PROCEDURES

Once an inhibitor has been identified it is necessary to determine if it is

PL-dependent. This step requires the addition of PL to the patient plasma to

Table 19.2 Lupus anticoagulant: screening tests

Intrinsic pathway Extrinsic pathway Final common pathway

APTT dPT dRVVT

KCT Textarin time
SCT Taipan venom time

Abbreviations: APTf, activated partial thromboplastin time; KCT, kaolin clotting time; SCT,
silica clotting time; dPT dilute prothrombin time; dRVVT, dilute Russell viper venom time

185
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
see if there is relative correction of the abnormal clotting times. A number of
sources of PL have been utilized including freeze-thawed platelets (platelet
neutralization procedure), excess PL, platelet vesicles, and hexagonal phase
PL8,17. When utilizing freeze-thawed platelets, or lyophilized platelets, heparin
may give a false-positive result (i.e. platelet factor 4 will neutralize heparin).
This may be a problem in the hospitalized patient in whom the presence of
heparin was not considered when an abnormal clotting study was obtained.
The hexagonal phase PL confirmatory study has proved to be quite sensitive
and specific for LA detection 17 . Hexagonal phase PL is part of an 'integrated
system' utilizing an APTT and incorporating in the sequential steps the addition
of a source of normal plasma as well as polybrene to neutralize heparin, if
present in the sample. Other examples of integrated tests include commercially
available dRVVT systems. These employ a sensitive dRVVT to screen and a
dRVVT reagent which contains high PL concentrations to confirm.

STEP 4: RULE OUT OTHER COAGULOPATHIES

Occasionally, patients are misdiagnosed as having LA when there are other

clinical conditions which need to be considered. Perhaps the most catastrophic
of these is to misidentify a factor VIII inhibitor as a LA. Consequently, factor
assays are indicated in some situations. Typically, patients with factor VIII
inhibitors have a history of clinical bleeding; nevertheless, they may be
misdiagnosed as LA. Factor assays in patients with LA often demonstrate
non-parallelism for factor assay plots (e.g. factors VIII, IX, XI, XII). When
diluting an LA-positive plasma the 'apparent' factor activity will increase with
progressive dilution.

LA STANDARDS AND REFERENCE MATERIALS

There are no available LA standards. Commercial LA-positive controls are

available. These are either frozen or lyophilized plasmas obtained from
LA-positive patients. The performance of these controls will vary depending
upon the choice of reagents used for LA detection. Various external quality
assessment schemes have evaluated laboratory performance of LA diagnosis.
The results have been disappointing due to a lack of adherence to the SSC
Subcommittee recommendations and marked reagent heterogeneity (i.e. variable
sensitivity to LA).

References
I. Triplett DA. Antiphospholipid-protein antibodies: laboratory detection and clinical relevance.
Thromb Res 1995; 78: 1-31.
2. Love PE, Santoro SA. Antiphospholipid antibodies: anticardiolipin and the lupus anticoagulant
in systemic lupus erythematosus (SLE) and in non-SLE disorders. Ann Intern Med 1990;
112: 692-8.
3. Roubey RAS, Hoffman M. From antiphospholipid syndrome to antibody-mediated thrombosis.
Lancet 1997; 350: 1491-3.

186
 LUPUS ANTICOAGULANT

4. Hughes GRY. The antiphospholipid syndrome: ten years on. Lancet 1993; 342: 341-4.
5. Harris EN, Gharavi AE, Boey ML, Patel BM, MacWorth-Young CG, Loizou S, Hughes GRY.
Anticardiolipin antibodies: detection by radioimmunoassay and association with thrombosis
in systemic lupus erythematosus. Lancet 1983; 2: 1211-14.
6. Ginsburg KS, Liang MH, Newcomer L, Goldhaber SZ, Schur PH, Hennekens CH, Stampfer
MJ. Anticardiolipin antibodies and the risk for ischemic stroke and venous thrombosis. Ann
Intern Med 1992; 117: 997-1002.
7. Kittner SJ, Gorelick PB. Antiphospholipid antibodies and stroke: an epidemiological perspective.
Stroke 1992; 23: 119-122.
8. Brandt JT, Triplett DA, A1ving B, Scharrer I. Criteria for the diagnosis of lupus anticoagulants:
an update. Thromb Haemostas 1996; 74: 1185-90.
9. Triplett DA. Screening for the lupus anticoagulant. Res Clin Lab 1989; 19: 379-89.
10. Sletnes K, Graven K, WisloffF. Preparation of plasma for the detection of lupus anticoagulants
and antiphospholipid antibodies. Thromb Res 1992; 66: 43-53.
II. Kaczor DA, Bickford NM, Triplett DA. Evaluation of different mixing study reagents and
dilution effect in lupus anticoagulant testing. Am J Clin Patho11991; 95: 408-11.
12. Johns AS, Charnley L, Ockelford PA, Pattison NS, McKay EJ, Corkill M, Hart H. Comparison
of tests for the lupus anticoagulant and antiphospholipid antibodies in systemic lupus
erythematosus. Clin Exp Rheumatol1994; 12: 523-6.
13. Schjetlein R, Wisloff F. An evaluation of two commercial test procedures for the detection
of lupus anticoagulant. Am J Clin Patho11995; 103: 108-11.
14. Triplett DA, Brandt JT, Maas RL. The laboratory heterogeneity of lupus anticoagulants.
Arch Pathol Lab Med 1985; 109: 946-51.
15. Clyne LP, White PF. Time dependency of lupus-like anticoagulants. Arch Intern Med 1988;
148: 1060-3.
16. Triplett DA. Simultaneous occurrence of lupus anticoagulant and factor VIII inhibitors. Am
J Hematol1997; 56:195-6.
17. Triplett DA, Barna LK, Unger GA. A hexagonal (II) phase phospholipid neutralization assay
for lupus anticoagulant identification. Thromb Haemostas 1993; 70: 787-93.

187
20
Heparin cofactor II
S. J. BAUMAN and F. C. CHURCH

INTRODUCTION

Heparin cofactor II (HCII) is a glycosaminoglycan-binding serine protease

inhibitor (serpin) found in plasma at a concentration of approximately 1.2
/lIIloVL 1. The putative physiological role of HCII is as an inhibitor of the serine
protease thrombin. To a lesser extent HCII is able to inhibit chymotrypsin and
cathepsin G.
Historically, HCII was first definitively isolated by Briginshaw and Shanberge
as a distinct protein from antithrombin lIe. They referred to it as heparin
cofactor B. In 1982 Tollefsen and co-workers purified and characterized a
thrombin inhibitor that was heparin-dependent. They were the first group to
refer to this protein as HCII 3. Later the same year, Wunderwald and co-workers
purified HCII and named it antithrombin BM4. In 1983, while studying a
thrombotic tendency within a family, Griffith et al. identified a second heparin-
binding antithrombin and called it heparin cofactor AS. Ragg rediscovered
HCII in 1986, yet referred to it as human leuserpin-26. Despite the many names,
this protein is most commonly called HCII.

STRUCTURAL ASPECTS
HCII is a 480 amino acid glycoprotein with a molecular weight of approximately
66000. HCII contains three possible N-glycosylation sites. At the amino terminal
end of the protein are two sulphated tyro sines. HCII contains three cysteines,
all of which are available for sulphydryl modification, indicating that none is
involved in disulphide bonds. The gene for HCII is located on chromosome
22. The protein is expressed in hepatocytes and secreted into the blood 7 ,8.
The reactive site loop of the serpin is the structure recognized by the serine
189
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
protease as a substrate. When the protease attempts to cleave this 'pseudo-
substrate' it becomes trapped in a stable 1:1 complex with the inhibitor. Hell
is unique among thrombin inhibitors because its reactive site bond is made of
the dipeptide Leu444-Ser445 as opposed to the usual Arg-Ser which are
designated as the PI-PI' residues9 . In particular, thrombin prefers Arg at the
PI position lO • Hell is a poor thrombin inhibitor in the absence of
glycosaminoglycan (2 x 104 M- I min-I). However, when glycosaminoglycans
such as heparin or dermatan sulphate are introduced to the inhibition reaction,
the rates increase dramatically to 3 x 108 and 6 x 108 M- I min-I, respectivelyli.
The glycosaminoglycan binding region of Hell has been localized to the
D-helix of the molecule. The D-helix is made up of residues Lysl71 through
Phe195. There are six basic residues within this helix which form a positively
charged surface-exposed structure (Lys173, Arg184, Lys185, Arg189, Argl92,
and Arg193). Although not well defined, it seems as though the residues at the
beginning of the helix (Lys173, Arg184, and Lys185) are involved in binding
to heparin, while the latter amino acids (Lys185, Arg189, Arg192, and Arg193)
are involved in dermatan sulphate binding I2- 15 •
Unlike other serpins, Hell contains a unique amino terminal structure. The
region between Gly54 and Asp75 contains two stretches of acidic amino acids.
This region, called the 'acidic domain', is similar to the carboxyl terminus of
the leech anticoagulant hirudin and the amino terminus of the tethered-ligand
thrombin receptor. Both of these similar structures are known to bind anion-
binding exosite-I of thrombin. An intramolecular interaction with the Rositively
charged D-helix and the acidic domain of Hell has been proposed 6,17.
The acidic domain is an integral component of the proposed mechanism of
glycosaminoglycan accelerated thrombin inhibition by He1l 18 . When glyco-
saminoglycan is present in the inhibition reaction, it binds the D-helix of Hell
and displaces the acidic amino terminal domain. In addition to the interaction
between the reactive site loop of Hell and the active site region of thrombin,
the acidic domain is free to interact with anion-binding exosite-I of thrombin.
The multiple points of contact between Hell, thrombin and the
glycosaminoglycan are believed to be responsible for the increase in inhibition
rates.

PATHOPHYSIOLOGICAL ASPECTS

Although a definitive role for Hell has not been found in vivo, the involvement
of HCII, or lack thereof, has been studied in many different clinical scenarios.
Two genetic variations have been described. Hell Oslo was identified by Blinder
et al. 5. The Hell Oslo gene contained a point mutation at base 651 that
resulted in an Arg189- His mutation. This protein lost all ability to accelerate
thrombin inhibition in the presence of dermatan sulphate. This patient was
identified from a pool of blood donors as having decreased Hell activity in
the presence of dermatan sulphate but not heparin. Otherwise the patient was
asymptomatic. Hell Awaji was a result of a thymine insertion within the gene
that resulted in a frame-shift mutation l9 • The mutated gene produced a truncated
protein that was rapidly degraded. The patient with Hell Awaji presented with
190
 HEPARIN COFACTOR II

a thrombotic tendency, although his sister, who was positive for the same
mutation, was asymptomatic.
In addition to genetic anomalies, changes in HCII levels have been
documented in many clinical conditions. Levels of HCII were reported to be
decreased in patients with disseminated intravascular coagulation, hepatic
failure, acute pancreatitis, human immunodeficiency virus, and diabetes type
I, as well as after elective surgeryl,20--23. Patients with chronic renal failure on
haemodialysis, and those with nephrotic syndrome, reportedly have increased
levels of HCII24,25. Females have higher levels of HCII than males. The levels
of HCII increase during pregnancy or oral contraceptive use26 . Preterm infants
have markedly decreased levels of HCII compared to term infants, who have
approximately 50% of adult HCII levels 27 .

Reference range
The normal concentration of heparin cofactor II for a healthy adult is 1.2 ±
0.4 (mean 2 SO) or approximately 80 IJg/mll. Healthy, full-term infants have
heparin cofactor II concentrations of approximately 50% of normal adult
values27 .

QUANTITATION

HCII can be assayed either when purified from plasma or directly in plasma
in a number of different ways. Antigen for HCII can be detected by elli.-ryme-
linked immunosorbent assay (ELISA), rocket immunoelectrophoresis (more
commonly referred to as Laurell rockets), and by Western immunoblotting
techniques. HCII activity can also be studied by looking at the ability to inhibit
thrombin in the presence of the glycosaminoglycan dermatan sulphate.

Methods of assay
ELISA for Hell
Principle
The principle of the HCII ELISA is to coat the microtitre plate with a polyclonal
antibody to HCII to capture the protein from the diluted sample. Next, an
HCII antibody that is conjugated to horseradish peroxidase is incubated in
each well. This antibody will react with the captured protein. A final reaction
is performed within each well to quantitate the amount of conjugate present
by its ability to develop a substrate. By comparing the amount of conjugate
in a series of standards with the test sample dilutions, an accurate amount of
HCII antigen can be quantitated. This assay is based on the method described
by Toulon et al. 28 .
Materials
1. Carbonate buffer: this buffer is used for making dilutions of samples. It is
50 mmollL sodium carbonate, pH 9.6.
191
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

2. Wash buffer: this buffer is used for all wash steps. It is made of the carbonate
buffer with ISO mmoVL NaCI and 0.1% polyoxyethylene-sorbitan mono-
laurate (Tween 20).
3. Blocking buffer: this buffer is used to block any available binding sites within
each well that have not bound capture antibody. It consists of carbonate
buffer with 10 mglml bovine serum albumin.
4. Dilution buffer: this buffer is used to dilute standard and test samples. It
consists of SO mmoVL sodium phosphate, pH 7.S, ISO mmoVL NaCl, 0.1 %
Tween 20, and 1 mglml bovine serum albumin.
S. Substrate reactants: the peroxidase reaction is done by adding a solution of
tetramethylbenzidine and hydrogen peroxide prepared according to the
instructions in the tetramethylbenzidine (TMB) Peroxidase EIA Substrate
Kit from BioRad laboratories.
6. Stop solution: a 1 moVL solution of H2S04 will stop the peroxidase reaction.
7. Capture antibody: this is a polyc1onal antibody against HCII. It can be found
commercially (American Diagnostica) or produced using methods described
previousli9 .
8. Conjugated antibody: this antibody is the capture antibody conjugated to
horseradish peroxidase using N-succinimidyl 3-(2-piridylthio)proprionate
(SPDP), as described by Carlsson et al. 3o, or the periodate method of Nakane
et al. 3l •
9. Assay plates: these plates must be of high quality to ensure homogeneity,
coating performance and uniform background absorbance. Two possible
sources are immunoassay plates from Titertek or polystyrene microtitre
plates type I from Nunc.
Method
1. Coating: coating of microtitre plates is done with 200 J.ll/well of 10 J.lglL
capture antibody to HCII diluted in carbonate buffer. Seal plates with
parafilm and incubate for 12 h at room temperature. The coating solution
is carefully removed from the wells and the plate is washed S x 3 min with
200 J.lVwell Wash Buffer. The plates are then blocked with 200 J.lVwell of
blocking buffer for 2 h at room temperature, after which the plates are
washed S x 3 min with 200 J.LVwell wash buffer.
2. Test samples: using the dilution buffer test samples were diluted 1:2000 and
1:4000; 200 J.Ll of diluted samples are added to the wells and incubated for
2 h at room temperature in a moist chamber. Wells are emptied and washed
5 x 3 min with 200 J.lVwell wash buffer. The conjugated antibody is then
added 200 J.LVwell diluted to a concentration of S J.lglml with dilution buffer
and incubated for 2 h at room temperature. Finally the plate is washed S x
3 min with 200 J.LVwell wash buffer. The peroxidase reaction is performed
using 200 J.LVwell of the tetramethylbenzidine and hydrogen peroxide solution
made according to the instructions provided with the TMB Peroxidase EIA
Substrate Kit. This solution incubates for 2 min at room temperature and
is quenched using 100 J.LVwell of stop solution. The chromogenic activity in
each well is measured at 4S0 nm on a microplate reader.
3. Calibration: calibrate this assay with pooled normal plasma samples diluted
in dilution buffer. The standard curve is created with 1: 1000, 1:2000, 1:4000,
192
 HEPARIN COFACTOR"

1:8000,1:16000,1:32000 dilutions. Each dilution is tested in duplicate. The

HCII concentration of each test sample is read from the standard curve
plotted on a double-logarithmic scale. The concentration of HCII is reported
in unitslml; 1 unitlml is the concentration of HCII present in 1 ml of pooled
normal plasma.

Laurell immunoelectrodiffusion
Principle
An agarose gel is made of uniform thickness containing a polyclonal antibody
to HCII. Exact volumes of pooled normal plasma dilutions and test plasma
are applied to the gel and electrophoresed. As the antibodies and samples
migrate through the gel, a rocket-shaped precipitation zone of antigen-antibody
complexes forms. Unbound HCII within the tip of the rocket redissolves the
complexes at the leading front as they migrate, until all of the free antigen is
completely complexed. This assay is based on the methods described by Laure1l32
and Tran and Duckert for HC1I 33 .

Materials
1. Barbital buffer: this buffer is used to make the gel used in this procedure. It
consists of sodium barbital (S,S'-diethyl-barbituric acid) buffer with an ionic
strength of 0.07 and 2 mmoVL calcium lactate at pH 8.6.
2. 1% agarose gel: this is used for electrophoresis of the standards and the test
samples; 1% (w/v) agarose is dissolved in the barbital buffer.
3. Hell antibody: this antibody is added to the agarose gel when the temperature
is approximately 40-45 °C. HCII polyc1onal antibodies are available
commercially, or can be produced using the method described in ref. 28.
The lowest amount of antibody needed to give well-defined rockets should
be used. This amount will usually be from 0.5% to 2% (v/v) of antibody in
the gel. The amount will, however, differ depending on the specific antibody
preparation used.

Method
1. Electrophoretic set-up: this method was first described by Laurell in 1966.
It entails forming a gel mould with two rectangular glass plates separated
by I.S mm spacer on each side and along the bottom. Mter securing the
apparatus the mould is filled with the 1% agarose with HCII antibody liquid
mixture (approximately 40 0q. Mter at least 30 min the top plate is removed
and small uniform holes are cut in the gel on a parallel line approximately
2 em from one of the long edges.
2. Electrophoresis of samples and standards: plasma samples are diluted 1:2
and standard samples of pooled normal plasma are made undiluted and
diluted 3:4, 2:4 and 1:4 in isotonic NaCl. The gel is arranged so the application
holes are at the cathodic end of the gel. Apply each sample of standard
dilution to a separate well in exactly the same volume (5 JJ.l). The gel is run
193
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
at 10 V/cm for 2-10 h depending on the amount of antigen versus antibody
applied to the gel. When the gel has completed running it can be stained
with Coomassie blue. The ideal height of the rockets is 2-5 cm.
3. Measurement and evaluation: a standard curve is drawn of the various
dilutions of pooled normal plasma based on their dilution factor versus the
height in millimetres of the associated rocket (measured from the centre of
the application hole to the tip of the rocket). The height of the test sample
plasma rockets can be used to deduce a percentage of HClI compared to
the pooled normal plasma standard curve.

Hell activity with dermatan sulphate

Principle
The determination of dermatan sulphate-accelerated HClI activity uses the
ability of HClI to inhibit thrombin, a 'trypsin-like' serine protease. Inhibition
of thrombin by HClI within a plasma sample can be exclusively measured by
the addition of the glycosaminoglycan dermatan sulphate, which specifically
accelerates the inhibition of thrombin by HClI. The amount of thrombin
amidolytic activity remaining (measured by absorbance of a chromogenic
substrate at 405 nm) after a specific incubation time with HClI and dermatan
sulphate is measured for a series of standards and test sample dilutions. By
comparing these values an accurate evaluation of the test sample can be made.
This assay is based on the method described by Tollefsen and Pestka I.

Materials
1. Dilution buffer: this buffer is used for making dilutions of plasma samples;
it comprises 20 mmollL Tris-HCI, pH 7.4, 150 mmollL NaCl, 0.1%
polyethylene glycol (pEG).
2. Reaction buffer: this buffer is used to ensure that the dermatan sulphate-
accelerated inhibitory activity of HClI is what is being measured. It consists
of dilution buffer with 4 ~g/ml polybrene, 133 ~g/ml dermatan sulphate
(nitrous acid-treated), 8 mmollL EDTA, and 1 mg/ml bovine serum albumin.
3. Thrombin solution: human a-thrombin is the enzyme used to measure HClI
inhibitory activity. The solution to be used in the assay is made up to
approximately 10 NIH Ulml in dilution buffer.
4. Chromogenic substrate solution: the chromogenic substrate solution is used
to measure the amount of thrombin amidolytic activity in a given sample.
The chromogenic substrate used in this assay is tosyl-Gly-Pro-Arg-p-
nitroanilide (Chromozym TH-Boehringer Mannheim). It is made up to 750
~mollL in dH 20.
5. Stop solution: glacial acetic acid is used to stop the hydrolysis of the substrate.

Method
1. Inhibition reaction: all reactions are carried out in 1.7 ml polypropylene
bullet tubes. Test plasma samples (1-4 ~l) are diluted to a 40 ~ volume with
194
 HEPARIN COFACTOR II

dilution buffer. Standard samples are prepared by diluting 0, 1,2, 3, and

4 ~ of pooled normal plasma to a 40 ~ volume with dilution buffer. Each
test plasma sample and standard sample dilution is prepared in duplicate.
Using reaction buffer the volume is brought to 100 Ill. The assay is started
by the addition and mixing of 20 ~ of the thrombin solution. After a 60 s
incubation, 60 1-11 of the chromogenic substrate solution is added, mixed
and incubated for 60 s. Finally 60 1-11 of the stop solution is added to the
reaction to terminate the reaction. A blank is prepared alongside each sample
(standard and test plasma), exactly as described above except with a
substitution of dilution buffer for the thrombin solution. Since some of the
glycosaminoglycans precipitate out of solution when the pH is decreased,
precipitates are removed by centrifugation of the bullet tubes for 10 min.
The absorbance of the supernatant is measured at 405 nm. The absorbance
of the blank corresponding to each standard or test sample is subtracted
from the absorbance of each sample to create a corrected absorbance value.
2. Evaluation: the corrected absorbances of each duplicated standard sample
dilution are averaged and plotted on a curve of absorbance at 405 nm
(residual thrombin amidolytic activity) versus the amount of plasma in the
dilution. The duplicate corrected absorbances of each test plasma sample
are averaged and compared to the standard curve to find the percentage of
activity compared to the pooled normal plasma.

Immunoblotting of Hell

Principle
Samples are run using sodium dodecyl sulphate polyacrylamide gel electro-
phoresis (SDS-PAGE) methodology and qualitatively compared to a protein
standard or a plasma-purified standard HCII sample to confirm the correct
molecular weight. The proteins are electrophoretically transferred to a nitrocel-
lulose membrane (or polyvinylidine difluoride membrane). The membrane is
then probed with an anti-HCII polyclonal antibody. If the antibody is conjugated
to horseradish peroxidase the activity can be developed and used to localize
the protein. This assay is based on the methods described by Laemmle4 and
Toulon et a1. 35 • Note: This method can also be used to qualitate complexes
formed by HCIl with thrombin and glycosaminoglycans.

Materials:
1. TBS (Tris buffered saline) buffer: this is the base buffer for all steps in
probing the transfer membrane. It is made of 20 mmoVL Tris-HCI, pH 7.5,
and 50 mmoVL NaCl.
2. Blocking buffer: this buffer will block all binding sites on the transfer
membrane where protein was not transferred. It is made up with TBS buffer
containing 5% (w/v) instant non-fat dried milk.
3. Wash buffer: this buffer is used to wash the blot after each step. It is made
of TBS buffer with 0.05% (v/v) Tween-20.
4. Primary (anti-Hell) antibody solution: a polyclonal antibody against HCII
195
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

will bind specifically to Hell transferred to the transfer membrane. This

antibody is made up to 0.1 % (v/v) in TBS buffer containing 1% (w/v) gelatin,
although the antibody concentration may have to be adjusted depending on
the specific preparation of antibody used. Hell polyc1onal antibodies are
available commercially, or can be produced using the method described
previously29.
5. Secondary antibody solution: this antibody is used to detect the primary
antibody binding. It can be obtained from a variety of sources. It is an
antibody that is directed against the species in which the primary antibody
was produced. In addition, it is conjugated to horseradish peroxidase so
that it can be detected enzymatically. It is diluted according to the manu-
facturer's specifications in TBS buffer.
6. Developing solution: this solution is used for detection of the secondary
antibody. The horseradish peroxidase conjugate can be developed with a
solution of 3,3'-diaminobenzidine (0.5 mglml) in methanol containing
0.012% (v/v) hydrogen peroxide.

Method
1. Gel electrophoresis: the plasma samples and standards can be run either
reduced or unreduced. Using SDS-PAGE methodolo@;; run protein out on
a 7.5% polyacrylamide gel containing 0.1% (v/v) SDS 4.
2. Electrophoretic transfer: the protein from the SDS gel is electrophoretically
transferred to the transfer membrane using the methods described by Towbin
et al. 36 •
3. Immunoblotting procedure: the transfer membrane is blocked using blocking
buffer for 2 h at room temperature. Next the blot is washed 3 x 10 min with
wash buffer. The blot is incubated at room temperature in primary antibody
solution for 1.5 h. The wash step is repeated. The blot is incubated in the
secondary antibody solution for 1.5 h. The wash step is again repeated.
Finally, the peroxidase activity is detected by applying the development
solution and observing colour development.

ADDITIONAL METHODS OF ASSAY

Hell can be obtained commercially or purified from plasma using the method
of Griffith et al. 37 or Tollefsen et al. 3. Hell can also be purified by immuno-
isolation as described by Toulon et al.2O. Purified Hell can be used as a standard
in all quantitative assays to calculate exact molar amounts of protein present
in patient samples.
Additional assays to evaluate HCII activity have been described using purified
Hell. These include assays for Hell inhibition of chymotrypsin, interaction
of Hell with neutrophil elastase and cathepsin G, leukocyte chemoattractant
activity, and inhibition of thrombin in the absence or presence of glyco-
saminoglycan (heparin or dermatan sulphate)1l,38-40.
196
 HEPARIN COFACTOR II
References
1. Tollefsen OM, Pestka CA. Heparin cofactor II activity in patients with disseminated
intravascular coagulation and hepatic failure. Blood 1985; 66: 769-74.
2. Briginshaw GF, Shanberge IN. Identification of two distinct heparin cofactors in human
plasma: II. Inhibition of thrombin and activated factor X. Thromb Res 1974; 4: 463-77.
3. Tollefsen OM, Majerus Ow, Blank MK. Heparin cofactor II: purification and properties of
a heparin-dependent inhibitor of thrombin in human plasma. I Bioi Chem 1982; 257: 2162-9.
4. Wunderwald P, Schrenk WI, Port H. Antithrombin BM from human plasma: an antithrombin
binding moderately to heparin. Thromb Res 1982; 25: 177-91.
5. Griffith MI, Carraway T, White GC, Oombrose FA. Heparin cofactor activities in a family
with hereditary antithrombin III deficiency: evidence for a second heparin cofactor in human
plasma. Blood 1983; 61: Ill-18.
6. Ragg H. A new member of the plasma protease inhibitor gene family. Nucl Acids Res 1986;
14: 1073-87.
7. Church FC, Shirk RA, Philips IE. Heparin Cofactor II. In: High KA, Roberts HR, eds.
Molecular basis of thrombosis and hemostasis. New York: MarcelOekker, 1995; 379-92.
8. Tollefsen OM. Insight into the mechanism of action of heparin cofactor II. Thromb Haemostas
1995;74: 1209-14.
9. Schecter I, Berger A. On the size of the active site in proteases I. Papain. Biochem Biophys
Res Commun 1967; 27: 157-62.
10. Griffith MI, Noyes CM, Tyndall lA, Church Fe. Structural evidence for leucine at the reactive
site of heparin cofactor II. Biochemistry 1985; 24: 6777-82.
II. Rogers SI, Pratt CW, Whinna HC, Church Fe. Role of thrombin exosites in inhibition by
heparin cofactor II. I Bioi Chern 1992; 267: 3613-17.
12. Whinna HC, Blinder MA, Szewczyk M, Tollefsen OM, Church FC. Role of lysine 173 in
heparin binding to heparin cofactor II. I Bioi Chern 1991; 266: 8129-35.
13. Whinna HC, Church FC. Characterization of a synthetic peptide from the glycosarninoglycan
binding site of heparin cofactor II. Letts Pept Sci 1994; I: 3-8.
14. Blinder MA, Tollefsen OM. Site-directed mutagenesis of arginine 103 and lysine 185 in the
proposed glycosaminoglycan-binding site of heparin cofactor II. I Bioi Chern 1990; 265:
286-91.
IS. Blinder MA, Andersson TR, Abildgaard U, Tollefsen OM. Heparin cofactor II Oslo. Mutation
of Arg-189 to His decreases the alfInity for dermatan sulfate. I Bioi Chem 1989; 264: 5128--33.
16. Blinder MA, Marasa IC, Reynolds Ch, Oeaven LL, Tollefsen OM. Heparin cofactor II:
cONA sequence, chromosome localization, restriction length polymorphism and expression
in Escherichia coli. Biochemistry 1988; 27: 752-9.
17. van Oeerlin VMO, Tollefsen OM. The N-terminal acidic domain of heparin cofactor II
mediates the inhibition of a-thrombin in the presence of glycosarninoglycans. I Bioi Chern
1991;266:20223-31.
18. Sheehan JP, Tollefsen OM, Sadler IE. Heparin cofactor II is regulated allosterically and not
primarily by template effects. Studies with mutant thrombins and glycosarninoglycans. I BioI
Chern 1994; 269: 32747-51.
19. Kondo S, Tokanaga F, Kario K, Matsuo T, Koide T. Molecular and cellular basis for type I
heparin cofactor II deficiency (heparin cofactor II Awaji). Blood 1996; 87: 1006-12.
20. Toulon P, Chadeuf G, Bouillot IL et al. Involvement of heparin cofactor II in chymotrypsin
neutralization and in the pancreatic proteinase-antiproteinase interaction during acute
pancreatitis in man. Eur I Clin Invest. 1991; 21: 303-9.
21. Toulon P, Larnine M, Ledjev I et al. Heparin cofactor II deficiency in patients infected with
the human immunodeficiency virus. Thromb Haemostas 1993; 70: 730-5.
22. Ouboscq C, Quintana I, Barros I, Kordich L. Heparin cofactor II in diabetic patients. Blood
Coag Fibronol. 1994; 5: 201-4.
23. Andersson TR, Berner NS, Larsen ML, Odegaard OR, Abildgaard U. Plasma heparin cofactor
II, protein C and antithrombin in elective surgery. Acta Chir Scand 1987; 153: 291-6.
24. Toulon P, Iacquot C, Capron L, Frydman MO, Vignon 0, Aich M. Antithrombin III and
heparin cofactor II in patients with chronic renal failure undergoing regular hemodialysis.
Thromb Haemostas 1987; 57: 263-8.

197
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
25. Toulon P, Gandrille S, Remy P, Chadeuf G, Jouvin MH, Aiach M. Significance of high levels
of heparin cofactor II in the plasma and urine of adult patients with nephrotic syndrome.
Nephron 1992; 60: 176-80.
26. Mackie IJ, Segal H, Burren T et af. Heparin cofactor II levels are increased by the use of
combined oral contraceptives. Blood Coag Fibrinol1990; 1: 647-51.
27. Andersson TR, Bangstad H, Larsen ML. Heparin cofactor II, antithrombin and protein C
in plasma from term and preterm infants. Acta Paediatr Scand 1988; 77: 485-8.
28. Toulon P, Costa JM, Amiral 1. An enzyme-linked immunosorbent assay for heparin cofactor
II (HCII). Application to the measurement of HCII in clinical materials. Clin Chim Acta
1992; 205: 65-73.
29. Harlow E, Lane D, eds. Antibodies: A laboratory manual. Cold Spring Harbor: Cold Spring
Harbor Laboratory, 1988; 726.
30. Carlsson J, Orevin H, Axen R. Protein thiolation and reversible protein-protein conjugation.
N- succinimidyI3-(2-pyridyldithio) proprionate, a new hetero-bifunctional reagent. Biochem
J 1978; 173: 723-37.
31. Nakane PK, Kawaoi A. Peroxidase-labeled antibody: a new method of conjugation. J
Histochem Cytochem 1974; 22: 1084-91.
32. Laurell C-B. Quantitative estimation of proteins by electrophoresis in agarose gel containing
antibodies. Anal Biochem 1966; 15: 45-52.
33. Tran TH, Duckert F. Heparin cofactor II determination-levels in normals and patients with
hereditary antithrombin III deficiency and disseminated intravascular coagulation. Thromb
Haemostas 1984; 52: 112-16.
34. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage
T4. Nature 1970; 227: 680--5.
35. Toulon P, Chadeuf G, Bouillot JL et af. Involvement of heparin cofactor II in chymotrypsin
neutralization and in the pancreatic proteinase-antiproteinase interaction during acute pacreatitis
in man. Eur J Clin Invest 1991; 21: 303-9.
36. Towbin HJ, Staehelin T, Gordon 1. Electrophoretic transfer of proteins from polyacrylamide
gels to nitrocellulose sheets. Procedure and some applications. Proc Nat! Acad Sci 1979; 76:
4350--4.
37. Griffith MJ, Noyes CM, Church Fe. Reactive site peptide structural similarity between heparin
cofactor II and antithrombin III. J Bioi Chem 1985; 260: 2218-25.
38. Church FC, Noyes CM, Griffith MJ. Inhibition of chymotrypsin by heparin cofactor II. Proc
Nat! Acad Sci USA 1985; 82: 6431-4.
39. Pratt CW, Tobin RL, Church Fe. Interaction of heparin cofactor II with neutrophil elastase
and cathepsin G. J Bioi Chem 1990; 265: 6092-7.
40. Hoffman M, Pratt CW, Corbin LW, Church FC. Characteristics of the chemotactic activity
of heparin cofactor II proteolysis products. J Leuk Bioi 1990; 48: 156-62.

198
21
Fibrinopeptide A (FPA)
A. HAEBERLI

INTRODUCTION

Fibrinopeptide A (FPA) is of particular interest, because it is released whenever

thrombin converts fibrinogen to fibrin. Human FPA is a 16 amino acid peptide
with a molecular weight of 1540, and it is cleaved from the N-terminal part of
the An-chain of fibrinogen by the action of thrombin. Since the fibrinogen
molecule has a dimer structure, two identical FPA molecules can be cleaved
from each fibrinogen molecule!.

PATHOPHYSIOLOGY

Any increased FPA concentration in plasma represents fibrin formation due

to thrombin activity and thus activation of the coagulation. Since, with the
measurement of FPA, a product of the last step of the activation of the coagu-
lation is measured, the determination of FPA may give an overall evaluation
of the activation of the coagulation cascade.
The measurement of FPA by radioimmunoassay has been introduced by
Nosse1 and co-workers 2,3. Since that time the radioimmunoassay, and later
the enzyme linked immunoassay4,5, have been widely used, and they are
considered by many authors as important and significant tests for the diagnosis
of prethrombotic and thrombotic states6-8.
The major diagnostic applications to date are the detection of acute deep
venous thrombosis 9-13, disseminated intravascularcoagulation lO,14,15, pulmon~
thrombosis and lun:§ embolism I6 ,17, angina and myocardial infarction l8- 2 ,
cerebral infarction28, and for the monitoring of anticoagulant theraprO,23,30-35.
Quite a number of reports have become available on the activation or liberation
of thrombin during fibrinolytic therapy of acute myocardial infarction36-40.
The determination of FPA has also been used to follow activation of the
199
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
coagulation in cancer41 --46, during pregnancy47, in diabetes48- 5o , in liver
cirrhosis51- 53. and during intake of oral contraceptives or oestrogen therap~57.
The activation of coagulation was also followed by measurinJ§ different activation
markers such as FPA and others during physical exercise .59.

METHODS OF ASSAY

Since normal concentrations are around or below 2 ng/ml « 1.3 nmoVL) and
pathological values very often in the range of 3-20 ng/ml (2-13 nmoVL), the
only sufficiently sensitive methods are radioimmunoassay (RIA) or enzyme-
linked radioimmunoassay (ELISA). Nossel and co-workers were the first to
develop a RIA for the determination of FPA2,3. The antibodies used by Nossel
became available to a broader community and therefore the number of publica-
tions studying thrombotic and prethrombotic activities in different diseases by
means of FPA measurements increased enormously. Since 1971 several modifica-
tions of the original RIA of Nosse1 have been published (e.g. refs 60 and 61).
Some of these modifications have been introduced into the RIA procedure
recommended below.
Two commercial RIA kits have been available for many years. The RIA-mat
FPA, originally manufactured by Mallinckrodt, then prepared and sold by
Byk-Sangtec (Dietzenbach, Germany) was a ready to use kit with all reagents
including the 125 labelled Tyr-FPA; it has since been discontinued.
The other RIA is prepared by Imco. Imco supplies only three vials containing
anti-FPA antiserum, FPA standard and Tyr-FPA for iodination with 1251. The
iodination of Tyr-FPA, all dilutions of the standards and the antiserum, as
well as all the solutions necessary for the performance of the RIA, have to be
prepared. It is at present not clear whether Imco will continue to supply the
reagents.
In 1980 a solid-phase ELISA for the measurement of FPA was published4,5.
This ELISA test is commercially available as a complete kit, Asserachrom FPA,
supplied by Diagnostica Stago. It is this assay which is now most frequently
used for the determination of FPA. Unfortunately it has recently been dis-
continued.
The FPA-ELISA distributed by Diagnostica Stago was rather unique in its
concept, and might not be easy to set up with reagents still available; it will
therefore be discussed only very briefly.
Since the Mallinckrodt kit RIA kit is no longer commercially available, the
methodological section below will preferentially deal with the RIA method as
described by Imco. The reagents supplied by Imco can easily be replaced by
reagents distributed by other companies. A list of companies supplying
anti-FPA-antiserum, fibrinopeptide A as standard and Tyr-fibrinopeptide A
for iodination is given at the end of this chapter.

PRECAUTIONS

Although the determination of FPA is an interesting method to document

activation of coagulation with the formation of fibrin, the method has several
200
 FIBRINOPEPTIDE A

important biological and methodological disadvantages which should be

considered:
1. The biological half-life of FPA is in the order of 3-5 min 3; therefore any
determination of the FPA concentration represents only the activity of a
short time interval. Several authors have tried to overcome this problem
with the determination of the FPA concentration in urine samples, thereby
obtaining an estimate of the activation over a longer time period62~6.
2. The collection of blood, and anticoagulant used, are extremely important.
Any incorrect procedure may easily lead to falsely elevated FPA concentra-
tions.
3. Most antibodies used so far for FPA determination also react with the
N-terminal sequence of the An-chains of the intact fibrinogen molecule,
thus giving rise to falsely elevated FPA values. For this reason fibrinogen
must be removed completely from the plasma before the determination of
FPA can be performed.

BLOOD SAMPLING AND PROCESSING

Blood is collected by flawless venipuncture through a 21-gauge needle at a

stasis of 50 mmHg into a syringe or evacuated tube (e.g. Vacutainer) or by
dripping into an open tube containing the anticoagulant solution. Artificially
elevated FPA values may easily occur after rough or repeated venipuncture, or
if the blood is not immediately mixed with the anticoagulant. Blood collection
into citrate is definitely not recommended. Instead special anticoagulant solu-
tions must be used67 . Two anticoagulant solutions for blood collection are
recommended:
1. A home-made anticoagulant solution: PPACK (prolyl-phenyl-arginyl-
chloromethylketone), 280 Ilg dissolved in I m1 of citrate. This anticoagulant
is used in a regular way, namely I volume anticoagulant + 9 volumes of
blood. PPACK is a very potent thrombin inhibitor, and used in this high
concentration recommended it also inhibits other serine proteases in sufficient
potency (including plasmin). The disadvantage of PPACK is its instability
in water. Therefore the anticoagulant solution has to be freshly made
immediately before use, or it must be stored frozen at -70°C until use.
Another excellent possibility is to lyophilize the anticoagulant in the syringe
(e.g. Monovette, Sarstedt) and to add water just before use. The lyophilized
anticoagulant is stable for at least 6 months. With the use of this anticoagulant
the collection of blood is only allowed to be made into evacuated tubes (e.g.
Vacutainer, Becton-Dickinson), or into a closed system with a diaphragm
separating the collecting tube from the needle (e.g. Monovette, Sarstedt),
both tubes containing the anticoagulant. Collection with regular syringes
containing the potentially hazardous anticoagulant is strictly forbidden.
2. Heparin-aprotinin anticoagulant. Heparin (e.g. Liquemin, Roche) and
aprotinin (e.g. Trasylol, Bayer) are dissolved in 0.9% sterile NaCI to a final
concentration of 1000 U heparin and 1000 U aprotinin per millilitre of
anticoagulant. An 0.5 m1 aliquot of this anticoagulant is used for the
201
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
collection of 4.5 ml of blood. With the use of this anticoagulant blood can
easily be collected, without any risk, into 5 ml syringes containing the
anticoagulant. This anticoagulant is recommended by Imco.
It is important that the blood is mixed quickly and thoroughly with the
anticoagulant and cooled immediately to 4 DC in an ice-bath. The plasma is
separated by centrifugation at 2000g for 20 min at 2-4 DC within 1 h after
collection. If not assayed immediately the plasma samples have to be stored
frozen at -20 DC or preferably lower temperature.

REMOVAL OF FIBRINOGEN FROM PLASMA

Unfortunately the anti-FPA antibodies supplied by the companies bind to a

considerable extent to the N-terminal sequence of the An-chains of intact
fibrinogen, thus giving rise to erroneously very high FPA values. Therefore
fibrinogen has to be removed from the plasma samples as completely as possible.
The adsorption of fibrinogen with bentonite has proved to be a satis-
factory method. This procedure was originally introduced by Kockum and
Frebelius61 • It is the method recommended in the Imco FPA RIA procedure.
In our experience it is the best method to eliminate fibrinogen almost
quantitatively from plasma. It is very reproducible and the recovery of FPA is
90% or more. It is highly recommendable. It should be noted that the bentonite
adsorption of fibrinogen performs well only with plasma and not with aqueous
fibrinogen solutions.

PRINCIPLE AND ASSAY CHARACTERISTICS OF THE IMCO FPA

RIA

The FPA RIA supplied by Imco is a competitive binding assay. It is based on

a competition between unlabelled FPA (standard or unknown) and the 1251
labelled Tyr-FPA (=tracer) for the limited amount of antibody-binding sites.
Tyr-FPA must be used for iodination since human FPA does not contain any
tyrosine residue. As the concentration of unlabelled FPA increases, less of the
tracer will be bound to the antibody. After the antibody antigen binding reaction
has been allowed to take place, free and antibody-bound FPA are separated
either by charcoal adsorption of the free FPA (recommended by Imco), or by
precipitating the antigen-antibody complexes with a second antibody (anti-
rabbit IgG). The precipitation of the antigen-antibody complexes is performed
by the addition of an immobilized second antibody (e.g. Sac-Cel, sold by IDS).
The radioactivity in the supernatant representing FPA-anti-FPA complexes
(charcoal system) or the radioactivity in the pellet representing FPA-anti-FPA-
anti-IgG complexes (second antibody system) is counted. The standard curve
is drawn by plotting the percentage of tracer bound in each standard versus
its concentration on a semi-log or on logit-Iog scale. The FPA concentration
in the unknown sample is determined by comparison of the percentage of
labelled FPA bound to the standard curve.
202
 FIBRINOPEPTIDE A

For laboratories familiar with RIA methods and equipped to perform a

iodination with 0.5-1.0 mCi (18.5-37 MBq), this RIA can easily be modified
and adapted to particular needs.
The following modifications, as compared to the assay protocol given by
Imco, have been introduced and successfully used in our laboratory for many
years:
1. Bentonite adsorption of fibrinogen. The bentonite suspension is prepared
exactly as described: 1 rnl of the bentonite suspension is added to 0.5 ml
plasma and incubated for 10 min on a rotating table at 4 0c. The mixture
is then centrifuged at 2000g and 4 °C for 20 min. The top 300 ,.d of the
supernatant are carefully aspirated and used for the RIA. The FPA values
obtained in the RIA must be corrected for the dilution caused by the bentonite
adsorption step. The dilution factor must be determined in each laboratory
because of differences depending on the quality of the bentonite.
2. The standards and samples are simultaneously mixed with the tracer and
the antibody and incubated at 4°C for 16-18 h instead of the recommenda-
tions by Imco, namely a sequential incubation of the standard or sample
with the antibody for 1 h, followed by the addition of the tracer with a
second incubation of 30 min.
3. The separation of the free and bound tracer is performed with an immobilized
second anti-rabbit IgG antibody (e.g. Sac-Cel), instead of the charcoal
separation. The amount of second antibody to be added must be tested
since it is dependent on the final dilution of the anti-FPA antiserum. The
incubation with the second antibody is carried out for 3 h at 4 °C on a
rotating or on a rocking table. The beads are then centrifuged at 2000g for
20 min at 4°C, the pellet is resuspended once with 1 ml of buffer I (Imco),
centrifuged again at 2000g for 10 min. The supernatant is discarded and the
pellet counted for radioactivity. The standard curve is drawn by plotting
percentage of tracer bound in each standard versus its concentration, either
on logit-Iog paper or by any computer curve fitting program (spline function).
The sensitivity of the assay is around 0.5-1 ng/ml. The intra- and inter-assay
variations are only 3-7%.
A more detailed description of the RIA method, iodination of Tyr-fibrinopeptide
A, all buffer solutions and a step-by-step description of the assay may be obtained
from the author.

PRINCIPLE AND ASSAY CHARACTERISTICS OF THE ELISA

ASSERACHROM FPA, STAGO

The FPA ELISA as supplied by Stago is a competitive binding assay. It is based

on a competition between a fixed amount of FPA immobilized in microtitre
plates and the FPA, standard or sample, mixed with the rabbit anti-FPA-
antibody solution and added to the microtitre plate.
For improvement of the sensitivity the standards or samples are first incubated
with a fixed amount of antibodies in a separate tube. During this incubation
step the FPA molecules bind to the free antibodies. An aliquot of this mixture
203
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
is then added to the FPA-coated plate and incubated. During this incubation
period the remaining free antibodies may bind to the immobilized FPA molecules
in the plate. Ample washing of the plate then removes the free FPA-anti-FPA-
antibody complexes and any free goat anti-FPA-antibody molecules. Finally,
the anti-FPA antibodies bound to the immobilized FPA molecules are revealed
by a goat anti-rabbit IgG coupled to peroxidase which is allowed to react with
its substrate a-phenylenediamine (OPD) in the presence of hydrogen peroxide.
The amount of colour produced is inversely proportional to the concentration
of FPA originally present in the standard or test sample. The standard curve
is drawn by plotting the optical density (492 nm) versus the concentrations of
the standards (nglml) on a semi-log scale. The FPA concentration in the
unknown sample is read off this standard curve.

REFERENCE VALUES AND QUALITY CONTROL

FPA concentrations may be expressed as nglrol or nmollL. To convert nglrol

to nmollL the value in nglrol is divided by the factor 1.S4 (e.g. 8 nglrol: 1.54 =
S.2nmoVL).
Normal FPA concentrations are always below 2 nglrol « 1.3 nmoVL) and
are independent of age and sex of the subject. Note: women on oral contraceptive
drug treatment have elevated FPA values 55,56.
The normal FPA values should be established in every laboratory performing
FPA determinations, because of the problems which may occur during blood
sampling or blood processing. The collection of blood from IS to 20 normal
healthy control subjects may help to assure the quality of the blood collection
and processing including the completeness of the elimination of fibrinogen by
bentonite.
Values higher than 2-2.5 nglrol found in healthy control persons, and values
found in patients higher than 60 nglrol, most probably point to difficulties in
blood sampling, blood processing or fibrinogen adsorption, provided the test
is adequately performed. The latter may easily be controlled by intra-assay
control samples.

References
I. Marder VJ, Francis cw, Doolittle RF. Fibrinogen, structure and physiology. In: Colman RW,
Hirsh J, Marder VJ, Salzman EW, eds. Haemostasis and thrombosis. Philadelphia: Lippincott,
1982; 145-63.
2. Nossel HL, Younger LR, Wilner GD, Procupez D, Canfield RE, Butler VP. Radioimmunoassay
of human fibrinopeptide A. Proc Natl Acad Sci USA 1971; 68: 2350-3.
3. Nossel HL, Yudelman I, Canfield RE, Butler VP, Spanondis K, Wilner GD, Qureshi GD.
Measurement of fibrinopeptide A in human blood. J Clin Invest 1974; 54: 43-53.
4. Soria J, Soria C, Ryckewaert JJ. A solid phase immuno enzymologica1 assay for the measurement
of human fibrinopeptide. A. Thromb Res 1980; 20: 425-35.
5. Amiral J, Walenga JM, Fareed 1. Development and performance characteristics of a competitive
enzyme immunoassay for fibrinopeptide A. Sem Thromb Haemostas 1984; 10; 228--42.
6. Bauer KA, Rosenberg RD. The pathophysiology of the prethrombotic state in humans:
Insights gained from studies using markers of hemostatic system activation. Blood 1987; 70:
343-50.

204
 FIBRINOPEPTIDE A
7. Hirsh 1. Blood tests for the diagnosis of venous and arterial thrombosis. Blood 1981; 57: 1-8.
8. Joist JH. Fibrinopeptide A in the diagnosis and treatment of deep venous thrombosis and
pulmonary embolism. Clin Lab Med 1984; 4: 363-80.
9. Douglas IT, Blamey SL, Lowe GDO, Carter DC, Forbes CD, Plasma beta-thromboglobu1in,
fibrinopeptide A and Bfl15-42 antigen in relation to postoperative DVT, malignancy and
stanozolol treatment. Thromb Haemostas 1985; 53: 235-8.
10. Leeksma OC, Meijer-Huizinga F, Stoepman-van-Dalen EA, van Ginkel CJW, van Aken WG,
van Mourik JA. Fibrinopeptide A and the phosphate content of fibrinogen in venous
thromboembolism and disseminated intravascular coagulation. Blood 1986; 67: 1460--7.
11. Nossel HL. Radioimmunoassay of fibrinopeptides in relation to intravascular coagulation
and thrombosis. N Engl J Med 1976; 295: 428-32.
12. Yudelman 1M, Nossel HL, Kaplan KL, Hirsh J. Plasma fibrinopeptide A levels in symptomatic
venous thromboembolism. Blood 1978; 51: 1189-95.
13. Sobel M, Sternbergh C, Marques D, Grimsdale AS. A comparative study of heparin responses
in arterial and venous thromboembolism using molecular markers for thrombosis. Circulation
1993; 88: 426--31.
14. Cronlund M, Hardin J, Burton J, Lee L, Haber E, Bloch KJ, Fibrinopeptide A in plasma of
normal subjects and patients with disseminated intravascular coagulation and systemic lupus
erythematosus. J Clin Invest 1976; 58: 142-51.
15. Mombelli G, Monotti R, Haeberli A, Straub pw. Relationship between fibrinopeptide A and
fibrinogen/fibrin fragment E in thromboembolism, DIC and various non-thromboembolic
diseases. Thromb Haemostas 1987; 58: 758-63.
16. Eisenberg PR, Lucore C, Kaufman L, Sobel BE, Jaffe AS, Rich S. Fibrinopeptide A levels
indicative of pulmonary vascular thrombosis in patients with primary pulmonary hypertension.
Circulation 1990; 82: 841-7.
17. Tulchinsky M, Zeller JA, Reba RC. Urinary fibrinopeptide A in evaluation of patients with
suspected acute pulmonary embolism. A prospective pilot study. Chest 1991; 100: 394-8.
18. Eisenberg PR, Sherman LA, Schectman K, Perez J, Sobel BE, Jaffe AS. Fibrinopeptide A: a
marker of acute coronary thrombosis. Circulation 1985; 71: 912-18.
19. Ga11ino A, Haeberli A, Baur HR, Straub pw. Fibrin formation and platelet aggregation in
patients with severe coronary artery disease: relationship with the degree of myocardial
ischemia. Circulation 1985; 72: 27-30.
20. Gallino A, Haeberli A, Hess T, Mombelli G, Straub pw. Fibrin formation and platelet
aggregation in patients with acute myocardial iufarction: effects of intravenous and subcutaneous
low-dose heparin. Am Heart J 1986; 112; 285-90.
21. Gensini GF, Rostagno C, Abbate R, Favilla S, Mannucci PM, Neri Semeri GG. Increased
protein C and fibrinopeptide A concentration in patients with angina. Thromb Res 1988; 50:
517-25.
22. Irie T, Imaaizumi T, Matuguchi T, Koyanagi S, Kanaide H, Takeshita A, Nakamura M.
Increased fibrinopeptide A during anginal attacks in patients with variant angina. J Am Coli
Cardio11989; 14: 589-94.
23. Mombelli G, 1m Hof V, Haeberli A, Straub pw. Effect of heparin on plasma fibrinopeptide
A (fpA) in acute myocardial iufarction. Circulation 1984; 69: 684--9.
24. Neri Semeri G, Gensini GF, Camovali M, Prisco D, Rogasi PG, Casolo GC, Fazi A, Abbate
R. Association between time of increased fibrinopeptide A levels in plasma and episodes of
spontaneous angina: a controlled prospective study. Am Heart J 1987; 113: 672-8.
25. Nichols AB, Owen J, Kaplan KL, Sciacca RR, Cannon PJ, Nossel HL. Fibrinopeptide A,
platelet factor 4, and fI-thromboglobulin levels in coronary heart disease. Blood 1982; 60:
650-4.
26. Theroux P, Latour JG, Leger-Gauthier C, De Lara J, Fibrinopeptide A and platelet factor
levels in unstable angina pectoris. Circulation 1987; 75: 156--62.
27. Van Hulsteijn H, Kolff J, Briet E, van de Laarse A, Bertina R. Fibrinopeptide A and
beta-thromboglobulin in patients with angina pectoris and acute myocardial infarction. Am
Heart J 1984; 107: 39-45.
28. Landi G, Barbarotto R, Morabito A, D'Angelo A, Mannucci PM. Prognostic significance of
fibrinopeptide A in survivors of cerebral iufarction. Stroke 1990; 21: 424--7.
29. Feinberg WM, Erickson LP, Bruck D, Kittelson 1. Hemostatic markers in acute ischemic
stroke: association with stroke type, severity, and outcome. Stroke 1996; 27: 1296--300.

205
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
30. Conway EM, Bauer KA, Barzegar S, Rosenberg RD. Suppression of hemostatic system
activation by oral anticoagulants in the blood of patients with thrombotic diseases. J Clin
Invest 1987; 80: 1535--44.
31. Yudelman I, Greenberg J. Factors affecting fibrinopeptide A levels in patients with venous
thromboembolism during anticoagulant therapy. Blood 1982; 59: 787-92.
32. Merlini PA, Ardissino D, Bauer KA et al. Persistent thrombin generation during heparin
therapy in patients with acute coronary syndromes. Arterioscler Thromb Vasc Bioi 1997; 17:
1325-30.
33. Melandri G, Semprini F, Cervi V et al. Benefit of adding low molecular weight heparin to the
conventional treatment of stable angina pectoris. A double-blind, randomized, placebo-
controlled trial. Circulation 1993; 88: 2517-23.
34. Rao AK, Sun L, Chesebro JH et al. Distinct effects of recombinant desulfatohirudin (Revasc)
and heparin on plasma levels of fibrinopeptide A and prothrombin fragment FI.2 in unstable
angina: a multicenter trial. Circulation 1996; 94: 2389-95.
35. Solymoss S, Bovill EG. Markers of in vivo activation of coagulation: interrelationships change
with intensity of oral anticoagulation. Am J Clin Patho11996; 105: 293-7.
36. Merlini PA, Ardissino D, Bauer KA et al. Activation of the hemostatic mechanism during
thrombolysis in patients with unstable angina pectoris. Blood 1995; 86: 3327-32.
37. Merlini PA, Bauer KA. Thrombin generation and activity during thrombolysis and concomitant
heparin therapy in patients with acute myocardial infarction. J Am CoIl Cardiol1995; 25:
203-9.
38. Rapold HJ, de Bono D, Arnold AER et al. Plasma fibrinopeptide A levels in patients with
acute myocardial infarction treated with alteplase. Correlation with concomitant heparin,
coronary artery patency, and recurrent ischemia. Circulation 1992; 85: 928-34.
39. Rapold HJ, Grimaudo V, Declerck PJ, Kruithof EKO, Bachmann F. Plasma levels of
plasminogen activator inhibitor type I, ~-thromboglobulin, and fibrinopeptide A before,
during, and after treatment of acute myocardial infarction with alteplase. Blood 1991; 78:
1490--5.
40. Scharfstein JS, Abendschein DR, Eisenberg PR et al. Usefulness of fibrinogenolytic and
procoagulant markers during thrombolytic therapy in predicting clinical outcomes in acute
myocardial infarction. Am J Cardio11996; 78: 503-10.
41. Auger MJ, Galloway MJ, Leinster SJ, McVerry BA, Mackie MJ. Elevated fibrinopeptide A
levels in patients with clinically localised breast carcinoma. Haemostasis 1987; 17: 336-9.
42. Edwards RL, Klaus M, Matthews E, McCullen C, Bona RD, Rickles FR. Heparin abolishes the
chemotherapy-induced increase in plasma fibrinopeptide A levels. Am J Med 1990; 89: 25-8.
43. Gugliotta L, Vigano S, D'Angelo A, Guarini A, Tura S, Mannucci PM. High fibrinopeptide
A (FPA) levels in acute non-lymphocytic leukemia are reduced by heparin administration.
Thromb Haemostas 1984; 52: 301-4.
44. Mombelli G, Roux A, Haeberli A, Straub pw. Comparison of 12sI-fibrinogen kinetics and
fibrinopeptide A in patients with disseminated neoplasias. Blood 1982; 60: 381-8.
45. Myers TJ, Rickels FR, Barb C, Cronlund M. Fibrinopeptide A in acute leukemia: relationship
of activation of blood coagulation to disease activity. Blood 1981; 57: 518-25.
46. Peuscher FW, Cleton FJ, Armstrong L, Stoepman-van-Dalen EA, van Mourik JA, van Aken
WG. Significance of plasma fibrinopeptide A (fpA) in patients with malignancy. J Lab Clin
Med 1980; 96: 5-14.
47. Douglas IT, Shah M, Lowe GOO, Belah JJF, Forbes CD, Prentice CRM. Plasma fibrinopeptide
A and ~-thromboglobulin in preeclampsia and pregnancy hypertension. Thromb Haemostas
1982; 47: 54-8.
48. Jones RL. Fibrinopeptide A in diabetes mellitus. Relation to levels of blood glucose, fibrinogen
disappearance, and hemodynamic changes. Diabetes 1985; 34: 836-43.
49. Rosove MH, Frank HJL, Harwig SSL. Plasma ~-thromboglobulin, platelet factor 4,
fibrinopeptide A, and other hemostatic functions during improved, short-term glycemic control
in diabetes mellitus. Diabetes Care 1984; 7: 174-9.
50. Ceriello A, Giacomello R, Stel G et al. Hyperglycemia-induced thrombin formation in diabetes:
the possible role of oxidative stress. Diabetes 1995; 44: 924-8.
51. Coccherri S, Mannucci PM, Palaretti G, Gervasoui W, Poggi M, Vigano S. Significance of
plasma fibrinopeptide A and high molecular weight fibrinogen in patients with liver cirrhosis.
Br J Haematol1982; 52: 503-9.

206
 FIBRINOPEPTIDE A
52. Mombelli G, Fiori G, Monotti R, Haeberli A, Straub PW: Fibrinopeptide A in liver cirrhosis.
Evidence against a major contribution of disseminated intravascular coagulation to
coagulopathy of chronic liver disease. J Lab Clin Med 1993; 121: 83-90.
53. Violi F, Ferro D, Basili S et al. Association between low-grade disseminated intravascular
coagulation and endotoxemia in patients with liver cirrhosis. Gastroenterology 1995; 109:
531-9.
54. Alkjaersig N, Fletcher AP, de Ziegler D, Steingold KA, Meldrum DR, Judd HL. Blood
coagulation in postmenopausal women given estrogen treatment: comparison of transdermal
and oral administration. J Lab C1in Med 1988; Ill: 224-8.
55. Inauen W, Baumgartner HR, Haeberli A, Straub PW: Excessive deposition of fibrin, platelets
and platelet thrombi on vascular subendothelium during contraceptive drug treatment. Thromb
Haemostas 1987; 57: 306-9.
56. Melis GB, Fruzzetti F, Paoletti AM, Carmassi F, Fioretti P. Fibrinopeptide A plasma levels
during low-estrogen oral contraceptive treatment. Contraception 1984; 30: 575-83.
57. Melis GB, Fruzzetti F, Ricci C, Carmassi F, Fioretti P. Oral contraceptives and venous
thromboembolic disease: the effect of the oestrogen dose. Maturitas, Suppl. 1988; 1: 131-9.
58. Herren T, Bartsch P, Haeberli A, Straub PW: Increased thrombin-antithrombin III complexes
after 1 h of physical exercise. J Appl Physiol1992; 73: 2499-504.
59. Blirtsch P, Welsch B, Albert M, Friedmann B, Levi M, Kruithof EKO. Balanced activation
of coagulation and fibrinolysis after a 2-h triathlon. Med Sci Sports Exerc 1995; 27: 1465-70.
60. Hoffman V, Straub PW: A radioimmunoassay technique for the rapid measurement of human
fibrinopeptide A. Thromb Res 1977; 11: 171-81.
61. Kockum C, Frebelius S. Rapid radioimmunoassay of human fibrinopeptide A - removal of
cross-reacting fibrinogen with bentonite. Thromb Res 1980; 19: 589-98.
62. Alkjaersig N, Fletcher AP. Catabolism and excretion of fibrinopeptide A. Blood 1982; 60:
148-56.
63. Gallino A, Haeberli A, Straub PW: Fibrinopeptide A excretion in urine in patients with
atherosclerotic artery disease. Thromb Res 1985; 38: 237--44.
64. Leeksma OC, Meijer-Huizinga F, Stoepman-van Dalen EA, van Aken WG, van Mourik JA.
Fibrinopeptide A in urine from patients with venous thromboembolism, disseminated
intravascular coagulation and rheumatoid arthritis - evidence for dephosphorylation and
carboxyterminal degradation of the peptide by the kidney. Thromb Haemostas 1985; 54:
792-8.
65. Wilensky RL, Zeller JA, Wish M, Tulchinsky M. Urinary fibrinopeptide A levels in ischemic
heart disease. J Am Coli Cardio11989; 14: 597-603.
66. Ardissino 0, Gamba MG, Mer1ini PA et al. Fibrinopeptide A excretion in urine: a marker of
the cumulative thrombin activity in stable versus unstable angina pectoris. Am J Cardio11991;
68: 58-63B.
67. Miller GJ, Bauer KA, Bazegar S et al. The effect of quality and timing of venepuncture on
markers of blood coagulation in healthy middle-aged men. Thromb Haemostas 1995; 73:
82-6.

207
22
Thrombin-antithrombin (TAT)
complexes
J.HARENBERG

INTRODUCTION

Blood coagulation terminates by the activation of prothrombin to thrombin

and subsequent conversion of fibrinogen to fibrin. During these processes
various activation peptides are released. Prothrombin fragments FI +2 are split
ofT from prothrombin by the prothrombokinase complex. Specific immunoassays
have been developed to determine the concentration of these peptides in plasma.
Antithrombin is an u2-globulin with a molecular weight of about 65 000
and is synthesized in the liver. Thrombin is released by activation of the
proenzyme prothrombin. Antithrombin inhibits thrombin by forming an inactive
protein/inhibitor complex 1• Tryptophan in the B-insertion loop of thrombin
is necessary for the interaction with antithrombin2.
Thrombin is inhibited by antithrombin, forming a thrombin-antithrombin
(TAT) complex. Therefore, the TAT complex also reflects the functional state
of the coagulation system. This complex is influenced to a smaller degree by
pre-analytical factors such as the technique of blood sampling. Therefore, TAT
has been adopted in a large set of clinical investigations, and represents a
diagnostic tool for the detection of hypercoagulability.

METHODS OF ASSAY
Principle and assay characteristics
The assay principle was described first by Neumann et al. 3 and Brower et al. 4
for the detection of elastase-ul-antiproteinase complexes, and is based on the
properties of the appropriate antibodies to bind selectively the corresponding
moieties of the complex. The assay for measuring the TAT complex consists
209
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
of antibodies directed against thrombin as well as antithrombin. Nanogram
quantities of TAT can be detected by these assays. Prothrombin, as well as
antithrombin, are not detected by the antibodies.
The TAT complex has a molecular we~t of 88,000 and elutes as a single
peak from a sodium dodecyl sulphate gel .

Materials
At present only one enzyme immunoassay test kit is commercially available.
This ELISA test system has been developed by Behringwerke (now Dade
Behring)6, and contains:
1. plastic tubes coated with antibodies to human thrombin, stability 4 weeks
at +4 DC;
2. antibodies to human antithrombin conjugated to horseradish peroxidase
(preservative phenol max I giL), stability 4 weeks at +4 DC or 3 months at
-20 DC;
3. buffered solution containing rabbit y-globulin and 5 mglml bovine serum
albumin (preservative phenol max 1 giL), stability 4 weeks at +4 DC or 3
months at -20 DC;
4. butTered solution for dilution of plasma samples 0.1 mol Tris-HCl pH 7.4
with 5 mglml bovine serum albumin (preservative sodium azide max I giL),
stability 4 weeks at +4 DC or 3 months at -20 DC;
5. PBS phosphate buffer saline solution containing 0.01 % Tween, stability as
given on the vial;
6. citrate buffer solution containing 0.1 moVL citrate and 0.2 moVL sodium
phosphate, pH 7.0 (preservative sodium-p-ethyl-mercury-mercapto-benzene
sulphonate max 0.1 giL);
7. 200 mglmll,2-phenylenediamine monohydrochloride, stability only in the
dark and 3 h at + 4 DC;
8. 0.5 N sulphuric acid, stability as given on the vial.
Plasma samples from human as well as various animal species (mouse, rat,
rabbit, guinea pig, dog, pig, sheep, baboon) can be analysed by this assay, but
notthe hen7 •

Procedure
Venous blood is withdrawn from individuals/animal species, in plastic syringes
or siliconized glass tubes containing 3.8% sodium citrate (9: 1 v/v). Samples are
centrifuged for 15 min at 2000g and at +4 DC; the obtained plasma is immediately
quick-frozen and stored at -30 DC.
The antibody specific to thrombin is bound to polystyrene plastic tubes.
Disposable tubes contain a volume of 2.0 ml or 0.35 ml on microtitre plates.
Plasma is incubated in the test tube. The thrombin of the TAT complex reacts
with the thrombin antibody on the surface of the plastic tube. The antithrombin
210
 THROMBIN-ANTITHROMBIN COMPLEXES

antibody is conjugated with peroxidase. This antibody reacts with the

antithrombin moiety of the TAT complex which was bound to the thrombin
antibody on the surface of the plastic tube. After washing the samples with
buffer o-phenylenediamine hydrochloride is added to the plasma sample and
thereafter hydrogen peroxide is added to the test tube. The reaction of the
release of o-phenylenediamine is stopped by adding H 2S04 and absorbance is
read at 492 nm.

EVALUATION OF RESULTS
Thrombosis and pulmonary embolism
Patients with deep vein thrombosis (DVT) and/or pulmonary embolism have
been studied to demonstrate the validity of the TAT assay. Elevated plasma
concentrations have been found in all patients with very recent clinical symptoms.
Normal TAT complexes in plasma exclude pulmonary embolism in patients
with clinically suspected pulmonary embolism8,9. In conclusion, many clinical
studies demonstrate the validity of the various activation variables of blood
coagulation. The sensitivity of all activation parameters ranges from 30% to
60% and the specificity from 55% to 85%. However, there is no conclusion, at
present, as to which activation variables possess the highest positive or negative
predictive value.

Postoperative care
TAT complexes have been measured in postoperative care medicine with the
aim of defining a predictive index for the development of DVT. TAT levels
were significantly higher in patients developing DVT when compared to patients
without DVT. Statistical analysis revealed no satisfactory discriminative power
for the diagnosis of developing DVT at any of the studied cut-off values for
TATIO. TAT levels were measured in plasma of patients who underwent total
knee replacement in order to determine the possible value of this assay in
screening for thromboembolic complications. No difference was observed in
patients who developed DVT and those who did notll.
Preoperative plasma levels of TAT complexes have been shown to correlate
with the development of venous thromboembolism after major hip or knee
surgery. However, even if much progress of the validity of TAT levels has been
made in perioperative medicine, further studies are needed concerning whether
rational recommendations about prophylaxis and screening for venous
thromboembolism can be made based on the results of preoperative TAT level.

Myocardial infarction and unstable angina

In patients with acute uncomplicated myocardial infarction and with unstable
angina pectoris elevated plasma levels of TAT were found 12 • Low-dose heparin
211
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
significantly reduces the rise in TAT complexes after myocardial infarction 13.
Elevated plasma levels in patients with coronary heart disease were also found by
others. These levels increased after a standardized exercise in these patients more
than in healthy controls, indicating that patients with coronary heart disease
possess a latent hypercoagulable state l4. However, no differences in TAT levels
were found in patients with unstable an~a who developed myocardial infarction
compared to a healthy control groupl . In patients with successful thrombolysis
of an occluded coronary vessel TAT levels decreased to the normal range, whereas
in patients with no successful thrombolysis TAT levels remained unchanged or
increased 16•

Disseminated intravascular coagulation and liver disease

Diseases with disseminated intravascular coagulation may also be related to
elevated levels of TAT. The sensitivity of the TAT levels is higher for detection
of disseminated intravascular coagulation than for myocardial infarction 17. In
patients with liver disease plasma levels of TAT were elevated, indicating an
activation of the coagulation system 18. The plasmin-a.rantiplasmin (PAP)
complexes (Chapter 29) are also increased in patients with liver disease. Therefore,
ratios were calculated to determine which effect of the two systems is more
important. Patients with sepsis and decompensated liver disease showed the
highest TATIPAP ratio l9 .

Malignancy, leukaemia and chemotherapy

TAT concentrations were investigated in patients with malignant diseases in
order to obtain information on activation of the coagulation system. Detailed
studies indicated that TAT levels are significantly higher in patients with
disseminated malignancy than in patients with limited disease. This indicates
a state of definite hypercoagulation in these diseases20. Activation of blood
coagulation has been described in patients with acute promyelocytic leukaemia
due to increased TAT levels21 . Treatment of patients with acute promyelocytic
leukaemia with L-asparaginase resulted in a further elevation of TAT complexes.
Patients who received antithrombin concentrate supplementation did not show
increases in TAT levels22 •

Diabetes mellitus
In patients with diabetes mellitus an activation of the coagulation system has
been described several times in the past, and has been attributed to the
morphological changes of the microvascular system. The levels of TAT were
significantly elevated in diabetes mellitus as compared to a healthy control
group23. However, as yet this has not been supported by others24 .
Thus, a variety of clinical conditions has been reported to be associated with
an increase in TAT concentration (Table 22.1).
212
 THROMBIN-ANTITHROMBIN COMPLEXES
Table 22.1 Elevated concentrations of TAT in plasma in relation to pathological conditions

Reference

Acquired diseases
acute deep vein thrombosis 9
acute pulmonary embolism 8
oral anticoagulation and heparinization 8-16
postoperative medicine 10,11
acute myocardial infarction and unstable angina 12-16
disseminated intravascular coagulation 17
liver disease 18,19
malignancy, leukaemia and chemotherapy 20,21
diabetes mellitus 23,24
Hereditary diseases 25,26

In summary, levels of TAT are elevated in acute stages of diseases, whereas

in subacute or chronic states of diseases contradictory results are reported.
Thus, TAT plasma levels are of clinical significance for screening hypercoagulation
in patients developing DVT or pulmonary embolism and to control the efficacy
of their anticoagulant regimen.

TAT levels in factor V Leiden mutation

Resistance to activated protein C due to factor V Leiden mutation has recently

been described in familial thrombophilia. Median TAT levels were above the
cut-off value in two-thirds of carriers compared to non-carriers. There were
no differences in median TAT levels among symptomatic and asymptomatic
factor V Leiden carriers25 • The frequency of elevated TAT levels is similar to
individuals with inherited deficiency of protein C and protein S26. However,
the data from the TAT levels and other markers of activated blood coagulation
still show large overlaps with those of normals and carriers of factor V Leiden
mutation or other inherited coagulation deficiencies.

DIFFICULTIES, SOURCE OF ERROR

The following points may influence the validity of the results: incorrect blood
sampling with coagulation of blood in vitro and incorrect storage of reagents.
Blood samples from patients with haemolysis, hypercholesterolaemia, hyper-
bilirubinaemia and high levels of rheumatic factors may influence the results.
The concentrations of TAT are influenced by technical problems during
withdrawal of blood. Blood sampling must be performed very carefully by a
clean venipuncture. The tourniquet of the forearm must be released after
puncture of a large vein of the forearm, and free-flowing blood must be taken
into a plastic syringe or into a siliconized glass tube containing 3.8% sodium
citrate. Despite careful blood sampling TAT levels in plasma have been described
to decrease within 3 months at -80 °C27 .
213
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Recently the determination of TAT levels has been extended to various
animal species, to new mechanical devices (artificial heart valves, coated
extracorporeal circuits), cell culture systems and to determine the efficacy of
new anticoagulants. In most investigations other variables for the measurement
of activation of blood coagulation are included. At present it remains open
which variable determines the activation of blood coagulation with the highest
sensitivity, specificity, negative or positive predictive value.

REFERENCE VALUE

The concentration of TAT in plasma samples of healthy subjects is not normallif'

distributed. Values range from 0.50 to 3.25 nglml with a peak at 1.25 nglml28,2 •

CONTROL AND CAUBRATION PROCEDURE

Standard curves are linear in a range 0.5--60 nglml. The recovery rate of purified
TAT added to normal plasma ranged from 90% to 110%. The intra-assay
coefficient of variation was 2.0--6.1 %. The inter-assay coefficient of variation
was between 2.4% and 4.7% when 2-60 nglml TAT were added to pooled
plasma. The linearity of plasma sample concentrations were determined after
various dilutions of the sample with TAT-poor pool plasma. The lower detection
limit of the assay is 0.5 nglml.

OTHER AVAILABLE METHODS OF ASSAY

The ELISA for the determination of TAT can be performed with larger
(ordinary test tubes, 5ml) and smaller volumes (microtitre plates) of the reagents.
The latter method has been developed in our laboratory and is now available
from Dade-Behring. The correlation between these two techniques is r 0.97. =
No other methods are available at the moment from other manufacturers.

References
I. Teitel 1M, Bauer KA, Lau HK, Rosenberg RD. Studies of the prothrombin activation pathway
utilizing radioimmunoassays for the F2FI +2 fragment and thrombin-antithrombin complex.
Blood 1982; 59: 1086-91.
2. Rezaie AR. Tryptophan 60-D in the B-insertion loop of thrombin modulates the thrombin-
antithrombin reaction. Biochemistry 1996; 35: 1918-24.
3. Neumann S, Gunzer G, Hennrich N, Lang H. PMN-elastase assay: enzyme immunoassay for
human polymorphonuclear elastase complexed with proteinase inhibitor. J Clin Chem Clin
Biochern 1984; 22: 693-7.
4. Brower MS, Harpel Oc. Alpha I-antitrypsin human leukocyte elastase complexes in blood:
quantification by an enzyme linked differential antibody immunosorbent assay and comparison
with alpha-2-plasmin inhibitor-plasmin complexes. Blood 1983; 61: 842-9.
5. Lau HK, Rosenberg RD. The isolation and characterization of a specific antibody population
directed against the thrombin-antithrombin complex. J Bioi Chern 1980; 255: 5885-93.

214
 THROMBIN-ANTITHROMBIN COMPLEXES

6. Enzygnost TAT ELISA test kit. Enzyme immunoassay for the determination of thrombin!
antithrombin III complex. Behringwerke AG, Marburg, Germany, 1997.
7. Ravanat C, Freund M, Dol F, Cadroy Y, Roussi J, Incardona F, Maffrand JP, Boneu B, Drouet
L, Legrand C. Cross-reactivity of human molecular markers for detection of prethrombotic
states in various animal species. Blood Coagul Fibrinol1995; 6: 446-55.
8. Carmassi F, Morale M, de Negri F, Puccetti R, Pistelli F, Mariani G, Pazzagli M, Palla A,
Giuntini e. Thrombin-antithrombin III complexes as an additional diagnostic aid in pulmonary
embolism. Haemostasis 1996; 26: 16-22.
9. Grosset AB, Spiro TE, Beynon J, Rodgers GM. Enoxaparin, a low-molecular-weight heparin
suppresses prothrombin activation more effectively than unfractionated heparin in patients
treated for venous thromboembolism. Thromb Res 1997; 86: 349-54.
10. Hoek JA, Nurmohamed MT, ten Cate Jw, Biiller HR, Knipscheer He. Thrombin-antithrombin
III complexes in the prediction of deep vein thrombosis following total hip replacement.
Thromb Haemostas 1989; 62: 1050-2.
11. Ginsberg JS, Brill-Eduards P, Panju A, Patel A, McGinnis J, Smith F, Dale I, Johnston M,
Ofosu F. Pre-operative plasma levels of thrombin-antithrombin III complexes correlate with
the development of venous thrombosis after major hip or knee surgery. Thromb Haemostas
1995; 74: 602-5.
12. Carvalho de Slus J, Azevedo J, Soria C, Barros F, Ribeiro C, Parreira F, Caen JP. Factor VII
hyperactivity in acute myocardial thrombosis - a relation to the coagulation activation. Thromb
Res 1988; 51: 165-73.
13. Pseja P, Lewandowski K, Turowiecka Z, Zozu1inska M, Tokarz A, Zawilska K. Fluctuation
of thrombin-antithrombin III complex in patients with acute myocardial infarction - influence
of low-dose heparin administration. Folia Haematol Leipz 1990; 117: 219-23.
14. van den Burg PJ, Hospers JE, van Vliet M, Mosterd WL, Bouma BN, Huisveld IA. Changes
in haemostatic factors and activation products after exercise in healthy subjects with different
ages. Thromb Haemostas 1995; 74: 1457-64.
15. Munkvad S, Gram J, Jespersen I A depression of active tissue plasminogen activator in
plasma characterizes patients with unstable angina pectoris who develop myocardial infarction.
Eur Heart J 1990; ll: 525-8.
16. Scharfstein JS, Abendschein DR, Eisenberg PR, George D, Cannon CP, Becker RC, Sobel
B, Cupples LA, Braunwald E, Loscalzo I Usefulness of fibrinogenolytic and procoagulant
markers during thrombolytic therapy in predicting clinical outcomes in acute myocardial
infarction. TIMI 5 Investigators. Thrombolysis in Myocardial Infarction. Am J Cardio11996;
78: 503-10.
17. Okajima K, Uchiba M, Murakami K, Okabe H, Takatsuki K. Determination of plasma
soluble fibrin using a new ELISA method in patients with disseminated intravascular
coagulation. Am J Hemato11996; 51: 186-91.
18. He S, Bremme K, Blombiick M. Acquired deficiency of antithrombin in association with a
hypercoagulable state and impaired function of liver and/or kidney in preeclampsia. Blood
Coagul Fibrinol1997; 8: 232-8.
19. Takahashi H, Tatewaki W, Wada K, Yoshikawa A, Shibata A. Thrombin and plasmin
generation in patients with liver disease. Am J Hemato11989; 32: 30-5.
20. Nakashima J, Tachibana M, Ueno M, Baba S, Tazaki H. Tumor necrosis and coagulopathy
in patients with prostate cancer. Cancer Res 1995; 55: 4881-5.
21. Avvisati G, ten Cate JW, Sturk A, Lamping R, Petti MG, Mandelle F. Acquired alpha-2-
antiplasmin deficiency in acute promyelocytic leukaemia. Br J Haematol1988; 70; 43-8.
22. Gugliotta L, D'Angelo A, Mattioli Belmonte M, Vigano-D' Angelo S, Colombo G, Datani
L, Gianni L, Lauria F, Tura S. Hypercoagulability during L-asparaginase treatment: the effect
of antithrombin III supplementation in vivo. Br J Haemato11990; 74: 465-70.
23. Takashahi H, Tsuda A, Tatewaki W, Wada K, Niwano H, Shibata A. Activation of blood
coagulation and fibrinolysis in diabetes mellitus. Evaluation by plasma levels of thrombin-
antithrombin III complex and plasmin-alpha-2-plasmin inhibitor complex. Thromb Res 1989;
55: 727-35.
24. van Wersch JWJ, Westerhuis LW, Venekamp WI Coagulation activation in diabetes mellitus.
Haemostasis 1990; 20: 263-9.

215
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
25. Simioni P, Scarano L, Gavasso S, Sardella C, Girolami B, Scudeller A, Girolami A. Prothrombin
fragment 1-2 and thrombin-antithrombin complex levels in patients with inherited APC
resistance due to factor V Leiden mutation. Br J Haemato11996; 92: 43~1.
26. Martinelli I, Bottasso B, Duca F, Faioni E, Mannucci PM. Heightened thrombin generation
in individuals with resistance to activated protein C. Thromb Haemostas 1996; 75: 703-5.
27. Iversen LH, Thorlacius-Ussing O. Short-time stability of markers of coagulation and fibrinolysis
in frozen plasma. Thromb Res 1996; 81: 253-{) I.
28. Pelzer H, Schwarz A, Heimburger N. Determination of human thrombin-antithrombin III
complex in plasma with an enzyme linked immunosorbent assay. Thromb Haemostas 1988;
59: 101-6.
29. Hoek JA, Sturk A, ten Cate JW, Lamping RJ, Berens F, Borm J1. Laboratory and clinical
evaluation of an assay of thrombin-antithrombin III complexes in plasma. Clin Chern 1988;
34: 2058-62.

216
23
Prothrombin fragment F1 +2
A.HAEBERLI

INTRODUCTION

The conversion of prothrombin to thrombin is a key event in the coagulation

of blood. Prothrombin fragment Fl +2 is an activation peptide released from
prothrombin during thrombin formation. Activation of prothrombin takes
place in the presence of factor Xa, factor Va, calcium ions and a phospholipid
suface (platelets). Prothrombin fragment FI +2 may be generated during
thrombin formation in two ways. Either prothrombin is cleaved to prethrombin
2 and prothrombin fragment FI +2. Subsequently prethrombin 2 is further
cleaved at one position to yield active thrombin. The alternative way of activation
is that prothrombin is cleaved to meizothrombin still containing the prothrombin
fragment Fl +2, and is then processed to thrombin, liberating the prothrombin
fragment FI + 2. The prothrombin fragment FI +2 may further be cleaved by
thrombin to fragments FI and F2.

PATHOPHYSIOLOGY

When the coagulation system is activated during pathological conditions only

a small amount of circulating prothrombin is activated to thrombin «1%).
Furthermore, the resulting enzyme is rapidly neutralized by antithrombin.
Therefore direct measurement of thrombin should be unable to detect a subtle
degree of thrombin generation. In contrast, quantitation of prothrombin
fragment FI +2 or thrombin-antithrombin (TAT) complexes (Chapter 22)
should allow monitoring of small degrees of the activation of coagulation with
the formation of thrombin. Part of the latter will then attack fibrinogen and
convert it to fibrin, with the liberation of fibrinopeptide A, another activation
marker of blood coagulation (see Chapter 21).
217
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Since the development of a radioimmunoassayl and later of two enzyme-
linked immunoassays2,3, this activation peptide has been tested in different
diseases known to exhibit activated blood coagulation. Since that time the
radioimmunoassay and enzyme-linked immunoassays have been widely used,
and they are considered by many authors to be important tests for the diagnosis
of prethrombotic and thrombotic states. The major diagnostic '¥'plications to
date have been in the detection of acute deep venous thrombosis4- ,disseminated
intravascular coagulation4, respiratory distress syndrome 8, angina9 ,lo and
myocardial infarction ll ,12, cerebral infarction 13 and for the monitoring of
anticoagulant therapyl4-2o. Some authors suggest that the measurement of
prothrombin fragment Fl +2 is an excellent method to monitor anticoagulation
with vitamin K-antagonists. It may be even better than the determination of
prothrombin time, but it is much more laborious and more expensive. Quite a
number of reports have become available on the activation or liberation of
thrombin during fibrinolytic therapy of acute myocardial infarction2l- 23 . The
determination of prothrombin fragment Fl +2 has also been used to follow
activation of the coagulation in factor V Leiden (APe resistancei4 and protein
S deficiencrs, in diabetes26 and during intake of oral contraceptives or oestrogen
therapy27. The activation of coagulation was also followed by measuring
different activation markers such as prothrombin fragment Fl +2 and others
during physical exercise28 .

METHODS OF ASSAY

The first test was developed in 1982 by the group of Teitel et a/. l and is a
radioimmunoassay using polyclonal antibodies against prothrombin fragment
Fl +2 and F2. This radioimmunoassay has since been used almost exclusively
by this group, and has not become available to a larger community. In 1991
Pelzer et al. presented an ELISA2which later became commercially available
(Enzygnost Fl +2, Dade-Behring). Since this test has been widely used in the
past 6 years it will be discussed in some detail below. Shortly afterwards two
other ELISA were developed for the measurement of prothrombin fragment
Fl +2. One was the Thrombonostika FI.2 (Organon-Teknika) using monoclonal
antibodies specific for prothrombin fragment FI +23. This test has been used
by several groups for the evaluation of activation of blood coagulation in
different diseases. This test is no longer commercially available. The last ELISA
kit was developed and sold by Baxter. It was in use for rather a short time and
is no longer available. Therefore, in the discussion of the method, only the test
still available, by Dade-Behring (Enzygnost FI +2) will be considered further.

BLOOD SAMPLING AND PROCESSING

As for other highly sensitive tests for low-grade activation of blood coaglllation,
e.g. fibrinopeptide A, blood sampling and processing is of importance l7, 29-34.
Although Dade-Behring recommends using citrate for blood collection, it is
advisable to add at least heparin to the anticoagulant solution for blood
218
 PROTHROMBIN FRAGMENT F1 +2

collection, since Tripodi et al. have shown a poor comparability between two
commercial prothrombin fragment Fl +2 ELISA methods35 . This was due at
least in part to the difference in the anticoagulants recommended by the
manufacturers, citrate versus heparin. Bauer and co-workers even recommend
using PPACK (phenyl-arginyl-prolyl-chloromethylketone), from Calbiochem,
a specific thrombin inhibitor, in addition to ensure an optimum quality of
blood samples29 • From our experience this recommendation is supported. In
our laboratory PPACK is used for all immunological determinations of
coagulation activation markers such as fibrinopeptide A (Chapter 21),
prothrombin fragment Fl +2, thrombin-antithrombin TAT complexes (Chapter
22) or plasmin-antiplasmin (PAP) complexes (Chapter 29).
In 1992 Soerensen and co-workers published an interesting paper on the
determination of prothrombin fragment Fl +2 in urine 36 • Unfortunately there
was no follow-up. Therefore it is not ensured that the measurement of the
fragment in urine is a worthwhile alternative.

PRINCIPLE AND ASSAY CHARACTERISTICS OF ELISA

ENZVGNOST F1 +2 DADE-BEHRING

The sandwich assay developed by Pelzer et al. 2 makes use of two polyclonal
antibodies against prothrombin and prothrombin fragment Fl + 2. The catching
antibodies, which are immobilized on micro titre plates, were raised in rabbits
against a 14 amino acid negatively charged synthetic peptide corresponding to
the C-terminal part of prothrombin fragment F1 +2. They do not cross-react
with intact prothrombin. The other antibody (polyclonal, rabbit) reacts with
prothrombin and prothrombin fragment Fl +2. It has been coupled to peroxidase.
The enzyme activity is determined with the substrate o-phenylenediamine (OPD)
in the presence of hydrogen peroxide. The test kit Enzygnost Fl +2 micro, supplied
by Dade-Behring, contains all the necessary reagents to perform the assay. One
minor disadvantage of the assay is that the incubation steps must be performed
at 37°C in a water bath or on a special heating plate. It is recommended to perform
the assay strictly according to the instructions given by the manufacturer. In
general the assay performs well with good reproducibility if no modifications are
introduced. The standard curve ranges from 0.04 to 10 nmollL. The optical density
read at 492 nm (490--500 nm) is plotted against the standard values on a double-
logarithmic graph. The limit of detection is close to 0.02 nmollL, the intra-assay
variations are 6.9-10.4% and the inter-assay variations are 6.7-11 %

REFERENCE VALUES AND QUALITY CONTROL

The normal values are variable to some extent depending on which assay was
used. For the Enzygnost F1 +2 Pelzer et al. 2 give a mean (± SD) of 0.67 ± 0.19
(range 0.32-1.2 nmoIlL). In our laboratory the normal values are 0.62 ± 0.21
nmollL (range 0.41--0.87 nmollL). Values higher than 1 nmollL mostly indicate
activation of blood coagulation. Values higher than 6 nmollL are rarely seen.
For other assays used formerly the following values were given: Thrombonostika
219
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

,
E
c
/
/
/

I
c
! V
/

0.1 V
v

0.1 10
nMorelL
Figure 23.1 Calibration curve, prothrombin fragment FI +2

F1.2 (Organon Teknika): 0.51 nmoVL (range 0.21-2.78 nmoVL)3 and radioim-
munoassay according to Teitel et al.: 1.97 ± 0.97 nmoVL 1 or 1.5 ± 0.7 nmoVL
(depending on another set of antibodies used)4. A minor increase with age
above 44 years, but no difference in sex, has been shown37 . 38.
The inter-assay variations should always be monitored, preferentially with
in-house control plasma. A reasonable control plasma is prepared by adding
to a pooled citrated plasma (10 donors or more) 100 Ilg PPACKlml, stored
frozen in aliquots until use.

References
I. Teitel lM, Bauer KA, Lau HI{, Rosenberg RD. Studies of tbe protbrombin activation patbway
utilizing radioimmunoassays for the F2IFI +2 fragment and tbrombin-antithrombin complex.
Blood 1982; 59: 1086-97.
2. Pelzer H, Schwarz A, Stiiber W Determination of human prothrombin activation fragment
1+2 in plasma witb an antibody against a synthetic peptide. Thromb Haemostas 1991; 65:
15J-9.
3. Hursting MI, Butman BT, Steiner JP et al. Monoclonal antibodies specific for prothrombin
fragment 1.2 and tbeir use in a quantitative enzyme-linked immunosorbent assay. Clin Chem
1993; 39:583-91.
4. Bauer KA, Rosenberg RD. The patbophysiology of tbe pretbrombotic state in humans: insights
gained from studies using markers of hemostatic system activation. Blood 1987; 70: 343-50.
5. Bouman CSC, Ypma ST, Sybesma IPHB. Comparison of the efficacy of D-dimer, fibrin
degradation products and prothrombin fragment I +2 in clinically suspected deep venous
thrombosis. Thromb Res 1995; 77: 225-34.
6. Estivals M, Pelzer H, Pichon SI, Boccalon H, Boneu B. Protbrombin fragment 1+2, tbrombin-
antithrombin III complexes and D-dimers in acute deep vein tbrombosis. Effects of heparin
treatment. Br I Haematol1991; 78: 421-4.
7. KyrJe PA, Eichinger S, Pabinger I et al. Prothrombin fragment Fl +2 is not predictive for
recurrent venous tbromboembolism. Thromb Haemostas 1997; 77: 829-33.
8. Yurdakok M, Yigit S, Gurakan B, Dundar S, Kirazli S. Thrombin-antithrombin III and
prothrombin fragment 1.2 levels in early respiratory distress syndrome. Am I Hemato11996;
51: 247-8.

220
 PROTHROMBIN FRAGMENT F1 +2
9. Kienast J, Thompson SG, Raskino C et aZ. Prothrombin activation fragment 1 + 2 and
thrombin-antithrombin III complexes in patients with angina pectoris: relation to the presence
and severity of coronary atherosclerosis. Thromb Haemostas 1993; 70: 550-3.
10. Rao AK, Sun L, Chesebro JH et aZ. Distinct effects of recombinant desulfatohirudin (Revasc)
and heparin on plasma levels of fibrinopeptide A and prothrombin fragment FI.2 in unstable
angina: a multicenter trial. Circulation 1996; 94: 2389-95.
II. Biasucci LM, Liuzzo G, Caligiuri G et aZ. Temporal relation between ischemic episodes and
activation of the coagulation system in unstable angina. Circulation 1996; 93: 2121-7.
12. Merlini PA, Bauer KA, Oltrona L et aZ. Persistent activation of coagulation mechanism in
unstable angina and myocardial infarction. Circulation 1994; 90: 61-8.
13. Catto A, Carter A, Ireland H et aZ. Factor V Leiden gene mutation and thrombin generation
in relation to the development of acute stroke. Arterioscler Thromb Vasc Bioi 1995; 15: 783-5.
14. Bruhn HD, Conard J, Mannucci M et al. Multicentric evaluation of a new assay for prothrombin
fragment FI +2 determination. Thromb Haemostas 1992: 68: 413-17.
15. Elias A, Bonfils S, Daoud-Elias M et af. Influence of long term oral anticoagulants upon
prothrombin fragment 1+2, thrombin-antithrombin III complex and D-dimer levels in patients
affected by proximal deep vein thrombosis. Thromb Haemostas 1993; 69: 302-5.
16. Granger CB, Miller JM, Bovill EG et aZ. Rebound increase in thrombin generation and activity
after cessation of intravenous heparin in patients with acute coronary syndromes. Circulation
1995;91: 1929-35.
17. Merlini PA, Ardissino D, Bauer KA et af. Persistent thrombin generation during heparin
therapy in patients with acute coronary syndromes. Arterioscler Thromb Vasc Bioi 1997; 17:
1325-30.
18. Millenson MM, Bauer KA, Kistler JP, Barzegar S, Tulin L, Rosenberg RD. Monitoring 'mini-
intensity' anticoagulation with warfarin: Comparison of the prothrombin time using a sensitive
thromboplastin with prothrombin fragment FI +2 levels. Blood 1992; 79: 2034-8.
19. Sobel M, Stembergh C, Marques D, Grimsdale AS. A comparative study of heparin responses
in arterial and venous thromboembolism using molecular markers for thrombosis. Circulation
1993; 88: 426--31.
20. Solymoss S, Bovill EG. Markers of in vivo activation of coagulation: Interrelationships change
with intensity of oral anticoagulation. Am J Clin Patho11996; 105: 293-7.
21. Baglin TP, Luddington R, Jennings I, Richards EM. Thrombin generation and myocardial
infarction during infusion of tissue-plasminogen activator. Lancet 1993; 341:504-5.
22. Merlini PA, Ardissino D, Bauer KA et af. Activation of the hemostatic mechanism during
thrombolysis in patients with unstable angina pectoris. Blood 1995; 86: 3327-32.
23. Scharfstein JS, Abendschein DR, Eisenberg PR et aZ. Usefulness of fibrinogenolytic and
procoagulant markers during thrombolytic therapy in predicting clinical outcomes in acute
myocardial infarction. Am J Cardiol1996; 78: 503-10.
24. Simioni P, Scarano L, Gavasso S et af. Prothrombin fragment 1+2 and thrombin-antithrombin
complex levels in patients with inherited APC resistance due to factor V Leiden mutation. Br
J Haematol1996; 92: 435-41.
25. Zoller B, Holm J, Svensson P, Dahlback B. Elevated levels of prothrombin activation fragment
I +2 in plasma from patients with heterozygous Arg (506) to Gin mutation in the factor V
gene (APC-resistance) and/or inherited protein S deficiency. Thromb Haemostas 1996; 75:
270-4.
26. Ceriellow A, Giacomello R, Stel G et al. Hyperglycemia-induced thrombin formation in
diabetes: the possible role of oxidative stress. Diabetes 1995; 44: 924-8.
27. Salah AA, Dorey LG, Dombrowski MP et aZ. Thrombosis and hormone replacement therapy
in postmenopausal women. Am J Obstet Gyneco11993; 169: 1554-7.
28. Bartsch P, Welsch B, Albert M, Friedmann B, Levi M, Kruithof EKO. Balanced activation
of coagulation and fibrinolysis after a 2-h triathlon. Med Sci Sports Exerc 1995; 27: 1465-70.
29. Bauer KA, Barzegar S, Rosenberg RD. Influence of anticoagulants used for blood collection
on plasma prothrombin fragment Fl +2 measurements. Thromb Res 1991; 63: 617-28.
30. Greenberg CS, Hursting MJ, Macik BG, Ortel TL, Kane WH, Moore BM. Evaluation of
preanalytical variables associated with measurement of prothrombin fragment 1.2. Clin Chern
1994; 40: 1962-9.
31. Iversen LH, Thorlacius-Ussing O. Short-time stability of markers of coagulation and fibrinolysis
in frozen plasma. Thromb Res 1996; 81: 253---{j I.

221
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

32. Leroy-Matheron C, Gouault-Heilmann M. Influence of conditions of blood sampling on

coagulation activation markers (prothrombin fragment 1+2, thrombin-antithrombin complexes
and D-dimers) measurements. Thromb Res 1994; 74: 399--407.
33. Miller GJ, Bauer KA, Barzegar S et al. The effect of quality and timing of venepuncture on
markers of blood coagulation in healthy middle-aged men. Thromb Haemostas 1995; 73:
82-6.
34. Rahr HB, Sorensen JV, Danielsen D. Markers of coagulation and fibrinolysis in blood drawn
into citrate with and without D-Phe-Pro-Arg-chloromethylketone (PPACK). Thromb Res
1994;73:279-84.
35. Tripodi A, Chantarangkul V, Bottasso B, Mannucci PM. Poor comparability of prothrombin
fragment I +2 values measured by two commercial ELISA methods: influence of different
anticoagulants and standards. Thromb Haemostas 1994; 71: 605-8.
36. Sorensen JV, Jensen HP, Rahr HB, Borris LC, Lassen MR, Knudsen F. FI +2 and FPA in
urine from patients with multiple trauma and healthy individuals. A pilot study. Thromb Res
1992;67:429-34.
37. Hursting MJ, Stead AG, Crout FV, Horvath BZ, Moore BM. Effects of age, race, sex and
smoking on prothrombin fragment 1.2 in a healthy population. Clin Chern 1993; 39: 683--6.
38. Lowe GDO, Rumley A, Woodward M et al. Epidemiology of coagulation factors, inhibitors
and activation markers: the Third Glasgow MONICA Survey 1. Illustrative reference ranges
by age, sex and hormone use. Br J Haematol1997; 97: 775-84.

222
24
Tissue-type plasminogen activator
(t-PA) activity
c. KLUFT, P. MEIJER, E. ERSDAL and S. ROSEN

INTRODUCTION

Tissue-type plasminogen activator (t-PA) is a major plasminogen activator of

the fibrinolytic system. t-PA is continuously released into the blood stream by
the endothelial lining, and also continuously and rapidly cleared by the liver.
The short half-life of around 5 min renders it a dynamic system that can be
influenced and changed in short time-frames.
t-PA is the only known haemostatic enzyme which is secreted in an active
form into circulating blood. However, activation of plasminogen is insignificant
in the absence of a fibrin surface which is necessary for exerting optimal activity.
However, t-PA in blood rapidly interacts with circulating inhibitors, the main
and most rapid inhibitor being plasminogen activator inhibitor-l (PAl-I). The
consequence of the continuous release and continuous inhibition and clearance
is a steady state in the circulation of active t-PA, in addition to the presence
of inactivated t-PA in complex with inhibitors l . The situation with regard to
PAI-l is illustrated in Figure 24.1, showing that the amount of active t-PA is
only a fraction of the total t-PA in circulation.
The interaction of t-PA with inhibitors in the blood also continues in vitro
when blood is collected in a routinely used anticoagulant, such as trisodium
citrate, and only after adequate measures to stop this interaction were introduced
did the assay of systemic t-PA activity become possible. It was discovered that
at a pH of 6 or below the rapid inhibition of t-PA by PAI-l was prevented2 ,3.
Subsequently, blood collection in anticoagulant was introduced that caused a
final pH of 6 or immediate acidification after normal collection3 • These develop-
ments date from around 1990. The t-PA activities documented and reported
before that time are only around 5-10% of the activities reported after proper
blood collection was introduced4 .
223
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

--~) t·PA antigen

+ t·PA activity

~ l·PA· PAI·l complex

----------------~) PAl· ' activity

Figure 24.1 Occurrence of t-PA and PAI-I in an average, healthy, young volunteer. The vertical
axis represents relative molar concentrations

In most t-PA activity assays the conversion of plasminogen to plasmin is

monitored. Therefore an additional, technical problem, besides the influence
of the plasminogen activator inhibitors during collection, was the interference
of PAI-l and plasmin inhibitors in the assay systems that usually required a
pH around 7-8. Initially, the irreversible denaturation of PAI-l and of plasmin
inhibitors with further acidification to around pH 4 was evaluated5•6, but later
a simplification was possible by isolating the t-PA from plasma at pH 6. This
isolation is achieved in a so-called biolo~cal immunoassay (BIA) in which t-PA
is captured by immobilized antibodies at pH 6, inhibitors are washed away,
and the assay at pH 7-8 follows 8•9 . In this chapter the principle and features
of its application are further described.

PRINCIPLE

The assay of t-PA captured by the antibody in the micro titre plate involves
two parallel reactions resulting from the simultaneous addition of plasminogen
and chromogenic plasmin substrate:
1. Activation of plasminogen in solution by active t-PA immobilized by
antibodies bound to the microtitre wall
Alr-t-PA + plasminogen -+ plasmin
(bound) (free)
2. Generated plasmin acts on the chromogenic substrate and the release of
p-nitroaniline (pNA) is detected at 405 nm
Plasmin + chromogenic substrate -+ pNA
The pNA generated after a fixed time interval is compared to corresponding
data of a calibration curve obtained by using different dilutions of a purified
t-PA standard. This t-PA standard should be calibrated against the WHO t-PA
standard, presently lot 86/670. The concentration ranges of t-PA that can be
evaluated are flexible and dependent upon the incubation times chosen for
224
 t-PAACTIVITY

binding t-PA and for plasmin generation. These times should in turn be selected
according to the petformance of the actual antibody used. With a high-petforming
antibody, 1 h-long incubation times are sufficient for the detection of as little as
0.02 IU/mL.

MATERIALS AND EQUIPMENT

1. Microtitre plate reader.
2. Microplate with wells coated with a monoclonal t-PA antibody which binds
t-PA without significantly impairing its activity (e.g. B-1O (lmco)).
3. Glu-plasminogen (quality equivalent to products from Chromogenix or
Biopool).
4. Chromogenicsubstrate(e.g.S-2403,pyroGlu-Phe-Lys-pNA. HCl(Chro-
mogenix).
5. t-PA standard calibrated against WHO t-PA standard, lot 86/670 (or a
reference plasma with pH 6.0).

SOLUTIONS
1. Distilled water.
2. 0.1 mollL Tris-acetate washing buffer, pH 8.4, 0.1% (v/v) Tween 80.
3. Assay reagents with 0.1 mg/ml Glu-plasminogen and 1 mmollL S-2403 in
washing buffer.
4. t-PA standard dilution range made in 0.1 mollL Tris-HCl pH 8.4, 0.15 mollL
NaCl, 5 mmollL disodium-EDTA, 0.1% (v/v) Tween 80.

ASSAY PROCEDURE
Blood collection 10
Select for venipuncture the sitting or lying position for the whole study and
allow for sufficient time to adjust to changes. Sitting or lying makes a difference
to haematocrit. The effects introduced by dilution with the usual liquid anti-
coagulants in blood tubes accentuate the difference between the two situa-
tions. Short-term effects of smoking, stress, exercise, alcohol, or beverages
containing caffeine on t-PA blood levels are known to exist; to minimize these,
subjects should refrain from smoking and the beverages mentioned for at least
1 h before venipuncture, and refrain for 18-24 h from alcohol. To minimize
the effects of stress, samples are taken from subjects who have been resting for
20 min. Fasting is preferred, but a light breakfast without fat and no tea or
coffee is permissible for morning samples. For sampling at other periods the
lunch and dinner times should be documented as well as the composition of
the meals. In view of the diurnal variations in t-PA, the time, or times, of blood
collection should be carefully chosen, standardized in studies, and be registered.
Subjects should have a normal or stable day/night rhythm in their lifestyle and
not be in a process of adaptation (unless this is the subject of the study).
225
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Blood should be collected in pre-cooled plastic or siliconized tubes containing
a specific anticoagulant. To prevent t-PA from being inactivated, blood should
reach pH 6 as soon as possible, which can be achieved with special anticoagulants
such as the commercially available Stabilyte®(Biopool) tubes 3 or rapid acidifi-
cation.
The acidified blood can be kept at room temperature, but preferably at 4°C,
and is centrifuged within 2-4 h after collection. The minimum condition is
centrifugation for 20 min at 1000g. The plasma collected can be stored frozen,
but the pH 6 condition requires freezing by a thorough and rapid method
(snap-freezing). For all storage the temperature should be constantly below
-30°C; storage at -70 °C or lower is strongly recommended for this type of
plasma.

Laboratory assay
1. 50,.u of undiluted plasma at pH - 6, and 100 ,.u distilled water are added
to a microtitre plate well coated with a monoclonal t-PA antibody (thus
maintaining a pH of 6 during the binding step).
2. Incubation for 1 h at room temperature.
3. Empty the well and wash three times with 300 III 0.1 mol/L Tris-acetate
washing buffer, pH 8.4, 0.1% (v/v) Tween 80.
4. Add 150 ,.u assay reagent containing a mixture of 0.08 mglml Glu-plasminogen
and 1 mmollL chromogenic substrate S-2403.
5. Read 00 at 405 nm directly after the addition in 4 and after 15, 30 or
60 min, depending on the expected range of activities (e.g. pre- or post-
occlusion; cf. Chapter 31).
6. Calculation of results: the t-PA activity of the test sample can be expressed
in IVlml from evaluation using a standard curve obtained from parallel runs
of t-PA standard dilutions.

(PATHO)PHYSIOLOGY

For most data reported before 1990, when adequate blood sampling for t-PA
activity was discovered, the variability in t-PA activity4 requires re-analysis
whether or not it concerned a true variation in circulating t-PA activity and
was not a consequence of variability of PAI-l expressed via effects in vitro.
Data on circulating t-PA activity are limited so far.
With the method described, analysis of Stabilyte plasma from 67 healthy
volunteers (aged 20-65 years; 22 males, 45 females) sampled at rest in the
morning showed a median value of 0.46 IVlml and a wide range from 0.07 to
1.3 IU/ml9 • For 40 of these individuals blood was also collected in parallel in
normal citrate showing a median value of 0.03 IU/ml (range 0-0.5), illustrating
the importance of the proper collection of the blood9 • Comparable values were
observed in other studies on 94 young women not using oral contraceptives
(unpublished), in healthy males (mean age 43)11; and in two groups of 40
healthy males and females (age 20-55 years)12. The mean values reported in
the studies mentioned were 0.385 (n =94); 0.29 (n =8)11; 0.34 (n =24)11; 0.29
(n =40)12 and 0.27 IUlml (n =40)12.
226
 t-PA ACTIVITY

For a very similar method (Chromolize™) systematically higher values in

control populations have been reported (see Pitfalls, 3), such as 0.698 (n =150)
(unpublished); 0.56 (n =23)13; 0.53 (n =24)13; 0.55 (n =14)14,15; 0.64 (n =124)
IUlml I6 .
The analysis of t-PA activity in acidified blood with the above or closely
related methods shows a number of variations that are briefly summarized.
The well-known diurnal variation in blood fibrinolytic activity is due to changes
in both t-PA and PAl-I. This variation is less prominent but still exists in the
t-PA activity when acid blood collection is used 13,17-19. Baseline t-PA activity
is not different in elderly individuals with different genotypes of the lID
polymorphism of the t-PA gene20; however, local and stimulated release appears
to be strongly associated with this polymorphism21 . Spontaneous longitudinal
fluctuations in t-PA activity are of similar magnitude to those in the determining
variables t-PA antigen and PAI-l antigen, and for 2-3 weeks show a correlation
coefficient of 0.58 22 . A small difference between males and females might
exist 18 ,19, but age does not seem to be a significant determinant 16,18.
t-PA activity was found to be negatively associated with body mass index
(BMI)12 in healthy middle-aged individuals and healthy elderly individuals 16.
In the latter group association is also significant with diastolic blood pressure
and high-density lipoprotein (HDL)16. Further negative associations with an
index of insulin resistance and triglycerides have been observed23 . Accord-
ingly, t-PA activity is significantly correlated with specific variables of insulin
resistance as determined by using a euglycaemic clamp24. Strenuous exercise
in healthy elderly men and women increases t-PA activity16.
The relation of t-PA activity to body fat was not significant in middle-aged
men, but rather cardiovascular fitness (Vo 2 max) was found to be indepen-
dently related in multivariate analysis25 , in accordance with other observations
of a relationship with physical activity16,26.
Job demand is negatively associated with t-PA activitY 7. Diets enriched in
fish oil 13 and omega-3 fatty acids28 do not influence t-PA activity; however, a
relation exists to ~-carotene (positive) and retinol (negative correlation)29.
For a recent review on t-PA in relation to angina pectoris patients, see ref
30. Both reduced t-PA activity and slightly, not significantly, higher levels in
survivors of myocardial infarction compared with control subjects have been
reported20,31. In insulin-dependent diabetics t-PA activity is found to be elevated
in normoalbuminuria while patients with clinical nephropathy have significantly
lower t-PA activityI4,15. t-PA activity is reduced in postmenopausal non-insulin-
dependent diabetes mellitus (NIDDM) patients with high BMI (mean 30.8)24,
and short-term oestrogen treatment in postmenopausal NIDDM did not affect
t-PA activity 32. t-PA activity is reduced during pregnancy33, but increased
shortly after delivery34. Androgenic-anabolic steroids enhance t-PA activity35.
t-PA activity is elevated in severe liver cirrhosis 36; and t-PA activity fluctuates
in the postoperative period37.

PITFALLS
1. For each variant of the method a check should be included for a number
of potential interfering and essential factors. These include:
227
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

(a) Efficacy of t-PA binding to the immobilized t-PA antibody at pH 6

versus pH 7-8. A comparison of incubation at pH 6 with one at pH
7-8, including in both cases antibodies to neutralize PAl-lor in PAI-l-
deficient or -depleted plasma, can provide the required information.
The interference of plasma should be evaluated from recovery studies
of purified t-PA in buffer or added to t-PAlPAI-free plasma.
(b) Interference by PAl-1: addition of PAl-lor PAl-I-neutralizing antibodies
to the sample should not change the result.
(c) Interference by u-PA: addition of u-PA and/or scu-PA to the sample
should not change the results.
(d) Binding of plasminogen to the coated plate: addition of exogenous
plasminogen to a sample with normal and low plasminogen should not
affect the results.
2. t-PA is secreted in a one-chain form with two different glycosylation patterns
and may be converted to a two-chain form in certain situations:
(a) Checks should be made on whether or not t-PA molecules with differ-
ences in glycosylation produce different activities when added in equal
molar amounts 38 •
(a) Checks should be made on whether or not conversion from one- to
two-chain t-PA results in difference in activity.
This relates also to the t-PA standard preparation used.
3. Normal values reported for different described methods are different. In a
direct comparison of the described method with Chromolize™ we found
=
a systematic difference of a factor 2.8 (correlation: y 2.8x - 0.09; n 42; =
=
r 0.942) (unpublished). This difference has not yet been explained.

COMMERCIAL AND OTHER METHODS

Biological immunoassay (Chromolize™ tPA, Biopool).

ELISA method (Immuno Free t-PA, Corvas Pharmaceuticals).

References
1. KIuft C. Constitutive synthesis of tissue-type plasminogen activator (t-PA) and plasminogen
activator inhibitor type I (PAl-I): conditions and therapeutic targets. Fibrinolysis 1994; 8
(Suppl. 2): 1-7.
2. Chmielewska J, Ranby M, Wiman B. Evidence for a rapid inhibitor to tissue plasminogen
activator in plasma. Thromb Res 1983; 31: 427-36.
3. Ranby M, Sundell IB, Nilsson TK. Blood collection in strong acidic citrate anticoagulant
used in a study of dietary influence on basal tPA activity. Thromb Haernostas 1989; 62: 917-22.
4. Verheijen JH. Tissue-type plasminogen activator activity assay. In: Jespersen J, Bertina RM,
Haverkate F, eds. ECAT assay procedures. A manual of laboratory techniques. Dordrecht:
KIuwer, 1992; 139--46.
5. Maat de MPM, KIuft C, Boer de K, Knot EAR, Jie AFH. Acid treatment of plasma for the
inactivation of plasminogen activator inhibitor-l (PAl-I). Thromb Res 1988; 52: 425-30.
6. Wejkum L, Rosen S, S0rskog L, Brandt B, Chmielewska I Blood sampling and determination
of tissue plasminogen activator activity with COA-SE~ t-PA. Fibrinolysis 1990; 4 (Suppl.
2): 152-4.
7. Mahmoud M, Gaffney PI Bioirnmunoassay (BIA) of tissue plasminogen activator (t-PA)
and its specific inhibitor (t-PAJINH). Thromb Haernostas 1985; 53: 356-9.

228
 t-PA ACTIVITY
8. Meijer P, Boon R, Jie AFH, Rosen S, Kluft C. Bioimmunoassay for tissue-type plasminogen
activator (t-PA) in human plasma: Evaluation of blood sampling and handling procedures
and comparison with other t-PA activity methods. Fibrinolysis 1992; 6 (Suppl. 3): 97-9.
9. Rosen S, Wejkum L, Billing-Claeson S, Ghosh R, Grdic K, Chmielewska J, Meijer P, Kluft
C, Tengborn L, Larsson G, Conkie J, Walker 1. Evaluation of a bioimmunoassay for t-PA
activity and its relation to PAl-I activity and antigen levels. 1998 (Submitted).
10. Kluft C, Meijer P. Update 1996: Blood collection and handling procedures for assessment of
plasminogen activators and inhibitors (Leiden Fibrinolysis Workshop). Fibrinolysis 1996; 10
(Suppl. 2): 171-9.
11. Velthuis-te Wierik EJM, Meijer P, Kluft C, Berg YD. Beneficial effect of a moderately energy-
restricted diet on fibrinolytic factors in non-obese men. Metabolism 1995; 44: 1548-52.
12. Velthuis-te Wierik EJM, Kluft C, Berg VD, Weststrate JA. Consumption of reduced-fat
products, haemostatic parameters and oral glucose tolerance test. Fibrinolysis 1996; 10: 159--66.
13. Marckmann P, Bladbjerg E, Jespersen J. Dietary fish oil (4 g daily) and cardiovascular risk
markers in healthy men. Arterioseler Thromb Vase Bioi 1997; 17: 3384-91.
14. Myrup B, Rossing P, Jensen T, Gram J, Kluft C, Jespersen J. The effect of the relationship
between tissue-type plasminogen activator and plasminogen activator inhibitor type I on
tissue-type plasminogen activator activity in insulin-dependent diabetes mellitus. Fibrinolysis
1994; 8 (Suppl. 2): 22-4.
15. Myrup B, Rossing P, Jensen T, Gram J, Kluft C, Jespersen J. Fibrinolysis in insulin-dependent
diabetic patients with and without nephropathy. Fibrinolysis 1996; 10: 331-5.
16. Schuit AJ, Schouten EG, Kluft C, de Maat M, Menheere PP, Kok FJ. Effect of strenuous
exercise on fibrinogen and fibrinolysis in healthy elderly men and women. Thromb Haemostas
1997; 78: 845-51.
17. Wa11nofer AE, van Griensven JM, Schoemaker HC, Cohen AF, Lambert W, Kluft C, Meijer
P, Kooistra T. Effect of isotretinoin on endogenous tissue-type plasminogen activator (t-PA)
and plasminogen activator inhibitor I (PAl-I) in humans. Thromb Haemostas 1993; 70:
1005-8.
18. Eliasson M, Evrin P-E, Lundblad D, Asplund K, Ranby M. Influence of gender, age and
sampling time on plasma fibrinolytic variables and fibrinogen. Fibrinolysis 1993; 7: 316-23.
19. Chandler WL, Schwartz RS, Stratton JR, Vitiello MY. Effects of endurance training on the
circadian rhythm of fibrinolysis in men and women. Med Sci Sports Exerc 1996; 28: 647-55.
20. van der Born JG, de KnijffP, Haverkate F, Bots ML, Meijer P, de Jong PT, Hofman A, Kluft
C, Grobbee DE. Tissue plasminogen activator and risk of myocardial infarction. The Rotterdam
Study. Circulation 1997; 95: 2623-7.
21. Jern C, Ladenva1l P, Wall U, Jern S. Association between a tissue-type plasminogen activator
(t-PA) gene polymorphism and forearm vascular release rate of t-PA. In: In vivo studies on
local release of tissue-type plasminogen activator from vascular endothelium. Wall University
thesis, Gotenborg University, Sweden, 1997.
22. Stegnar M, Mavri A. Reproducibility of fibrinolytic response to venous occlusion in healthy
subjects. Thromb Haemostas 1995; 73: 453-7.
23. Lindahl B, Asplund K, Eliasson M, Evrin P-E. Insulin resistance syndrome and fibrinolytic
activity: the Northern Sweden Monica study. Int J Epidemiol1996; 25: 291-9.
24. Brussaard HE, Gevers Leuven JA, Krans HMJ, Meijer P, Buytenhek R, Kluft C. Non-esterified
fatty acids are related with hypofibrinolysis in type 2 diabetes mellitus. Fibrinolysis 1994; 8
(Supp!. 2): 25-7.
25. Szymanski LM, Durstine JL, Davis PG, Dowda M, Pate RR. Factors affecting fibrinolytic
potential: cardiovascular fitness, body composition, and lipoprotein(a). Metabolism 1996; 45:
1427-33.
26. Shaukat N, Douglas IT, Bennett JL, Bono de DP. Can physical activity explain the differ-
ences in insulin levels and fibrinolytic activity between young Indo-origin and European
relatives of patients with coronary artery disease? Fibrinolysis 1995; 9: 55-63.
27. Ishizaki M, Tsuritani I, Noborisaka Y, Yamada Y, Tabata M, Nakagawa H. Relationship
between job stress and plasma fibrinolytic activity in male Japanese workers. Int Arch Occup
Environ Health 1996; 68: 315-20.
28. Eritsland J, Seljeflot I, Abdelnoor M, Arnesen H. Long-term influence of omega-3 fatty acids
on fibrinolysis, fibrinogen, and serum lipids. Fibrinolysis 1994; 8: 120-5.

229
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
29. Eliasson M, Asplund K, Evrin P-E, Huhtasaari F, Johansson I. Plasma fibrinogen, fibrinolysis
and (pro)vitamins; is there a connection? Fibrinolysis 1995; 9: 87-92.
30. Juhan-Vague I, Pyke SDM, Alessi MC, Jespersen J, Haverkate F, Thompson MA on behalf
of the ECAT Study Group. Fibrinolytic factors and the risk of myocardial infarction or
sudden death in patients with angina pectoris. Circulation 1996; 94: 2057-63.
31. Chandler WL, Stratton JR. Laboratory evaluation of fibrinolysis in patients with a history
of myocardial infarction. Am J Clin Patho11994; 102: 248-52.
32. Brussaard HE, Gevers Leuven JA, Frolich M, KIuft C, Krans HM. Short-term oestrogen
replacement therapy improves insulin resistance, lipids and fibrinolysis in postmenopausal
women with NIDDM. Diabetologia 1997; 40: 843-9.
33. Ishii A, Yamada S, Yamada R, Hamada H. T-PA activity in peripheral blood obtained from
pregnant women. J Perinat Med 1994; 22: 113-17.
34. Stegnar M, Meglic A, Meglic L, Novak-Antolic Z. Fibrinolysis after delivery: Caesarean
section versus vaginal delivery. Fibrinolysis 1994; 8: 270--5.
35. Pernot C, Hartgens F, Keizer HA, Kuipers H, Hamulyak K. Effects of self-administration
of high dose androgenic-anabolic steroids on fibrinolytic activity in non-elite bodybuilders.
Fibrinol Proteo11996; 10 (Suppl. 2): 53-4.
36. Ferro D, Quintarelli C, Saliola M, Allessandri C, Basili S, Bonavita MS, Violi F. Prevalence
of hyperfibrinolysis in patients with liver cirrhosis. Fibrinolysis 1993; 7: 59-62.
37. Nielsen JD, Gram J, Fabrin K, Holm-Nielsen A, Jespersen J. Lack of correlation between
blood fibrinolysis and the immediate or post-operative blood loss in transurethral resection
of the prostate. Br J Haematol1997; 80: 105-10.
38. Huisman LGM, Griensven van JMT, KIuft C. Usefulness and limitations of Actilyse for
standardisation of tissue-type plasminogen activator (t-PA) assays in blood. Fibrinolysis 1996;
10 (Suppl. 2): 125-7.

230
25
Tissue-type plasminogen activator
antigen (t-PA Ag)
M. C. ALESSI and I. JUHAN-VAGUE

INTRODUCTION
Physiological role
Tissue-type plasminogen activator (t-PA) constitutes an important agent in the
fibrinolytic pathway. It is synthesized and released continuously into the blood
by endothelial cells and then rapidly cleared by the liver. The secreted form, a
single-chain glycoprotein composed of 530 acids (70 kD), can be converted by
plasmin or kallikrein into a double-chain form with one disulphide bond.
Several domains have been distinguished on the molecule: a finger domain
homologous to the type I fingers of fibronectin, an epidermal-growth-factor-like
domain, two 'kringle' domains similar to the triple disulphide bound structures
found in many plasma proteins, a connecting peptide between the second
kringle and the plasmin cleavage site and the carboxyl terminal domain
constituting the chain bearing the active site.
The physiological role of t-PA is to activate plasminogen to plasmin which
degrades fibrin to soluble degradation products. Fibrinolysis appears to be
regulated by specific molecular interactions between t-PA and fibrin, as well
as between plasmin and plasmin inhibitor. Because fibrin is required as a
cofactor, the activation of plasminogen can occur only on the fibrin surface
(for review see ref. 1).

Pathophysiological role
Circulating t-PA levels are thought to have a major effect on fibrinolytic capacity,
and proper regulation of the t-PA is essential. Increased plasma levels of t-PA
antigen have been found to be predictive of arterial ischaemic events 2 • It is
231
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
noteworthy that this increase in circulating t-PA antigen does not reflect an
increase in fibrinolytic capacity but rather the opposite. This has been attributed
to a parallel evolution of PAI-I which results in t-PAlPAI-1 complexes rather
than a rise in active t-PA.
Decreased release of t-PA has been associated with thrombosis. Various tests
have been designed to evaluate t-PA release. Different aspects of the response
are probably measured depending on the stimulus used, e.g. venostasis (Chap-
ter 31)3, desamino-8-o-arginine vasopressin (DDAVP) injection4 or physical
exerciseS. The capacity of the endothelium to release t-PA can be assessed by
comparing either t-PA activity or antigen level before and after stimulation.
Circadian variations in t-PA antigen have been described in relation with
variations in plasminogen activator inhibitor 1 (PAl-I). Increase in t-PA antigen
concentration has been reported in the elderly and in inflammatory syndrome,
in association with several clinical conditions involving increased PAI-I
concentration. The association of hyperfibrinolysis with haemorrhage has been
attributed to a congenital lack of inhibitor, but a tendency to bleeding has also
been observed in patients displaying high t-PA release. Venostasis revealed
defects in t-PA release in patients with severe von Willebrand disease and
abnormally low release of t-PA has been reported in 10-20% of patients with
recurrent deep vein thrombosis (for review see ref. 6, and see Chapter 31).

METHODOLOGY
t-PA activity in whole blood or plasma varies due to reactions between t-PA
and inhibitors resulting in formation of t-PA inhibitor complexes. Immunological
determination of t-PA allows a total quantification of circulating t-PA.
Efforts to develop assay procedures for t-PA antigen in plasma have been
under way for many years and several techniques have been described7- 17• The
earliest methods were radioimmunoassays but they were soon replaced by
enzyme-linked immunosorbent assay (ELISA) which have the advantage of
using a stable enzyme label. The switch to ELISA was concomitant with
improvements in calibrator preparation, in antibody sources and in the buffer
as well as the tracer. Current commercially available kits contain stable and
reproducible reagents (see Table 25.1).

Techniques
Radioimmunoassay
Radioimmunoassay based on 12sI-labelled antibodies against t-PA ('2sI_Ab)
was the first method to be used. Briefly samples were incubated in polystyrene
tubes of microtitre plates coated with IgG from goat or rabbit antiserum. After
washing, 125I_Ab was added. Finally, after another wash, radioactivity on the
walls of the recipient was counted in a y-scintillator. Results were evaluated by
comparison with a standard curve constructed with pure t-PA.
In one variant of this technique, the sample, antiserum and l2SI-t-PA were
232
 t-PA ANTIGEN

first incubated. Next anti-IgG immobilized covalently to Sepharose was added,

and the bound and free fractions were separated by sedimentation through
10% sucrose or centrifugation. Finally radioactivity in the solid phase was
counted.

ELISA
ELISA techniques rapidly superseded radioimmunoassay because they are so
much easier to perform. Specific IgG or the F(ab')2 fragments were coupled
on polystyrene wells in which samples were incubated for a few hours. After
washing, peroxidase or [3-D-galactosidase-labelled antibodies were added.
Specific substrate (orthophenylenediamine methylumbelliferyl [3-D galactoside)
was then added for various periods of time depending on the assay. The
relationship between t-PA and fluorescence intensity or coloration was measured
using a spectrofluorometer (ex 360 nm, em 450 nm) or a spectrophotometer
(492 nm) respectively.
Rijken et al. 9 described a three-step ELISA procedure with the last step
consisting of the use of a rabbit anti-(goat) IgG labelled with alkaline phos-
phatase. Holvoet et al. 14 proposed a two-site ELISA using a coating phase
comprising the IgG fractions of two monoclonal antibodies directed against
different exposed epitopes on the t-PA molecule. They also proposed an ELISA
for free t-PA using a murine monoclonal antibody directed against the active
site of t_PA I8 • Several groups have also developed ELISA techniques to quantify
t-PA complexed with PAI-I 19,20.

Materials
Calibrator t-PA
t-PA is synthesized and secreted at a much higher level by numerous cultured
human cell lines than by endothelial cells. t-PA purified from the culture medium
of Bowes melanoma cell line has served as a rich source of calibrator t-PA.
Several purification processes have been described. t-PA was first purified using
immunoaffmity chromatography with antibodies directed against porcine t-PA,
chromatography on arginine Sepharose and gel filtration. Then extraction of
t-PA was achieved by three chromatographies using zinc chelate Sepharose and
concanavalin A Sepharose, followed by filtration on Sephadex G50. More
recently t-PA has been purified by immunoaffinity chromatography using
monoclonal antibodies directed against human t-PA.
Wun and Capuano 17 obtained calibrator t-PA from Hela cells stimulated
with 12-0-tetradecanoyl phorbol13 acetate by chromatography on Erythrina
latissima inhibitor Sepharose 4B. The resulting product was a mixture of the
single- and double-chain forms. Purification in the presence of aprotinine
allowed production of a 90% pure single-chain form (sct-PA). t-PA has been
produced by DNA recombinant technology. Concentrations in the purified
preparation were determined by protein quantification and amino acid analysis.
233
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
International standard NIBSC for t-PA was first prepared in 1983 from
melanoma cells (coded 83/517) and then in 1986 (coded 86/670). This latter
was prepared by high-performance liquid chromatography on a gel exclusion
column. Preparation contained 98% sct-PA2 !.

Antibodies
All the assays described up to now have made use of specific antibodies raised
in goats or rabbits. IgG were purified by ammonium sulphate precipitation or
from protein A Sepharose. Monoclonal antibodies have also been produced in
order to obtain stable and reproducible antibodies.

Quality control and reference values

Gaffney and Curtis22 analysed data of ELISA for t-PA performed in 13 different
laboratories and observed a high degree of variability in results. Discrepancies
were greatest when normal plasma was assayed using the intemationallaboratory
standard established in 1983. The International Committee for Thrombosis
and Hemostasis (Fibrinolysis Subcommittee) has recommended that plasma
with a known level of t-PA be made available to supplement the current t-PA
international standard for the measurement of t-PA in plasma. The mean of
reference values is around 6-7 ng/ml

Sample collection
Blood samples should be taken after a 20-min rest period (see also Chapter 3).
Since t-PA levels are subject to circadian variations, timing should be carefully
standardized and subjects should have a normal day/night rhythm. A good
clear venipuncture should be made using minimal stasis. Both free flow and
vacuum techniques can be used. The first few millilitres of blood should be
discarded. Cooling does not seem to be necessary. Any anticoagulant is suitable.
Interestingly, t-PA antigen in plasma is not essentially influenced by the pH of
the anticoagulant or by the presence of platelet-stabilizing agent (CTAD).
Anticoagulant should be consistently applied when chosen. Preparation of
plasma is not very critical as long as haemolysis and clotting in the blood is
prevented, and cellular elements are largely removed. Blood can be kept at
temperatures between 25°C and 4 °C for at least 2 days; significant losses are
encountered after 24 h when 37°C is used. For prolonged storage, plasma
aliquots can be frozen and thawed for assay. Reported snap-freezing and rapid
thawing is allowed23 •

Problems
False-positives
The presence of an excess of non-immune IgG can reduce the risk of false-
positives resulting from the presence of antibodies against IgG or rheumatoid
factors in the plasma samples.
234
 t-PA ANTIGEN
Calibrator

Potential discrepancies exist between features of Bowes melanoma t-PA and

those of endothelial t-PA as the secretory pathways of these cells are different.
t-PA secreted by Bowes melanoma cell line consists of two types differing by
three amino acid residues. In the same way recombinant t-PA through the
homogeneous polypeptide chain could be aberrantly glycosylated. The
concentration of single-chain t-PA versus double-chain t-PA should be assessed.
Care should be taken when comparing various preparations of t-PA.

Crypticity

Two major problems with current immunoassays were poor recovery and lack
of parallelism between the dilution curves of plasma and buffer. However, t-PA
in serum free cell culture media requires no manipulation for full detection of
antigen and addition of EDTA, poly-L-Iysine to plasma improves results24 •
Wun and Capuano 17 have shown that the addition of lysine or arginine (final
concentration 0.5 moVL) and slight acidification of plasma are highly efficient
in revealing the plasma t-PA antigen. This portion has been termed cryptic.
This phenomenon has been attributed to the efficiency of the reaction between
t-PA antigen in citrated plasma and immobilized t-PA-specific antibodies. The
reaction rate can be considerably enhanced by adding EDTA, lysine or acid.
It has been speculated that a large portion of the t-PA in citrated plasma is in
a complexed form. EDTA, lysine or acid would have the capacity to dissociate
these complexes, making t-PA accessible to the antibody.

COMMERCIALLY AVAILABLE KITS

A suitable assay for t-PA antigen is an ELISA using polyclonal antibodies and
peroxidase substrate (Biopool). This assay appears to be particularly reliable,
as background due to non-specific immunoglobulin, i.e. rheumatoid factors,
is eliminated by determining two optical densities: one is the presence of normal
goat IgG (N-well) and the other is the presence of goat anti t-PA IgG (A-well).
The t-PA signal corresponds to the difference between the N-well and the
A-well. In order to enhance recovery and suppress t-PA crypticity, a dilution
buffer composed of saline phosphate containing Ig/L Tween 20 and 5 mmoVL
EDTA is used. This composition is as effective as cidification or lysine. The
assay allows direct determination of the immunoreactivity of t-PA alone or
complexed with purified inhibitors, i.e. PAl-I, PAI-2, plasmin inhibitor, CI
inhibitor. Consistently precise results have been obtained by different laboratories,
the 'within' and 'between' assay coefficients being 5% and 8% respectively.
Several other ELISA kits are now available for t-PA antigen quantification
in plasma. Some use F(ab')2 antibody fragments to prevent interference from
rheumatoid factors. Others use monoclonal or polyclonal antibodies. Various
types of calibrator t-PA are used; it may be obtained from melanoma cells or
recombinant technology. Since most of these assays use an adequate buffer,
they feature low detection limits and satisfactory reproducibility (Table 25.1).
235
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Table 25.1 Leading manufacturers

Firm Kit name

Biopool Immulyse t-PA

Tintelize t-PA
Cabru t-PA assay kit
Chromogenix Coaliza t-PA
Imco t-PA assay kit
Innogenetics Innotest t-PA
Diagnostica Stago Asserachrom t-PA
Monozyme t-PA assay kit

CONCLUSION

Methods for measuring t-PA antigen in plasma have improved in the past few
years. Several techniques have been proposed but few comparative data are
available. A complete characterization of each is needed to determine reactivity
against all the forms of t-PA. This should lead to better inter-laboratory
correlation.

References
I. Bachmann F. Fibrinolysis. In: Verstraete M, Vermylen J, Lijnen R, Arnout J, eds. Thrombosis
and haemostasis. Leuven: Leuven University Press, 1987; 227--65.
2. Juhan-Vague I, Pyke SDM, Alessi MC, Jespersen J, Haverkate F, Thompson SG. Fibrinolytic
factors and the risk of myocardial infarction or sudden death in patients with angina pectoris.
Circulation 1996; 94: 2057.
3. Robertson B, Pandolfi M, Nilsson 1M. Fibrinolytic capacity in healthy volunteers at different
ages as studied by standardized venous occlusion of arms and legs. Acta Med Scand 1972;
191: 199.
4. Brommer EJP, Barrett-Bergshoeff MM, Allen RA, Schicht L, Bertina RM, Shalekamp
MADH. The use of desmopressin acetate (DDAVP) as a test of the fibrinolytic capacity of
patients, analysis of responders and non-responders. Thromb Haemostas 1982; 48: 156.
5. Cash JD. Control mechanism of activator release. In: Davidson JF, Rowan RM, Samama
MM, Desnoyers PC, eds. Progress in chemical fibrinolysis and thrombolysis, vol. 3. New York:
Raven Press, 1978; 65.
6. Juhan-Vague I, Aillaud MF, Alessi Me. Biological variation in t-PA activity and antigen. In:
KIuft C, ed. Tissue-type plasminogen activator (t-PA): physiological and clinical aspects, vol.
2. Boca Raton, FL: CRC Press, 1988; 69.
7. Holmberg L, Kristoffersson A, Lecander I, Wallen P, Astedt B. Immunoradiometric
quantification of tissue plasminogen activator secreted by fetal organs. Comparison with
urokinase. Scan J Clin Lab Invest 1982; 42: 347.
8. Mac Gregor IR, Prowse Cv. Tissue plasminogen activator in human plasma measured by
radioimmunoassay. Thromb Res 1983; 31: 461.
9. Rijken DC, Juhan-Vague I, DeCock F, Collen D. Measurement of human tissue-type
plasminogen activator by a two-site immunoradiometric assay. J Lab Clin Med 1983; 101:
274.
10. Matsuo 0, Kato K, Matsuo C, Matsuo T. Determination of tissue plasminogen activator by
an enzyme-immunoassay method. Analyt Biochem 1983; 135: 58.
II. Bergsdorf N, Nilsson T, Wallen P. An enzyme linked immunosorbent assay for determination
of tissue plasminogen activator applied to patients with thromboembolic disease. Thromb
Haemostas 1983; 50: 740.

236
 t-PA ANTIGEN
12. Urden G, Blombiick M. Determination of tissue plasminogen activator in plasma samples
by means of a radioimmunoassay. Scand J Clin Invest 1984; 44: 495.
13. Rijken DC, Van Hinsbergh VWM, Sens EHe. Quantitation of tissue-type plasminogen
activator in human endothelial cell cultures by use of an enzyme immunoassay. Thromb Res
1984; 33: 145.
14. Holvoet P, Cleemput H, Collen D. Assay of human tissue type plasminogen activator (t-PA)
with an enzyme linked immunosorbent assay (ELISA) based on three murine monoclonal
antibodies to t-PA. Thromb Haemostas 1985; 54: 684.
15. Kaizu T, Kojima K, Iwasaki K, Yamashita T. One-step sandwich enzyme-linked immunosorbent
assay of human tissue plasminogen activator using monoclonal antibodies. Thromb Res 1985;
40: 91.
16. Takada A, Shizume K, Ozawa T, Takahashi S, Takada Y. Characterization of various antibodies
against tissue plasminogen activator using highly sensitive enzyme immunoassay. Thromb
Res 1986; 42: 63.
17. Wun TC, Capuano A. Immunoradiometric quantitation of tissue plasminogen activator-
related antigen in human plasma: crypticity phenomenon and relationship to plasma fibrinolysis.
Blood 1987; 69: 1348.
18. Holvoet P, Boes J, Collen D. Measurement of free, one chain tissue-type plasminogen activator
in human plasma with an enzyme-linked immunosorbent assay based on an active site specific
murine monoclonal antibody. Blood 1987; 69: 284.
19. Amiral J, Plassart V, Grosley M, Mimilla F, Contant G, Guy-Ader AM. Measurement of
t-PA and t-PAlPAI-1 complexes by ELISA, using monoclonal antibodies: clinical relevance.
Thromb Res 1988 (Suppl. VII): 99.
20. Rijken DC, Juhan-Vague I, Collen D. Complexes between tissue-type plasminogen and
proteinase inhibitors in human plasma identified with an immunoradiometric assay. J BioI
Chem 1982; 101: 285.
21. Gaffney PJ, Curtis AD. A collaborative study to establish the 2nd international standard for
tissue plasminogen activator (t-PA). Thromb Haemostas 1987; 58: 1085.
22. Gaffney PI Specific assays of plasminogen activators and inhibitors. European School of
Hematology: Fibrinolysis, 2--6 October 1989.
23. KIuft C, Meijer P. Update 1996: Blood collection and handling procedures for assessment of
plasminogen activators and inhibitors (Leiden Fibrinolysis Workshop). Fibrinolysis 1996; 10
(Suppl. 2): 171.
24. Ranby M, Nguyen G, Scarabin PY, Samama M. Immunoreactivity of tissue plasminogen
activator and of its inhibitor complexes. Thromb Haemostas 1989; 61: 409.

237
26
Plasminogen activator inhibitor-1
(PAI-1) antigen
P. J. DECLERCK

INTRODUCTION

Plasminogen activator inhibitor-l (PAl-1) is a major inhibitor of the fibrinolytic

system. PAl-1 exerts its activity through inhibition of the plasminogen activators
(PA), t-PA (tissue-type PA) and/or u-PA (urokinase-type PA). Increased levels
of PAI-I are associated with various thrombotic diseases 1. Up to now only a
few cases of PAI-l deficiencies have been reported, and were associated with
an increased bleeding tendency2-6. In addition transgenic mice, overexpressing
human PAl -1, and PAl -1 deficient mice have further demonstrated the role of
PAI-l in thrombosis, thrombolysis, fibrin deposition and related events 7 ,8.
Initially the pathophysiological importance of PAI-I was established using
activity methods. However, over the past decade a number of immunological
assays have been developed for the sensitive and accurate measurement of
PAI-l antigen 2,9--15. Concomitant biochemical and structural research on PAI-l
revealed a unique conformational flexibility in this protein. To date at least
three different conformations (with different functional properties) of intact
PAI-l have been identified: an active form, a latent form and a substrate
form l 6-18. Yet the active and the latent conformations appear to be the most
relevant free PAI-l forms in biological fluids. The active form reacts with PA
resulting in the formation of covalent complexes (i.e t-PAlPAI-1 or u-PAlPAI-I).
In addition the active form binds to vitronectin, resulting in a stabilization of
the protein I9 ,20. Alternatively PAI-I may occur as an inactive cleaved derivative
(cleaved at PIP I'). Taken together, the following forms should be considered
in the evaluation of PAI-I antigen in biological fluids: active PAl-I, latent
PAl-I, complexed PAI-I (t-PA/PAI-l in plasma, u-PAIPAI-l in tissue-derived
samples), active PAI-l associated to vitronectin (either in plasma or in the
extracellular matrix).
239
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
During the initial phase of the development of immunological assays for
PAI-l it soon became apparent21 ,22 that, because of the large conformational
differences between these different forms, the specificity (towards particular
conformations) of the antibodies used (even with polyclonal reagents) may, at
least in part, influence the outcome of the results. This is of particular importance
considering the fact that compartmentalization of PAl-1 (i.e. plasma vs. platelets)
is associated with pronounced differences in concentration, as well as in relative
proportions, of the various conformations. Indeed, whereas both plasma and
platelets each account for approximately 50% of PAI-l activity in blood, only
10% of total PAI-l antigen is present in plasma while 90% is present in
platelets 12,14,15 (Table 26.1). Evaluation of the specific activities revealed that
in plasma PAI-l occurs mainly in the active form, whereas platelet PAI-l occurs
mainly in a latent form. As a consequence immunological determination of
PAI-l antigen in plasma requires collection of the blood, and subsequent
preparation of the plasma, under very stringent conditions 23 (see below).
Moreover, the specificity of the assay with respect to the different conforma-
tions should always be evaluated.
General principle
All methods described (see ref 24 and references therein) are either based on
the principles of a radioimmunoassay (RIA; competitive; using polyclonal
antibody preparations) or a sandwich-type enzyme-linked immunosorbent
assay (ELISA; non-competitive, using monoclonal antibodies or polyclonal
antisera or a combination of the two). An overview of the commercially available
and/or published methods and their characteristics is presented in Table 26.2.

Calibration
To date, no official calibrator for PAI-l antigen is available. Two NIBSC prepara-
tions have been made, i.e 87/512 and 921654. NIBSC 87/512, an interim standard
prepared by adding melanoma PAI-l to human plasma, contains a PAI-l
antigen level of 175 nglampoule24,25. NIBSC 92/654 was prepared by adding
recombinant PAI-l to human plasma26 and contains approximately 200 ng
PAI-l antigen/ampoule. However, a recent multi-laboratory calibration study
did not result in the assignment of a consensus value for preparation NIBSC
92/654 (reported at the ISTH-SSC Subcommittee meeting on Fibrinolysis,
Florence 1997). Anyway, for the time being these values may serve as a guide.

Table 26.1 Relative amounts of PAI-l antigen and PAI-l activity in plasma and platelets

PAI-I Blood Plasma Platelets

Antigen 100 5-10 90--95

Activity 6-15 3--6 3--9

Expressed relative to total PAI-l antigen present in blood (arbitrarily set at '100')

240
 PAI-1 ANTIGEN
Table 26.2 Available methods for the measurement of human PAI-l antigen in plasma

Type Antibody Specificity

MacGregor and Booth 10 ELISA Polyclonal Active, latent, complexed

Declerck et al. 15 ELISA Monoclonal Active, latent> complexed
Declerck et al. 21 ELISA Monoclonal Latent, complexed> active
Declerck et al. 21 ELISA Monoclonal Active, complexed» latent
Kruithof et al. 12 RIA Polyclonal Active, latent, complexed
Urden et al. 9 RIA Polyclonal Complexed :!!: active, latent
Asserachrom® PAI-! ELISA Monoclonal Active, latent, complexed
(Diagnostica Stago)
Coaliza® PAI-! ELISA Monoclonal Active, latent, complexed
(Chromogenix)
Imubind® PAI-! (American ELISA Monoclonal Active, latent:!!: complexed
Diagnostica)
Imulyse® PAI-l (Biopool) ELISA Monoclonal Active, latent:!!: complexed
PAI-! antigen (Technoclone) ELISA Monoclonal Active, latent, complexed
PAI-! Elisa kit (Monozyme) ELISA Monoclonal Latent:!!: complexed> active
Tintelize® PAI-l (Biopool) ELISA MonocionaU Active, complexed ;;. latent
polyclonal

;;'1.5-3-fold difference in reactivity; >, 3-l5-fold difference in reactivity; », 30-50-fold

difference in reactivity

Blood collection, sample preparation

Problems associated with blood collection and sample preparation for measurement
of PAl-I antigen are mainly associated with (a) the occurrence of a diurnal
variation of PAl-I levels and (b) the compartmentalization of PAl-I (i.e. platelet
PAl-I vs. plasma PAl-I). Therefore, apart from the general rules that should
be respected during blood collection and sample preparation, the following
precautions should be taken:
I. The time of sampling should be standardized throughout the course of a
study; this variable should also be considered when comparing data from
different studies.
2. Platelet activation, during blood collection as well as during sample
preparation, should be avoided. Even though not strictly necessary it is
strongly advised to add platelet-stabilizing agents to the anticoagulant23 •
3. Further processing of the sample after blood collection should be done
as soon as possible. Blood tubes should be kept on ice if preparation of
the plasma cannot be achieved within 30 min after collection of blood.
Blood is then centrifuged for 30 min at 1800g using a swing-out rotor at
4°C. After centrifugation the middle layer of the plasma is collected
carefully, avoiding 'contamination' with cellular components (e.g. platelets)
situated at the interphase. Plasma can be stored frozen ( -30°C or lower)
until analysis.

A more detailed description of the procedures for collection and handling

of blood has been described elsewhere27, and see Chapter 3.
241
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Reference
Levels of PAl-1 antigen in r.latelet-poor plasma from healthy individuals range
between 5 and 60 ng/mI 12, 4,15 and increase with age 12 . Mean PAI-l levels in
platelets are 0.3-0.7 fg/platelet I2 ,14,15 and are independent of PAI-l antigen
levels in platelet-poor plasma. During pregnancy PAI-I levels increased up to
three- to six_fold15 ,28.

Assay dependency
Initial studies had shown that PAl-1 antigen values also strongly depend upon
the assay used. Therefore it is obvious, that when reporting PAI-l antigen
values in a study: (a) the type of the assay and its characteristics should be
indicated and (b) samples from a normal, age-matched population should be
included in the analysis.
The reason for this assay dependency may be found in two parameters: (a)
the differential reactivity of the antibodies towards the different forms or (b)
the calibration of the standard. A multicentre study24 on the evaluation of
seven commercially available kits (Table 26.2) for the determination of PAI-I
antigen in plasma revealed differences ranging between 4- and 10-fold in various
normal samples. Importantly, it was demonstrated that, after normalization
of the data (relative to a well-defined, amino-acid-calibrated PAI-I sample),
for six out of seven kits no significant differences could be observed. One kit,
even after normalization of the values, yielded data that were significantly
different from the normalized data obtained with the other kits. This indicated
that the major cause of differences in 'absolute' antigen values obtained with
the different kits is mainly due to differences in calibration of the standard.
The results also suggested that, even though PAl-l can adopt various conforma-
tions, and that various monoclonal antibodies are used in the assay kits, the
potential differential reactivity of the different PAl-l forms may, in the majority
of cases, be considered to be of minor importance.
It is important to realize that these observations were obtained with
'normal' plasma samples and may be extrapolated to several clinical situa-
tions. Usually, t-PA antigen and PAI-l antigen in plasma correlate quite
well, and it has been shown that levels of active PAI-l and t-PA/PAI-l
complexes are not independent 29 • However, due to differences in relative
sensitivities between active PAl-I, t-PAlPAI-l complexes and latent PAl-I,
it cannot be excluded that in particular disease states in which the coordination
between t-PA and PAI-l levels is disturbed 3o , or in which a specific increase
in platelet PAI-l is observed 31 , the different kits may respond differently
and correlations may be less pronounced. Therefore, in particular clinical
situations putative, small differences between the kits may become much
more pronounced. A similar remark is valid for analysis of PAI-I in samples
not from plasma origin (e.g. tissue extracts, cell culture supernatants) in
which PAI-I may occur predominantly in only one form (e.g endothelial cell
culture supernatants containing mainly latent PAl-I).
242
 PAI-1 ANTIGEN
Evaluation of PAI·1 antigen from other species
A number of experimental animal models (mice, rats, rabbits, pigs) have been
developed to study thrombosis, atherosclerosis32, and are being used to evaluate
the role of PAl-I in various disease states. In general the immunological reagents
raised again human PAI-I are not suitable for the development of sensitive
assays for the specific immunological ~uantitation of PAI-I from other species.
Recently, PAI-I from other species 3-36has been expressed, purified and
characterized. Subsequently, monoclonal antibodies were raised against murine
PAl-I, rat PAI-I and porcine PAl-I, and are being applied for the construction
of monoclonal antibody-based ELISAs allowing the specific detection of PAI-I
antigen in the respective species37- 39•

References
I. Declerck PI, Iuhan Vague I, Felez I, Wiman B. Pathophysiology of fibrinolysis. I Intern Med
1994:236;425-32.
2. Schleef RR, Higgins DL, Pi11emer E, Levitt LJ. Bleeding diathesis due to decreased functional
activity of type I plasminogen activator inhibitor. I Clin Invest 1989: 83 : 1747-52.
3. Dieval I, Nguyen G, Gross S, Delobel I, Kruithof EK. A lifelong bleeding disorder associated
with deficiency of plasminogen activator inhibitor type I. Blood 1991; 77: 528-32.
4. Lee MH, Vosburgh E, Anderson K, McDonagh J. Deficiency of plasma plasminogen activator
inhibitor I results in hyperfibrinolytic bleeding. Bleed 1993; 81 : 2357-62.
5. Fay WP, Shapiro AD, Shih IL, Schleef RR, Ginsburg D. Brief report: complete deficiency of
plasminogen-activator inhibitor type 1 due to a frame-shift mutation. N Engl I Med 1992;
327: 1729-33.
6. Fay WP, Parker AC, Condrey LR, Shapiro AD. Human plasminogen activator inhibitor I
(PAl I) deficiency: characterization of a large kindred with a null mutation in the PAl I gene.
Blood 1997; 90: 204-8.
7. Erickson LA, Fici GI, Lund IE, Boyle TP, Polites HG, Marotti KR. Development of venous
occlusions in mice transgenic for the plasminogen activator inhibitor-I gene. Nature 1990;
346:74-76.
8. Carmeliet P, Stassen 1M, Schoonjans L, Ream B, van den Oord JJ, De Mol M, Mulligan RC,
Collen D. Plasminogen activator inhibitor-I gene deficient mice. II. Effects on hemostasis,
thrombosis and thrombolysis. I Clin Invest 1993; 92: 2756-60.
9. Urden G, Chmielewska I, Carlsson T, Wiman B. Immunological relationship between
plasminogen activator inhibitors from different sources. Thromb Haemostas 1987; 57: 29-34.
10. MacGregor IR, Booth NA. An enzyme-linked immunosorbent assay (ELISA) used to study
the cellular secretion of endothelial plasminogen activator (PAl-I). Thromb Haemostas 1998;
59: 68-72.
II. Rijken DC, Sprengers ED, van der Hoeven RJ, Kluft C, Engesser L. Plasminogen activator
inhibitor 1 (PAl-I) antigen levels in human plasma samples. Fibrinolysis 1988;2 (Suppl. 2) :
79-82.
12. Kruithof EK, Nicolosa G, Bachmann F. Plasminogen activator inhibitor 1: development of
a radioimmunoassay and observations on its plasma concentration during venous occlusion
and after platelet aggregation. Blood 1987; 70 : 1645-53.
13. Lund LR, Georg B, Nielsen LS, Mayer M, Dano K, Andreasen PA. Plasminogen activator
inhibitor type I: cell-specific and differentiation-induced expression and regulation in human
cell lines, as determined by enzyme-linked immunosorbent assay. Mol Cell Endocrinol 1988;
60: 43-53.
14. Booth NA, Simpson AJ, Croll A, Bennett B, MacGregor IR. Plasminogen activator inhibitor
(PAl-I) in plasma and platelets. Br J Haematol1988; 70: 327-33.
15. Declerck PI, Alessi MC, Verstreken M, Kruithof EK, Juhan Vague I, Collen D. Measurement
of plasminogen activator inhibitor I in biologic fluids with a murine monoclonal antibody-
based enzyme-linked immunosorbent assay. Blood 1988; 71: 220-5.
243
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
16. Declerck PJ, De Mol M, Vaughan DE, Collen D. Identification of a conformationally distinct
form of plasminogen activator inhibitor-I, acting as a noninhibitory substrate for tissue-type
plasminogen activator. I Bioi Chern 1992; 267: 11693-6.
17. Urano T, Strandberg L, Johansson LB, Ny T. A substrate-like form of plasminogen-activator-
inhibitor type I. Conversions between different forms by sodium dodecyl sulphate. Eur J
Biochem 1992; 209: 985-2.
18. Munch M, Heegaard CW, Andreasen PA. Interconversions between active, inert and substrate
forms of denatured/refolded type-l plasminogen activator inhibitor. Biochim Biophys Acta
1993; 1202: 29-37.
19. Declerck PJ, De Mol M, Alessi MC, Baudner S, Paques EP, Preissner KT, Muller Berghaus
G, Collen D. Purification and characterization of a plasminogen activator inhibitor I binding
protein from human plasma. Identification as a multimeric form of S protein (vitronectin). J
Bioi Chern 1988; 263: 15454-61.
20. Sigurdardottir 0, Wiman B. Complex formation between plasminogen activator inhibitor I
and vitronectin in purified systems and in plasma. Biochim Biophys Acta 1990; 1035: 56-61.
21. Declerck PJ, Verstreken M, Collen D. Measurement of different forms of plasminogen
activator inhibitor 1 (PAl-I) using various monoclonal antibody-based enzyme-linked
immunosorbent assays. Fibrinolysis 1990; 4 (Suppl. 2): 132-3.
22. Kluft C, Jie AF. Comparison of specificities of antigen assays for plasminogen activator
inhibitor I (PAl-I). Fibrinolysis 1990; 4 (Suppl. 2): 136-7.
23. Juhan Vague I, Alessi MC, Fossat C, Declerck PJ, Kruithof EK. Plasma determination of
plasminogen activator inhibitor I antigen must be performed in blood collected on antiplateletl
anticoagulant mixture. Thromb Haemostas 1987; 58: 1096.
24. Declerck PJ, Moreau H, Jesperson J, Gram J, Kluft C. Multicenter evaluation of commercially
available methods for the immunological determination of plasminogen activator inhibitor-l
(PAl-I). Thromb Haemostas 1993; 70: 858-63.
25. Lijnen HR. Report of the meeting of the subcommittee on Fibrinolysis, 6-9 July 1992,
Miinchen, Germany, 1992.
26. Gaffney PJ, Edgell TA. The international standard for plasminogen activator inhibitor I
(PAl-I) activity. Thromb Haemostas 1996; 76: 80-83.
27. Kluft C, Verheijen JH. Leiden fibrinolysis working party: blood collection and handling
procedures for the assessment of tissue-type plasminogen activator and plasminogen activator
inhibitor 1. Fibrinolysis 1990; 4 (Suppl. 2): 155-{i1.
28. Kruithof EK, Tran Thang C, Gudinchet A, Hauert J, Nicoloso G, Genton C, Welti H,
Bachmann F. Fibrinolysis in pregnancy: a study of plasminogen activator inhibitors. Blood
1987; 69: 460-{i6.
29. Alessi MC, Juhan-Vague I, Declerck PJ, Anfosso F, Gueunoun E, Collen D. Correlations
between t-PA and PAI-I antigen and activity and t-PAlPAI-1 complexes in plasma of control
subjects and of patients with increased t-PA of PAI-I levels. Thromb Res 1990; 60: 509-16.
30. Andreotti F, de Bart AC, Regan T, Maseri A, Kluft C. Factors affecting the relation between
plasminogen activator inhibitor (PAl) activity and PAI-I antigen levels in normals and in
patients with arterial disease. Fibrinolysis 1990; 4 (Suppl. 2): 34--5.
31. Alessi MC, Juhan-Vague I, Declerck PJ, Collen D. Molecular forms of plasminogen activator
inhibitor-I (PAl-I) and tissue-type plasminogen activator (t-PA) in human plasma. Thromb
Res 1991; 62: 275-85.
32. Badimon L. Models to study thrombotic disorders. Thromb Haemostas 1997; 78: 667-71.
33. Bijnens AP, Knockaert I, Cousin E, Kruithof EK, Declerck PI Expression and characterization
of recombinant porcine plasminogen activator inhibitor-I. Thromb Haemostas 1997; 77:
350-{i.
34. Ngo TH, Knockaert I, Declerck PI Expression, purification and characterization of
recombinant rat plasminogen inhibitor 1. Fibrinol Proteol 1997; II: 37-43.
35. Hofmann KJ, Mayer EJ, Schultz LD, Socher SH, Reilly CF. Purification and characterisation
of recombinant rabbit plasminogen activator inhibitor-I expressed in Saccharomyces cerevisiae,
Fibrinolysis 1992; 6: 263-72.
36. Carmeliet P, Kieckens L, Schoonjans L, Ream B, Van Nuffelen A, Prendergast G, Cole M,
Bronson R, Collen D, Mulligan RC. Plasminogen activator inhibitor-I gene-deficient mice.
I. Generation by homologous recombination and characterization. J Clin Invest 1993; 92:
2746-55.

244
 PAI-1 ANTIGEN
37. Declerck PJ, Verstreken M, Collen D. Immunoassay of murine t-PA, u-PA and PAI-1 using
monoclonal antibodies raised in gene-inactivated mice. Throm Haemostas 1995; 74: 1305-9.
38. Ngo TH, Verheyen S, Knockaert I, Declerck PI Monoclonal antibody-based immunoassays
for the specific quantitation of rat PAI-1 antigen and activity in biological samples. Thromb
Haemostas. 1998; 79: 808-12.
39. Leng H, Brouwers E, Knockaert I, Declerck PI Monoclonal antibody based immunoassays
for the specific quantitation of porcine PAI-1 antigen and activity in biological fluids. (In
preparation).

245
27
Plasminogen activity
P. J. GAFFNEY

INTRODUCTION

Plasminogen is present in plasma as a proenzyme with respect to its major

catalytic activity, namely the hydrolysis of fibrin and other components in
blood. Other activities of plasminogen (e.g. its binding to fibrin, histidine-rich
glycoprotein HRGP, and plasmin inhibitor PI), are partly or fully expressed
in plasma by the proenzyme form and will not be discussed further here. The
measurement of the plasminogen molecule in plasma has presented difficulties
for many years, due to its presence in plasma as a proenzyme and the interference
of a number of plasmin inhibitors (notably PI) with the measurement of the
activated form of the proenzyme, namely plasmin. Plasma plasminogen levels
can be measured directly or indirectly. Indirect methods are based on the
activation of the inactive plasminogen by streptokinase (SK) or urokinase
(UK) to active plasmin which, being a proteolytic enzyme, can be assayed using
a variety of methodologies and substrates. The substrates most commonly used
have been gelatin', fibrinogen-fibrin2-s, casein6-lOand esters of arginine and
lysine"- l4 . Further details of methods and substrates for the assay of plasmin
have been reviewed 's . Most direct assay methods for plasminogen are based
on its affinity for a plasminogen antiserum, and have also been reported in
the literature l 6-20. A listing of some of the assays with references is given in
Table 27.1

INDIRECT METHODS

Determination of plasminogen in plasma by an indirect method requires the

removal or inactivation of the anti-plasmins in the plasma. This can be done
by acidification 2- 9 and by precipitation with ammonium sulphate l7 or
acetone 30 . These procedures remove the antiplasmins in the supernatant, and
247
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Table 27.1 Indirect and direct methods of plasminogen assay

Direct
I. Affinity chromatography Zoltan et al. 21
2. Haemaggiutination inhibition Ludlam and Das22
3. I=unoelectrophoresis Ganrot and Nilehn l6
4. Latex flocculation Wuetal.20
5. Radial i=unodiffusion Storiko23
6. Radioi=unoassay Rabiner et al. 18
7. Amidolytic (SK-plgn complex) Friberger et al.24
Indirect
1. Caseinolytic Re=ert and Cohen6*
2. Esterolytic Troll et al. 11
3. Fibrinolytic (clot lysis) Berg et al. 25
4. Fibrinolytic (plate assay) Wolf26
5. Fluorescence Bell et al. 27 Pochron et al. 28
6. Spectrophotometric Smith et al. 29

*Many workers have since extended and modified this method particularly, and others in this
table. The references are intended to provide the source of the original or definitive method

the precipitate is reconstituted and assayed as plasmin, following an appropriate

activation procedure. It is the influence of this plasminogen activation procedure
on the resultant plasmin activity which is the greatest problem in all indirect
assays of plasminogen. Much of the problem relates to the near-impossibility
of fully activating all the plasminogen to active plasmin without plasmin
autolysis (Figure 27.1). The dynamic molecular balance between plasminogen,
plasminogen intermediate (pLG-i), active plasmin and inactive plasmin renders
it difficult to convert all the plasminogen to active plasmin without some
digestion of the active site L-chain. It has been suggested3 ! that the indirect
assay of plasminogen in plasma from the viewpoint of tedium and accuracy
is somewhat incompatible with the requirements of routine laboratory practice.

Direct assays
Table 27.1 lists some direct assays with appropriate references, should details
of these be required. Since the level of plasminogen in plasma is reasonably
constant (approximately 0.16 mglml plasma), and since the normal level is high
compared with other components of haemostasis, methods of great sensitivity
(e.g. radioimmunoassay, latex flocculation and haemagglutination inhibition)
seem unnecessary. Thus the most practical and quantitative methods are those
which can be applied to most components of blood which are present in reason-
ably large amounts; namely, radial immunodiffusion using some modification
of the procedure of Mancini et al. 35 and the immunoelectrophoretic or 'rocket'
procedure36 . Since these are described in many textbooks it is necessary here
only to emphasize their simplicity and reproducibility. Obviously the only
problem in these assays is that denatured plasminogen and degraded forms of
plasminogen will be measured as biologically intact molecules. The user must
assess the clinical situation of the patient whose plasma is being assayed before
a judgement can be made concerning the suitability of these assays.
248
 PLASMINOGEN ACTIVITY

PL. [90] PLI-j [IS]

StlP 1
+ 5 60
UK
s I
1~:~:..
I 25 A~,--

PL [85]
50 ..!!... PL 60 t
"
II
I Is
,
-
I

~ •..:r- I 25 t L
L 8
GJ
I.AnIVE ACTIVE
Figure 27.1 Schematic representation of the two-step activation of plasminogen followed by the
autodigestive step from active to inactive plasmin. This scheme was compiled from data reported
elsewhere32- 34• PLG, PLG-i PL and UK denote plasminogen, plasminogen intermediate, plasmin
and urokinase. The numbers in brackets and numbers above the polypeptide chains (denoted by
straight lines) are molecular weights x 103• Small arrows denote the approximate locations at which
urokinase cleaves plasminogen to form an intermediate and finally plasmin. Hand L denote the
heavy and light chains of active plasmin, while H' and L' are their digested equivalents in the
auto-digested inactive form of plasmin. A token number of disulphide bonds (S) is shown to
explain the total loss of only 5000 MW peptides during the activation of plasminogen, while the
molecular weights of the subunits show more significant changes. Step I shows the loss of 5000
MW peptide material during the formation of plasminogen intermediate, step 2 shows the cleavage
of an arginylvaIyl bond to form the two disulphide-bonded subunits of active plasmin, and plasmin
converts active plasmin to an inactive form by the cleavage of the light and heavy chains, most
probably at their C-terminal ends. (Taken from ref. 31).

Fortunately one method which can be described as direct and yet measures
functional activity has been developed 37 • This depends on the fact that a
streptokinase-plasminogen (SK-plgn) complex can hydrolyse the presumed
plasmin-specific chromogenic substrates S-2251 (Chromogenix) and
Chromozym-PL (Pentapharm). Since this complex is only poorly inhibited by
PI and other plasma inhibitors it is possible to convert all the plasminogen in
plasma to the SK-plgn complex following a IO-min incubation period. The
complex, as measured by the hydrolysis of chromogenic substrate, is a direct
reflection of the level of plasminogen in the test plasma. This assay obviates
the problems of plasminogen activation referred to above, and has allowed
plasminogen assays to be conducted in plasma with some degree of confidence.
Althoup a slight inhibition of the SK-plgn complex by excess SK has been
reported 3 , this does not affect the validity of assays for plasminogen when
comparing one plasma with another. Purified plasminogen preparations are
affected more by this inhibition, and this is probably the cause of the non-
parallelism found when plasma plasminogen is compared in a dose-response
manner with a purified glutamic acid-plasminogen (glu-plgn) standard (78/646)
249
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
(Figure 27.2). It was found 3 ! essential to maintain a constant plgn/SK ratio
during the dose-response analysis of test and standard preparations of
plasminogen (Figure 27.3). Thus, while recommending this procedure as the
only reliable direct plasma plasminogen assay, certain precautions must be
taken when calibrating the assay using the glu-plgn standard (78/646). Problems
have been reported when this type of assay is applied to patient plasma samples,
notably patients with disseminated intravascular coagulation (DIC) and elevated
fibrin degradation products39,40, and patients with elevated levels of fibrinogen
associated with ischaemic heart disease4 !. The fact that fibrino~en has been
shown to enhance the enzymatic activity of the SK-plgn complex 9,40 has been
used to obviate these effects and avoid an overestimation of plasminogen levels
in the plasma of patients with DIe and elevated levels of fibrinogen. The assay
to be described below was initially designed for the assay of plasminogen in
plasma using a calibrated standard for glu-plgn (78/646). Due to the effect of
fibrinogen and fibrin-degradation products on the activity of the SK-plgn
complex on chromogenic substrates we recommend that the purified plasminogen
standard should be used in the presence of 1.0 mglml of plasminogen-free
fibrinogen which is incorporated into buffer A below to give buffer B.
Furthermore, assay of plasminogen in plasma from patients undergoing
thrombolytic therapy is not recommended by this procedure. Under these

PLG GLUVSP
1.00

0.70

CD 0.40
"'"a.0
'"
CD
II:
0>
0
...J
0.10

-0.20

)(

-0.50 +----.----r-----,,----r----,----.---,
1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90

Log Dose
)( STD + UNKl

Figure 27.2 The log/log dose-response curves of the SK-plgn complexes formed in two plasmas
(+, n) compared with that formed from purified glu-plgn (x). The response variable is the rate
(x 10-3 S-I) of optical density change of S-2251 at 405 nm31

250
 PLASMINOGEN ACTIVITY

Plasminogen Assay: Incubation with SK Before Dilution

0.8

E
c::
It)
0.6
....
0

iii
"'"
'"Q;
0.

'c::co"
0> 0.4
.!:

"
"'cco"
.0
0
'"
.0 0.2
~
0>
0
....J

0.0

-0.2 +---r----,---,------r---..,---r----,r-----,
1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00
Log (Vol of PLGN/~I)

o Purified Glu Plasminogen A Test Plasma

Figure 27.3 Comparison of the log/log dose-response curves of glu-plgn standard and normal
test plasma using a consistent SK/plgn ratio in all dilutions3 !

circumstances much of the plasminogen will be converted to plasmin, which

in turn will be complexed with plasmin inhibitor (PI) (Chapter 29). Since we
know that SK will form a complex with pure plasmin as well as pure plasminogen
it would be quite difficult to interpret the data in terms of levels of intact
plasminogen in the plasma. Indeed the assay is only fully reliable in plasmas
where no great overt fibrinolysis has taken place. Recommended details of the
assay procedure are given below.

DETAILS OF RECOMMENDED DIRECT ASSAY OF PLASMINOGEN

IN PLASMA
Principle of assay
When SK is added to a solution of plasminogen (i.e. plasma) an SK-plgn
complex is formed, in which the active site of the complex is exposed to such
251
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
an extent that it can hydrolyse the chromogenic substrate H-D-Val-Leu-Lys-pNA
(S-2251 ) releasing the yellow chromophore p-nitroaniline (PNA). An alternative
chromogenic substrate known as Chromozym-PL (Pentapharm) can be used
to detect the complex. The activity of the activator complex on S-2251 is not
affected by the fibrinolytic inhibitors in plasma, notably PI. If the amount of
SK added to the sample is in excess of the plasminogen present, then an
equimolar complex (SK-plgn) is formed, the amount of which is determined
by the quantity of plasminogen. Thus, under conditions of excess SK, the rate
of pNA production is a direct reflection of the amount of plasminogen in the
sample. The pNA production can be followed by a recorder (initial rate method)
or read after stopping the reaction with acetic acid (end-point method). The
final suggested outline of this assay, given below, was an amalgam from various
reports in the literature37,42.

Equipment
Spectrophotometer (405 nm wavelength setting); semi-microcuvettes (1 ml);
centrifuge; water bath (37°C); stopwatch; disposable plastic test tubes.
Extra equipment for initial rate method: photometer cuvette housing at
37°C; recorder.

Reagents
1. Plasminogenfreefibrinogen: freeze-dried reagent from IMCO. Reconstitute
with distilled water to give a solution (2%) in 0.3 moUL NaCl.
2. Buffer A: 50 mmoVL Tris HCl-12 mmoVL NaCl. Add 6.1 g Tris and 0.7 g
NaCI to distilled water; adjust to pH 7.4 at room temperature with 1 moUL
HCl and make up to 1000 ml with distilled water. Stable for 2 months at
2-6°C.
3. Buffer B: buffer A with plasminogen-free human fibrinogen added at 10.0
mg/ml. Prepare fresh on day of use.
4. Chromogenic substrate:
Alternatives:
(a) H-D-Val-Leu-Lys-pNA, 2 HCl (S-2251)
(b) Tos-Gly-Pro-Lys-pNA, 2 HCl (Chromozym - PL)
Dissolve contents of manufacturer-supplied ampoule in sufficient distilled
water to make a 3.0 mmoVL solution. This is stable at 0-6 °C for 6 months.
Add 1 part of this solution to 8 parts buffer, mix well; this solution is
subsequently referred to as the substrate and is 333 ~oUL. The substrate
should be prepared on the day of assay and incubated at 37°C.
5. Activator (SK): streptokinase (Chromogenix) or Streptase from
Dade-Behring is reconstituted with distilled water to give a concentration
of 10 000 iulml.
6. Plasminogen standard: glutamic acid plasminogen (glu-plgn) standard (code
78/646) is available from the National Institute for Biological Standards and
Control, Potters Bar, Hertfordshire EN6 3QG, UK. It contains 10 units per
252
 PLASMINOGEN ACTIVITY
ampoule and these units are the same as those defined by the second
international reference preparation for plasmin. Reconstitute ampoule with
4 mI of buffer. This solution will be referred to as the standard, and contains
2.5 units per mI.
7. Acetic acid (500/0): this is used only in the end-point method.

Assay procedure
Establishment of a calibration curve using the standard and the assay of the
test plasma involves the same procedure; thus the construction of an assay for
each will be described simultaneously.
1. To 100 III of the standard and 100 J.1l of a 1:1 diluted buffer A test plasma
add 400 J.1l of the activator (SK) solution and incubate at 37°C for 10 min.
2. Dilute the SK test plasma and SK standard 111, 112, 113 and 114 with buffers
A and B respectively.
3. To 100 J.1l of each dilution of the SK standard and SK test plasma add
900 J.1l of the chromogenic substrate solution and do this in a manner which
is dependent on whether (a) continuous OD recording and initial rate is
available or (b) the end-point procedure is followed:
(a) Initial rate OD: use disposable plastic cuvettes and stagger the addition
of substrate so that there is time to follow the OD at 405 nm for about
1 min in a thermostatically controlled (37°C) spectrophotometer. Each
dilution of SK standard and SK test plasma will give a different OD/
min, and these data give a linear plot over a range of 0.2-2.0 units of
plasminogen. The blank used in each case is 900 III of chromogenic
substrate with 100 III of distilled water.
(b) End-point: add chromogenic substrate to each dilution of SK standard
and SK test plasma in disposable plastic test tubes and incubate for
exactly 3 min before stopping each reaction mixture by the addition of
100 J.1l of acetic acid (50%). Read the colour generated in each reaction
mixture in a 1 mI plastic cuvette (light path = 1 em). Use a blank made
up of 900 J.1l of substrate, 100 J.1l of distilled water and 100 III 50% acetic
acid.
Note: buffer B is recommended in DIe plasma samples. However, it is
worthwhile checking the assay using buffer A only, since the cost of the
buffer B reagent is quite considerable.

Calculations
1. Plot the log OD against the log of the amount of plasminogen in units
contained in each dilution of the standard measured. The graph is linear
over a wide range (0.2-2.0 units/mI) (see Figure 27.3).
2. Each dilution of the test plasma generated an OD/min reading. The four
separate values can be read from the calibration curve of the plasminogen
standard. The average of these dilution-adjusted values can be used to
express the plasminogen concentration in plasma as units of plasminogen/mI.
253
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
References
1. Christensen LR, McLeod MC. A proteolytic enzyme of serum. Characterisation, activation
with inhibitors. J Gen Physiol 1949; 28: 559-83.
2. Nilsson 1M, Skanse B, Gydell K. Fibrinolysis in Boeck's sarcoid. Acta Med Scand 1958; 159:
463-70.
3. Lassen M. The estimation of fibrinolytic components by means of the lysis time method.
Scand J Clin Lab Invest 1958; 10: 384-89.
4. Kowalski E, Kopec M, Niewiarowski S. An evaluation of the euglobulin method for the
determination of fibrinolysis. J Clin Patho11959; 12: 215-18.
5. Vignal A, Blatrix C, Steinbuch M. Fibrinogen free of plasminogen as substrate for fibrinolytic
assays. Nature, Lond 1962; 193: 693.
6. Remmert LF, Cohen P. Partial purification and properties of a proteolytic enzyme of human
serum. J Bioi Chem 1949; 181: 431-48.
7. Alkjaersig N, Fletcher AP, Sherry S. The mechanism of clot dissolution by plasmin. J Clin
Invest 1959; 38: 1086-95.
8. Derechin M. The assay of human plasminogen with casein as substrate. Biochen J 1961; 78:
443-8.
9. Hedner U, Nilsson 1M. Determination of plasminogen in human plasma by a casein method.
Thromb Diathes Haemorrh 1965; 14: 545-61.
10. Johnson AJ, Kline DL, Alkjaersig N. Assay methods and standard preparations for plasmin,
plasminogen and urokinase in purified systems, 1967-1968. Thromb Diathes Haemorrh 1969;
21: 259-72.
II. Troll W, Sherry S, Wachman 1. The action of plasmin on synthetic substrates. J Bioi Chern
1984; 208: 85-93.
12. Kline DL, Fishman JB. Plasmin. The humoral protease. Ann NY Acad Sci 1957; 68 : 25-36.
13. Lassen M. The esterase activity of the fibrinolytic system. Biochem J 1958; 69: 360--6.
14. Roberts PS. Measurement of rate of plasmin action on synthetic substrates. J Bioi Chem 1958;
232: 285-91.
15. Heimburger N. Plasminogen assay. A review and evaluation of various methods. In: Davidson
JF, Samana MM, Desnoyers PC, eds. Vol. 1. New York: Raven Press, 173-90.
16. Ganrot PO, Nilehn JE. Immunochemical determination of human plasminogen. Clin Chim
Acta 1968; 22: 335-40.
17. Hedner U, Nilsson 1M. Comparison between a direct and indirect method for determining
plasminogen. Thromb Diathes Haemorrh 1971; 26: 289-94.
18. Rabiner SF, Goldfine ID, Hart A. Radioimmunoassay of human plasminogen and plasmin.
J Lab Clin Med 1969; 74: 265-73.
19. Wu KK, Jacobsen CD, Hoak JC. Highly sensitive method for the assay of plasminogen. J.
Lab Clin Med 1973; 81: 484-8.
20. Wu KK, Jacobsen CD, Hosk JC. Assay of plasminogen using latex flocculation. Proc Soc
Exp Bioi Med 1973; 144:391-3.
21. Zoltan RP, Mertz ET, Russell HT. Assay of human plasminogen in plasma by affinity
chromatography. Clin Chem 1972; 18: 654-7.
22. Ludlam CA, Das Pc. Plasminogen assay by haemagglutination inhibition technique. J Clin
Patho11971; 24: 136-42.
23. Storiko K. Normal values for 23 different human plasma proteins determined by single radial
immunodiffusion. Blut 1968; 6: 200-8.
24. Friberger P, Knos M, Gustavsson S, Aurell L, Claeson G. Methods for determination of
plasmins, antiplasmin and plasminogen by means of substrate S-2251. Haemostasis 1978; 7:
138--45.
25. Berg W, Korsan-Bengtsen K, Y gge 1. Determination of plasminogen in human plasma by the
lysis time method. Thromb Diathes Haemorrh 1966; 16: 1-17.
26. Wolf P. Modification of the fibrin plate for measurement of the components of the fibrinolytic
system. Thromb Diathes Haemorrh 1968; 20: 50--65.
27. Bell P J, Dziobowski CT, Englert ME. A sensitive fluorometric assay for plasminogen, plasmin
and streptokinase. Anal Biochem 1974; 61: 200-8.
28. Pochron SP, Mitchell GA, Albareda I, Huseby RM, Gargiulo RJ. A fluorescent substrate
assay for plasminogen. Thromb Res 1978; 13: 733-9.

254
 PLASMINOGEN ACTIVITY
29. Smith RE, Bissell ER, Mitchell AR, Pearson KW. Direct photometric or fiuorometric assay
of proteineases using substrates containing 7-amino-4-trifiuoromethylcoumarin. Thromb Res
1980; 17:392-402.
30. Brakman P. Fibrinolysis. A standardized fibrin plate method and a fibrinolytic assay of
plasminogen. Amsterdam: Scheltema Holkema, 1967.
31. Gaffney PJ. Standardisation of plasminogen assays. Haemostasis 1988; (Suppl. 1): 47-60.
32. Rickli EE, Otavsky UI. Release of an N-terminal peptide from human plasminogen during
activation with urokinase. Biochim Biophys Acta 1973; 295: 381-4.
33. Walther PI, Steinman HM, Hill RL, McKee PA. Activation of human plasminogen by
urokinase. Partial characterization of a preactivation peptide. I BioI. Chem 1974; 249: 1173-81.
34. WIman B, Wallen P. Activation of human plasminogen by an insoluble derivative or urokinase.
Eur J Biochem 1973; 36: 25-31.
35. Mancini G, Carbonara AO, Heremans IF. Immunochemical quantitation of antigens by single
radial immunodiffusion. Immunochemistry 1965; 7: 261-4.
36. Laurell CB. Quantitative estimation of proteins by electrophoresis in agarose gel containing
antibodies. Anal Biochem 1966; IS: 45-9.
37. Friberger P, Knos M. Plasminogen determination in human plasma. In: Chromogenic peptide
substrates: chemistry and clinical usage. Edinburgh: Churchill Livingstone, 1979; 128-40.
38. Gaffney PJ, Philo RD. Measurement of some haemostatic components using biological and
amidolytic assays. In: Davidson JF, Nilsson 1M, Asted B, eds. Progress in fibrinolysis, Vol 5.
Edinburgh: Churchill Livingstone, 1981; 196-202.
39. Soria C, Soria J, Bertrand 0, Dunn F, Samama M, Bachmann F. The amidolytic activity of
the SK-plasminogen complex is enhanced by a potentiator which is generated in the presence
of vascular plasminogen activator - role of fibrin degradation products. Thromb Haemostas
1982;47: 193-6.
40. Gram J, Jespersen J. A functional plasminogen assay utilising the potentiating effect of
fibrinogen to correct for the overestimation of plasminogen in pathological plasma samples.
Thromb Haemostas 1985; 53: 255-9.
41. Gram J, Munkvad S, Jespersen J. Elevated plasma concentrations of fibrinogen assay cause
an overestimation of functional plasminogen. Thromb Haemostas 1989; 61: 154.
42. Gaffney PJ, Philo RD. A commentary on new methodology in haemostasis using chromogenic
substrates. In: Fareed J, Mesmore HL, Fenton JW, Brinkhous KM, eds. Perspectives in
haemostasis. New York: Pergamon Press, 1981; 405-17.

255
28
Plasmin inhibitor activity (previously
u2-antiplasmin)
C. KLUFT and P. MEIJER

INTRODUCTION

The glycoprotein, plasmin inhibitor, is a serine protease inhibitor of molecular

weight 65-70 kD, present in plasma l in a concentration of approximately
1 J.1mollL. The inhibitor has previously been described under various names,
including u2-antiplasmin (as in the previous edition2 and urplasmin inhibitor;
the present official name is plasmin inhibitor3 .
The protein is synthesized in the liver and has a catabolism corresponding
to a plasma half-life of about 2.5 days4,5. With respect to plasmin inhibition,
the inhibitor occurs in blood in two principal molecular forms: a plasminogen-
binding (PB) and a non-plasminogen-binding (NPB) form 6 • On average the
ratio PB:NPB is 2:17.
The inhibitor further loses its N-terminal 12 amino acid peptide in the
circulation 8,9 and acquires a capacity of crosslinking to fibrin catalyzed by
coagulation factor XIIIa8,1O.
The PB form is a very fast-acting plasmin inhibitor; NPB reacts 50-100
times more slowlyll-14. The PB form of plasmin inhibitor is responsible for
the rapid plasmin inactivation observed in plasma. The assay principle of the
immediate plasmin inhibition test, IPIT l 5-17, described in the previous edition2
was designed principally to detect the rapid inhibition in plasma and
predominantly to report on the level of PB plasmin inhibitor.
The IPIT used a reversed scheme of reagent addition (substrate before
enzyme) to obtain specificity, while the usual scheme 18 has been adapted in
most commercial methods and for automation. However, the usual scheme
may show problems with specificity19-21, especially relevant for detection of
low levels of plasmin inhibitor. The specificity problems were found to be related
mainly to the use of too-high plasmin concentrations, and subsequently it was
257
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
also possible to define a specific method using the reversed addition scheme
with low plasmin (specific plasmin inhibition test, SPIT20.22.23). This special
variant, the SPIT, and as an example its specific application in our laboratory
for one of the automates, is described in this chapter.
For any further variant, or for any other automated version of this principle,
a check should be made on whether specificity is violated. To this end, within
the framework of the SSCIISTH, a working group has defined a set of criteria
and methods of checking adherence to these criteria24•

PRINCIPLE

The assay involves two reactions:

1. Reaction of plasmin inhibitor (PI) in diluted plasma with a known excess
of plasmin
PI + plasmin ~ PI-plasmin complex + plasmin
(excess) (free)
2. Determination of the residual free plasmin by its amidolytic activity on a
synthetic tripeptide chromogenic substrate (p-nitroaniline (pNA) release
detected at 405 nm)
Plasmin + chromogenic substrate - pNA
(free)
The rate of pNA release is compared to similar data from a calibration curve
constructed by using different dilutions of pooled or reference plasma standard.
The content of the plasma standard is set at 100% or I arbitrary U/ml.

MATERIALS AND EQUIPMENT

1. STA analyser (Diagnostica Stago).
2. Plasmin. In our example (22 nkat/vial) from Chromogenix.
3. Plasmin solvent (containing 50% (v/v) glycerol and 2 mmoVL HCI).
4. Methylamine hydrochloride (Sigma).
5. Tween 80 (Merck, pro analysis).
6. Chromogenic substrate, in our example: S-2403, pyroGlu-Phe-Lys-pNA.
HCI, from Chromogenix.
7. Normal reference plasma, normal and abnormal control plasmas, com-
mercially available from Dade-Behring (CTS control plasma N; CTS control
plasma PI), Chromogenix (normal reference plasma, normal and abnormal
control plasma), Organon Teknika (verify normal citrate), Biopool (Hae-
mostasis reference plasma) and Diagnostica Stago (Stachrom antiplasmin
control).

SOLUTIONS
1. Tris buffer: 40 mmoVL Tris-HCI buffer, pH 7.4 (at 25°C) containing 86
mmoVL NaCI, 120 mmoVL methylamine and 0.008% (v/v) Tween 80.
258
 PLASMIN INHIBITOR ACTIVITY

2. Plasmin substrate stock: chromogenic substrate (8.4 mg) is reconstituted in

6 ml distilled (preferably sterile) water to give a final concentration of 2.4
mmoVL. If not contaminated by microorganisms the solution can be stored
refrigerated at 2-8 DC for 6 months, otherwise it should be stored frozen
(-20 DC). Prior to use the substrate solution should, after thawing, be kept
in ice-water for 1 h. The working solution contains 1.4 mglml.
3. Plasmin stock solution: plasmin is dissolved in 10 ml plasmin solvent and
allowed to stand for at least 30 min at room temperature before preparing
further dilutions. Further dilutions are prepared in buffer and plasmin
solvent holding a final ratio of 2: 1 (stable for 3 days at 2-8 DC). In our
example a plasmin working solution of 0.4 nkat/ml is used, and it is
ascertained that this reagent has reached the temperature of the reagent
drawer of the instrument before starting the assay.
4. In-house pooled citrated plasma: pooled plasma for the calibration curve is
obtained by mixing platelet-poor citrated plasma from 15 to 20 healthy
volunteers and is stored frozen.
5. Test sample: platelet-poor citrate plasma. No serum.

ASSAY PROCEDURE

The STA analyser performs the test as follows:

1. Plasma dilution of 1:8 with buffer.
2. Mixing 25 ~l of the diluted plasma with 175 ~l plasmin solution of 0.4
nkatlml.
3. Incubation of the mixture at 37 DC for 40 s.
4. Addition of 100 ~l chromogenic substrate working solution (2.4 mmoVL).
5. The optical density at 405 nm is read after 10 and 130 s.
6. Calculation of the results: the plasmin inhibition activity of the test sample
can be expressed as a percentage of the concentration in the calibration
plasma (100%) by comparing the changes in OD 405 between 10 and 130 s
with those of the standard curve.

(PATHO)PHYSIOLOGY

In normal individuals, aged between 20 and 50 years, and sex ratio approximately
1, plasmin inhibitor activity assayed with the IPIT showed, in our laboratory,
a narrow range: 104 ± 11% (SD), n = 1, range 85-139% without differences
between males and females. This compared well with experience with the SPIT
of 95 ± 7% (SD), n = 25, range 83-108%, and another estimation of the range
=
(80-136% n 90f5. Very similar methods consistently show a narrow normal
range for plasmin inhibitor activity with a standard deviation of 9% in a
= =
meta-anal~sis in 1982 (n 265)16 and with SD of 17% (n 50)26 and 15-17%
(n = 100)2 in some other reports. Variations with season28 and age28,29 may
exist.
The lower limit of the normal range allows a clear distinction of heterozy~otes
for plasmin inhibitor deficiency from normals and normal family members 0-32.
The upper limit resembles that used by others (l20-140%? (see above).
259
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Plasmin inhibitor is an acute-phase reactant1,34,35, and can be found to be
elevated to about 130-140% of normal. Elevated plasmin inhibitor has been
observed in some cases with thrombotic complications25 ,33,36,37 and in cases
with type II hyperlipoproteinaemia38 and progressive renal failure 39 . Reduced
plasma levels of plasmin inhibitor activity can occur due to congenital
deficiencies type I and II40,41 and is considered 'an overlooked cause of haemor-
rhage,42. Acquired deficiencies are known for thrombolytic therapy, liver diseases,
nephrotic syndrome, disseminated intravascular coagulation (in relation to
antithrombin changes43), amyloidosis, leukaemia (especially acute promyelocytic
leukaemia26,44), L-asparaginase therapy45, the postoperative period35 and
extracorporeal circulation40 . In some cases the reduction might involve
inactivation by elastase 26,44,46.

PITFALLS
1. Quality ofplasmin: the rapid interaction between plasmin and the PB-plasmin
inhibitor involves a second site interaction between the two molecules, which
requires in plasmin an intact function of the site. As reported for some types
of plasmin preparations47, plasmin can lose the functional integrity of this
site, and lose its suitability for the assay. At present most, if not all,
commercially available plasmin preparations are suitable, but outdated
material may contain larger proportions of degraded plasmin. The presence
of substantial amounts of degraded plasmin is observed as a deviation of
the calibration curve at higher inhibition levels. It is chosen to work only in
an area of 0-50010 inhibition of plasmin, and accordingly to further dilute
samples with high activity.
2. Viscosity of the plasmin solution: it should be recognized that the plasmin
solution has increased viscosity and its handling should be carefully con-
trolled, as prescribed.
3. Spontaneous activity of plasma on the chromogenic substrate: in clinical
samples such as after recent activation of fibrinolysis, plasmin captured by
u2-macroglobulin may occur in plasma. Enzymes captured br u2-macro-
globulin retain significant activity on chromogenic substrates4 . The spon-
taneous activity of plasma can in such special circumstances be elevated,
and should be subtracted from the recorded activities in the assay to avoid
underestimation of plasmin inhibitor activity. The spontaneous activity can
be easily assayed itself in the procedure described, replacing plasmin by
buffer. Alternatively, the spontaneous activity can be irreversibly inhibited
by the addition of small synthetic inhibitors (see point 4t9.
4. Activation in vitro: during and shortly after thrombolytic therapy, high levels
of thrombolytics occur in collected plasma samples. Activation of
plasminogen can continue in vitro, also at 0 °C, and can artificially reduce
plasmin inhibitor activity5o. Additions to blood and plasma of PPACK51
or GGACK52 for inactivation of t-PA and u-PA, respectively, do not disturb
the assay of plasmin inhibitor activity53.

260
 PLASMIN INHIBITOR ACTIVITY
OTHER METHODS
Technical variations of the activity assay
Several different chromogenic substrates are in use 54 • Commercially available
methods include Coamatic® Plasmin Inhibitor, Spectrolyse® a2-antiplasmin,
Actichrome® a2-antiplasmin, Berichrome® arantiplasmin, a2-antiplasmin,
Dade, arantiplasmin Instrumentation Laboratory, Chromostrate
~-antiplasmin, Orthokrome Antiplasmin, Stachrom® Antiplasmin, and Unitest
a2-antiplasmin. Protocols for the use of these methods on several instruments
are available. For each performance it should have been documented and
checked that the method is specific, according to the consensus procedures24 •
Immunochemical methods (radial and electroimmunodiffusion, enzyme
immunoassays) may assess more or all molecular forms of plasmin inhibitor.
It should be recognized that these methods may be complicated by differences
in titre of antisera for the various molecular forms, as shown for the PB and
NPB form 55 • Correlation between activity and antigen methods can accord-
ingly be quite good 56,57, but may also be poor7, depending upon the specificity
of the antibodies used.

Other functional aspects of plasmin inhibitor

Binding to fibri,.;sa
During coagulation, part of the plasmin inhibitor becomes bound to fibrin
through the action of activated factor XIII. This binding involves about 20%
of the total plasmin inhibitor when serum is formed in vitro. It was found to
be confined to the PB-form of plasmin inhibitor (35% bound), rendering the
functional assay suitable to document the effect59 . This binding can be deter-
mined as the difference between plasma and serum (or clotted plasma).

Binding to plasminogen
Binding to plasminogen can be evaluated with plasma samples using a modified
crossed immunoelectrophoresis method7 •

References
1. Wiman B. Human urantiplasmin. Methods Enzymo11981; 80: 395-408.
2. Kluft C. ~-Antiplasmin activity. In: Jespersen J, Bertina RM, Haverkate F, eds. ECAT assay
procedures: a manual of laboratory techniques. Dordrecht: Kluwer, 1992; 165-71.
3. Blombiick M, Abildgaard U, van den Besselaar AM, Clementson K1, Dahlbiick B, Exner T,
Francis cw, Gaffney P, Gralnick H, Hoyer LW et al. Nomenclature of quantities and units
in thrombosis and haemostasis (recommendation 1993). A collaborative project of the Scientific
and Standardization Committee of the International Society on Thrombosis and Haemostasis
(ISTH/SSC) and the Commission/Committee on Quantities and Units (in Clinical Chemistry)
of the International Union of Pure and Applied Chemistry-International Federation of
Clinical Chemistry (IUPAC-IFCC/CQU(Cc». Thromb Haemostas 1994; 71: 375-94.
261
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
4. Collen D, Wiman B. Turnover of antiplasmin, the fast-acting plasmin inhibitor of plasma.
Blood 1979; 53:313-24.
5. Knot EAR, Drijfhout HR, Ten Cate JW, De Jong E, Iburg AHC, Kahle LH, Grijm R.
u2-Plasmin inhibitor metabolism in patients with liver cirrhosis. J Lab Clin Med 1985; 105:
353--61.
6. Oemmensen I. Different molecular forms u2-antiplasmin. In Collen D, Wiman B, Verstraete
M, eds. The physiological inhibitors of coagulation and fibrinolysis. Amsterdam: Elsevier!
North-Holland, 1979; 131--6.
7. Kluft C, Los P, Jie AFH, Van Hinsbergh VWM, Vellenga E, Jespersen J, Henny e. The mutual
relationship between the two molecular forms of the major fibrinolysis inhibitor u 2-antiplasmin
in blood. Blood 1986; 67: 616-22.
8. Koyama T, Koike Y, Toyota S, Miyagi F, Suzuki N, Aoki N. Different NH2-terminal form
with 12 additional residues of uz-plasmin inhibitor from human plasma and culture media
of Hep G2 cells. Biochem Biophys Res Commun 1994; 200: 417-22.
9. Bangert K, Johnsen AH, Christensen U, Thorsen S. Different N-terminal forms of uz-piasmin
inhibitor in human plasma. Biochem J 1993: 291: 623-5.
10. Sumi Y, Ichikawa Y, Nakamura Y, Miura 0, Aoki N. Expression and characterization of pro
uz-plasmin inhibitor. J Biochem 1989: 106: 703-7.
II. Wiman B, Collen D. On the kinetics of the reaction between human antiplasmin and plasmin.
Eur J Biochem 1978; 84: 573-8.
12. Wiman B, Boman L, Collen D. On the kinetics of the reaction between human antiplasmin
and a low-molecular-weight form of plasmin. Eur J Biochem 1978; 87: 143--6.
13. Petersen LC, Clemmensen I. Kinetics of plasmin inhibition in the presence of synthetic
tripeptide substrate. Biochem J 1981; 199: 121-7.
14. Christensen U, Bangert K, Thorsen S. Reaction of human az-antiplasmin and plasmin.
Stopped-flow fluorescence kinetics. FEBS Lett 1996; 387: 58---{j2.
15. Teger-Nilsson AC, Friberger P, Gyzander E. Determination of a new rapid plasmin inhibitor
in human body blood by means of a plasmin specific tripeptide substrate. Scand J Clin Lab
Invest 1997; 37: 403-9.
16. Friberger P. Chromogenic peptide substrates. Their use for the assay of factors in the fibrinolytic
and the plasma kallikrein-kinin systems. Scand J Clin Lab Invest 1982; 42 (Suppl. 162): 41-7.
17. Gallimore MJ, Amundsen E, Aasen AO, Larsbraaten M, Lyngaas K, Svendsen L. Studies on
plasma antiplasmin activity using a new plasmin specific chromogenic tripeptide substrate.
Thromb Res 1979; 14: 51-60.
18. Naito K, Aoki N. Assay of u 2-plasmin inhibitor activity by means of a plasmin specific
tripeptide substrate. Thromb Res 1978; 12: 1147-56.
19. Matsuda T, Ogawara M, Miura R, Seki T, Matsumoto T, Teramura Y, Nakamara K. Selective
determination uz-plasmin inhibitor activity in plasma using chromogenic substrate. Thromb
Res 1984;33:379-88.
20. Meijer P. Westerveld W, Huisman LGM. An improved method for the detection of
u2-antiplasmin on the automated coagulation analyzer MLA Electra 1000e. Fibrinolysis
1994; 8 (SuppI.2): 160-2.
21. Woodham B (Letter to the editor) and Meijer P (Reply). Fibrinolysis 1995; 9: 129-31.
22. Meijer P, Kluft C. Criteria for an automated specific plasmin inhibitor test (SPIT) in human
plasma. Fibrinolysis 1996; 10 (Suppl. 2): 121-3.
23. Billing Clason S, Meijer P, Kluft C, Ersdal E. A novel method, Coamatic Plasmin Inhibitor,
for the specific determination of plasmin inhibitor activity in plasma. Fibrinolysis 1996; 10
(SuppI2): 165-7.
24. Meijer P, Hanss M, Christensen U, Wiman B, Kluft e. Criteria for the specific measurement
of plasmin inhibitor activity. (Submitted).
25. Engesser L. Thrombophilia. Disorders of blood coagulation and fibrinolysis. Thesis, Leiden
University, 1988.
26. Aoki N. Hemostasis associated with abnormalities of fibrinolysis. Blood Rev 1989; 3: 11-17.
27. Dick W, Cullmann W. Automatisierung eines neuen amidolytischen Verfahrens zur
u z-Antiplasmin-Bestimmung im Plasma. Lab Med 1983; 7: 51-4.
28. Stout RW, Crawford VLS, McDermott MI, Rocks MJ, Morris TCM. Seasonal changes in
haemostatic factors in young and elderly subjects. Age Ageing 1996; 25: 256-8.
29. Nowak-GOttl, Funk M, Mosch G,Wegerich B, Kornhuber B, Breddin HK. Univariate tolerance

262
 PLASMIN INHIBITOR ACTIVITY
regions for fibrinogen, antithrombin III, protein C, protein S, plasminogen and uz-antipiasmin
in children using the new automated coagulation laboratory (ACL) method. Klin Padiatr
1994; 206: 437-9.
30. Kluft C, Vellenga E, Brommer EJP, Wijngaards G. A familial hemorrhagic diathesis in a
Dutch family: an inherited deficiency of uz-antiplasmin. Blood 1982; 59: 1169-80.
31. Miles LA, Plow EF, Donnelly KJ, Hougie C, Griffm IH. A bleeding disorder due to deficiency
of uz-antiplasmin. Blood 1982; 59: 1246-51.
32. Kluft C, Nieuwenhuis HK, Rijken DC, Groeneveld E, Wijngaards G, Van Berkel W,
Dooijewaard G, Sixma JJ. uz-Antiplasrnin Enschede: dysfunctional uz-antiplasmin molecule
associated with an autosomal recessive hemorrhagic disorder. J Clin Invest 1987; 80: 1391-400.
33. Kluft C, Leebeek FWG. az-Antiplasmin and thrombosis: results of a questionnaire. Fibrinolysis
1988: 2 (Suppl. 2): 47-8.
34. Matsuda M, Wakabayashi K, Aoki N, Morioka Y: uz-Antiplasmin inhibitor is among acute-
phase reactants. Thromb Res 1980; 17: 527-32.
35. Kluft e. Fibrinolytic shut-down after surgery. In: Sawaya R, ed. Fibrinolysis and the central
nervous system. Philadelphia, PA: Hanley & Belfus, 1990; 127-40.
36. Brommer EJP, Gevers Leuven JA, Kluft C, Wijngaards G. Fibrinolytic inhibitor in type II
hyperlipoproteinaemia. Lancet 1982; 1:1066.
37. RatnoffOD. The role of haemostatic mechanisms. Clin Haematol1981; 10: 261-81.
38. Lowe GDO, Stromberg P, Forbes CD, McArdle BM,Lorirner AR, Prentice CRM. Increased
blood viscosity and fibrinolytic inhibitor in type II hyperlipoproteinaemia. Lancet 1982; I:
472.
39. Gordge MP, Faint RW, Rylance PB, Kluft C, Nield GH. Abnormal fibrinolysis in progressive
renal failure due to a reduction in circulating tissue plasminogen activator. Br J Haematol
1988; 69: 133.
40. Saito H. uz-Plasmin inhibitor and its deficiency states. J Lab Clin Med 1988; 112: 671-8.
41. Ikernatsu S, Fukutake K, Aoki N. Heterozygote for plasmin inhibitor deficiency developing
hemorrhagic tendency with advancing age. Thromb Res 1996; 82: 129-36.
42. Griffin GC, Mammen EF, Sokol RJ, Perrotta AL, Stoyanovich A, Abildgaard CF.
uz-Antiplasmin deficiency. An overlooked cause of hemorrhage. Am J Pediatr Hematol Oncol
1993; 15: 328-30.
43. Okajima K. Clinical relevance of determination of plasma ATIII and uz-PI activities in
patients with DIC. Application of the molecular markers for the analysis of pathophysiology
of DIe. Jpn J Clin Patho11994; 42: 45-55.
44. Avvisati G, Ten Cate Jw, Sturk A, Lamping RJ, Petti MC, Mandelli F. Acquired az-antip1asrnin
deficiency in acute promyeloytic leukaemia. Br J Haematol1988; 70: 43-8.
45. Vellenga E, Kluft C, Mulder NH, Wijngaards G, Nieweg HO. The influence of L-asparaginase
therapy on the fibrinolytic system. Br J Haematol1984; 57: 247-54.
46. Brower MS, Harpel PC. Proteolytic cleavage and inactivation of uz-plasmin inhibitor and
CI-inactivator by human polymorphonuclear leukocyte elastase. J Bioi Chern 1982; 257:
9849-54.
47. Kluft C, Traas DW, Jie AFH, Hoegee-de Nobel E. The suitability of various plasmin prepara-
tions for the functional assay of uz-antiplasmin in plasma. Thromb Haemostas 1982; 48:
320-4.
48. Gyzander E, Teger-Nilsson AC. Activity of the uz-macroglobulin-plasrnin complex on the
plasmin specific substrate H-o-Val-Leu-Lys-p-nitroanilide. Thromb Res 1980; 19: 165-75.
49. Gram J, Jespersen 1. On the significance of antithrombin III, uz-rnacroglobulin, az-antiplasrnin,
histidine-rich glycoprotein, and protein C in patients with acute myocardial infarction and
deep vein thrombosis. Thromb Haemostas 1985; 54: 503-5.
50. Rijken DC, Seifried E, Barrett-Bergshoeff MM, Dooijewaard G. Plasminogen activation at
low temperatures in plasma samples containing therapeutic concentrations of tissue-type
plasminogen activator or other thrombolytic agents. Thromb Haemostas 1990; 64: 47-52.
51. Seifried E, Tanswell P. Comparison of specific antibody, D-Phe-Pro-Arg-CHz CI and aprotinin
for prevention of in vitro effects of recombinant tissue-type plasminogen activator on
haemostasis parameters. Thromb Haemostas 1987; 58: 921--{).
52. Lijnen HR, Van Hoef B, Collen D. Differential reactivity of Glu-Gly-Arg-CHz CI, a synthetic
urokinase inhibitor, with single-chain and two-chain forms of urokinase-type plasminogen
activator. Eur J Biochem 1987; 162: 351--{).

263
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
53. Oethinger MD, Seifried E. In vitro effects of urokinase - prevention by different inhibitors.
Thromb Haemostas 1990; 64: 402-6.
54. Kluft C, Jie AFH, Los P, Dooijewaard G, Traas DW. Expression of the two forms of
urantiplasmin in functional and immunochemical assays. In: Davidson JF, Bachmann F,
Bouvier CA, Kruithof EKO, eds. Progress in fibrinolysis, vol VI. Edinburgh: Churchill
Livingstone, 1983: 386--7.
55. Svendsen LG, Fareed J, Walenga JM, Hoppensteadt D. Newer synthetic peptide substrates
in coagulation testing: some practical considerations for automated methods. Semin Thromb
Hemostas 1983; 9: 250-62.
56. Matsuda T, Ogawara M, Miura R, Seki T, Matsumoto T, Teramura Y, Nakamara K. Selective
determination of ~-p1asmin inhibitor activity in plasma using chromogenic substrate. Thromb
Res. 1984; 33:379-88.
57. Aoki N, Yamanaka T. The u2-plasmin inhibitor levels in liver diseases. Clin Chim Acta 1978;
84: 99-105.
58. Sakata Y, Aoki N. Cross-linking of ~-plasmin inhibitor to fibrin by fibrin-stabilizing factor.
J Clin Invest 1980; 65: 290-7.
59. Kluft C, Los P, Jie AF. The molecular form of u2-antiplasmin with affmity for plasminogen
is selectively bound to fibrin by factor XIII. Thromb Res 1984; 33: 419-25.

264
29
Plasmin-a2 -antiplasmin complexes
(plasmin-plasmin inhibitor
complexes)
E. HATTEY, M. HAUMER, M. R. GRIFFITHS, V. CARROLL and
B. R. BINDER

INTRODUCTION

u2-Antiplasmin (new name: plasmin inhibitor) is the most important plasmin

inhibitor. Its rapid reaction with plasmin results in the formation of an inactive
complex composed of one molecule of each component. Two steps are involved
in this process: first a reversible complex is formed between the lysine-binding
sites of plasmin and complementary sites on the carboxyterminal end of the
plasmin inhibitor molecule. In a second step an irreversible complex is generated,
associated with the cleavage of a peptide bond in the inhibitor!. During
fibrinolysis and thrombolysis an equilibrium exists between formation of
plasmin-plasmin inhibitor complexes, occurring preferentially in the fluid phase,
and binding and action of plasmin on the fibrin surface. Bound to fibrin,
plasmin is protected against inhibition by plasmin inhibitor2,3. However,
whenever fibrin is completely dissolved the plasmin liberated from the fibrin
surface is immediately complexed by plasmin inhibito~,3.
Whenever fibrin forms in the circulation, this process will be accompanied
by activation of the fibrinolytic system, because of the well-known effects of
fibrin on tissue plasminogen activator (t_PA)4,5. Plasmin generated thereby will
in part become complexed with u2-antiplasmin, leading to increased levels of
plasmin-a2-antiplasmin (PAP) complexes (new name: plasmin-plasmin inhibitor
(PPI) complexes). Such increased levels of PPI complexes have therefore been
found in many situations in which fibrin formation is increased. In severe cases
of disseminated intravascular coagulation (DIC) as well as during thrombolytic
therapy, extensive activation of the fibrinolytic system can lead to massive
plasmin formation in the fluid phase and maximal PPI complex formation 6 ,7.
265
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Especially non fibrin-specific plasminogen activators such as streptokinase or
urokinase can cause complete consumption of plasmin inhibitors, resulting in
an increased bleeding tendency due to plasminaemia. Thereby plasmin action
is no longer restricted to its specific substrate fibrin but extended to non-specific
substrates such as fibrinogen and other coagulation factors. Determination of
plasmin-plasmin inhibitor complexes can therefore serve as an indicator of
general plasminaemia during hyperfibrinolytic states accompanied by fibrinogen
and plasmin inhibitor consumption and possible bleeding tendency.
In recent years analysis of PPI complex formation has rained increasing
importance for the prognosis of various tumour diseases6- 1. In addition, in
several cases of vascular diseases or thrombophilia - as in old age, myocardial
infarction, stroke or DIc 12- 25 - determination of PPI complexes, or rather the
ratio of thrombin-antithrombin (TAT) comrlexes to PPI complexes (TATIPPI
ratio), seems to be of prognostic relevance 3,26-28. Last but not least, evalu-
ation of the PPI content is used, among others, as a marker for activation of
the haemostaticlfibrinolytic system under various experimental conditions29-33.

METHODS OF ASSAY

For the detection of PPI complexes in plasma most investigators use a com-
bination of antibodies against the two complex-forming proteins in a sandwich
system, with either radioactive or enzymatic labelling. There are currently
several ELISA systems commercially available (Dade-Behring, Technoclone,
Teijin). In addition several investigators use their own in-house test systems,
e.g. RIA34 or ELlSA35 in the human system, and for studies in animal models36 .
Thereby the use of different antibodies, different standards, varying test condi-
tions (incubation temperature, addition of inhibitors) has led to discrepancies
in the literature with respect to the normal range values (Table 29.1). Despite
the great difference between some of the reported baseline values Meijer et al.
found a significant correlation when the same samples were tested with the
three commercially available assay systems37 . It is important to note that the
mean baseline values obtained with these assays (Technoclone: 142 ng/ml,
Dade-Behring: 368 ng/ml, Teijin: 758 ng/ml) correspond with the respective
normal values published for these assays, and reflect the same baseline fibrinolytic
activities. The difference in the standards used, and perhaps in some cases even
the small sample size, seem to account for the variation in the reported basal
values for healthy donors. However, a systematic difference might exist between
a number of publications reporting rather lower mean baseline levels and others
with rather higher baseline levels. In the latter cases higher sample dilutions
(1:50-1:800) were used, thereby shifting the equilibrium between plasminogen
activators and inhibitors in plasma towards plasminogen activation ('diluting
out inhibitors'). If such diluted samples are then incubated for a prolonged
time (1 h or more) at 37°C during the assay procedure increased PPI formation
might occur ex vivo. We therefore recommend addition of inhibitors to the
sample dilution buffer and incubation at lower temperature, or only for a short
time. The avoidance of PPI formation ex vivo is especially important when the
TATIPPI ratio is determined.
266
 Table 29.1 Reference values reported for various PPI kits

Company name Reported values Range Sample Fibrinolysis inhibitors Incubation Sample Reforence
"tJ
dilution size (n)
):
C/J
Commercial Assays :s:::
Technoclone l31 nglml (median) 10-514 nglml 1:10 Aprotinin 2000 KIU/mI, 4°C overnight 87 Carroll et al. 41 Z
benzamidine 20 mmollL ~
II)

Technoclone 142 nglml (mean) 1:10 As above As above 113 Meijer et al. 37 ):.
Z
Dade-Behring 210 ± 88 (mean) 80-470 nglml 1:2 Aprotinin 10 KIU/mi 15 min at 37°C 178 Pelzer et al. 42 --I
I\)
:;;
!?l Dade-Behring 290 nglml (median) 120-700 nglml 1:2 None 15 min at 37°C 466 Company
instructions ~:s:::
Teijin 758 nglml (mean) 1:800 None? 1 h at 37°C 113 Meijer et al. 37 Z
()
Non-commercial assays 0
:s:::
RIA 3.5 nmollL (mean) 2-8nmollL 1.5 SBTI 100 Ilglml, 4hatRT* 15 Levi et al. 34 "tJ
I
(-490 nglml) (280-1120 nglml) benzamidine 10 mmoVL m
X
ELISA 573 ± 131 nglml 350-790 nglml 1:50 EACAO.l mollL 1 h at 37°C 30 Montes et al. 35 m
(mean) C/J

*RT =room temperature

 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

As mentioned above, there are several test kits commercially available for
the determination of PPI complexes in plasma. For those who wish to establish
their own test system the following procedure is described:

Test principle
The test described here is a solid-phase enzyme immunoassay in which MPW7AP,
a specific monoclonal antibody directed against the neoantigen of the PPI
complex, is adsorbed on plastic microtitre plates38,39. During incubation with
test samples PPI complexes are selectively bound, and after washing away
unbound material the complexes are detected by MPW2PG POX, a peroxidase-
labelled monoclonal antibody against the kringle 1-3 region of the plasmin(ogen)
part of the complex40 • Quantification of labelled antigen-antibody conjugates
is achieved by ABTS, a chromogenic substrate for peroxidase.

Materials and equipment

1. Flat-bottom microtitre plates (96-well) with high binding capacity (e.g. NUNC
immunoplate maxisorp 4-93454 or GREINER ELISA plates No. 655061),
plate sealers (e.g. COSTAR 3095).
2. ELISA reader for 405 nm and 492 nm wavelength (e.g. Anthos reader
2001, Fresenius Pharma).
3. Antibody to PPI neoantigen MPW7AP and antibody to kringle 1-3 region
of plasmin(ogen) POX-labelled, MPW2PG POX (Technoclone).
4. PPI complex standard plasma and PPI complex depleted plasma (Tech-
noclone).
5. ABTS [2-2' azinobis(3-ethylbenz-thiazolinesulphonic acid)] from Boehringer
Mannheim).
6. Aprotinin (TrasyloIR) from Bayer.
7. 2-[(ethylmercury(n)-thio]-benzoic acid sodium salt (ThimerosaIR) 818957,
Merck.
8. Polyoxyethylene sorbitan monooleate (Tween 20) P 13 79, Sigma.
9. Serum albumin bovine, purified (BSA) ORHO 20/21, Dade-Behring.
10. Benzamidine chloride 820122, Merck.

Solutions
I. Coating buffer: 1.59 g Na2C03.1 OH20, 2.93 g NaHC03, 100 mg thimerosal
with distilled water to I L, pH 9.6.
2. Coating solution: 20 Ilg/ml MPW7AP in coating buffer; 100 j.tlIwell.
3. Phosphate buffered saline (PBS): 8 g NaCI, 0.2 g KH 2P0 4 , 1.44 g
Na2HP042H20 with distilled water to I L, pH 7.4.
4. Washing buffer: PBS containing 0.5% Tween 20.
5. Blocking solution: PBS containing 1% BSA.
6. Dilution buffer I (DBI): PBS containing 1%(w/v) BSA, 2000 KIV/ml
aprotinin and 20 mmol IL benzamidine chloride or a specific inhibitor for
t-PA, e.g. PPACK.
268
 PLASMIN-a2-ANTIPLASMIN COMPLEXES
7. Dilution buffer 2 (DB2): OBI containing 1% PPI depleted plasma.
8. Dilution buffer 3 (DB3): OBI containing 10% PPI depleted plasma.
9. Detecting solution: 10 Ilg/ml MPW2PG POX in OBI; 100 Ill/well.
10. Substrate buffer: 1.29 g citric acid monohydrate, 1.375 g Na2HP042H20
with distilled water to 100 ml, pH 4.0.
11. Substrate solution: 1 mg ABTS/ml solution containing 1 III H 20 2 30%/ml
solution in substrate buffer; 100 Ill/well.
12. Stop solution: 320 mg NaFIlOO ml distilled water; 100 Ill/well.

Stability of the reagents

Coating buffer and coated plates containing an antimicrobial substance are

stable at 4°C. Other protein-containing solutions and washing buffer should
be prepared freshly, or kept under sterile conditions to prevent microbial growth.
To avoid adverse reactions with the peroxidase, antibiotics should not be used
in these cases.

Procedure
Preparation of plates

The wells of an ELISA plate are filled with 100 Ill/well coating solution,
preferably by use of a multichannel micropipette. The plate should remain
covered with a self-adhesive plastic foil for at least 16 h at 4 °C but can also be
stored that way for a prolonged period. Before use, the plate is emptied and
refilled with 100 Jll/well of blocking solution and incubated for 1 hat 37°C to
block excessive reactive groups on the plate surface. For all incubation steps
the plate remains covered with the plastic foil.

Sample preparation

Normal citrated or ethylenediaminetetra-acetic acid (EOTA) plasma can be

used, but one should be aware that uninhibited plasminogen activators will
lead to formation of PPI complexes in vitro. Especially in monitoring
thrombolytic therapy, blood samples should therefore always be collected into
anticoagulants containing inhibitors, e.g. 2000 KIVlml aprotinin + 20 mmollL
benzamidine final concentrations. The plasma samples are diluted 1: lOin OBI
for low concentrations and 1:100 for high concentrations of PPI complexes;
100 Ill/well are needed and at least duplicate determinations are recommended.

Standard preparation

PPI complex standard plasma is reconstituted with distilled water. Serial

dilutions from 1:100 to 1:800, and blanks, are prepared in DB3 for use with
low PPI concentration samples and in DB2 for use with samples containing
high PPI concentrations; again 100 Jll/well and duplicates are also necessary.
269
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Washing of the plates
Between the incubation steps the plates are washed three times with approxi-
mately 300 ~Vwell washing buffer, either manually or by use of an automatic
plate washer. After each emptying step the plate is carefully tapped dry on
absorbent paper towels.

Flow sheet of the procedure

1. Coating of ELISA plates overnight at 4°C.
2. Incubation with blocking solution 1 hat 37°C.
3. Washing step.
4. Incubation of samples overnight at 4 0C.
5. Washing step.
6. Incubation with detecting solution 2 h at 37°C.
7. Washing step.
8. Incubation with substrate solution 30 min at room temperature, protected
from light.
9. Addition of stop solution.
10. Reading of the plate in an ELISA reader with 405 nm test wavelength and
492 nm reference wavelength within 2 h.
11. Plotting of standard values on linear scale and reading of sample values
from the respective standard curve for high and low PPI concentrations;
multiplication with the dilution factor.

Evaluation of results
For evaluation of the results the reading of the samples is compared with those
of a reference PPI complex preparation.

Standardization
Since PPI complexes are not easily prepared in a stable purified form - the
complex is susceptible to proteolytic degradation and different epitopes might
be generated whether the complex is formed in excess of plasmin or of inhibitor l
- we decided to prepare PPI complexes directly in plasma and to calibrate them
with PPI complexes generated from purified components. For this purpose
citrated plasma was incubated with more than saturating concentrations of
urokinase to produce maximum amount of PPI complexes. Thereafter the
reaction was terminated by addition of benzamidine and Trasylol. The complexes
so formed are stable either frozen at -70°C or lyophilized at 4 0C. To determine
the actual amount of complexes in the standard plasma the readings in the
ELISA assay were compared with those of PPI complexes generated from
purified components. For this purpose plasmin was prepared freshly from
purified gIu-plasminogen by incubation with urokinase bound to Sepharose
270
 PLASMIN-a2 -ANTIPLASMIN COMPLEXES

and, after removal of urokinase-Sepharose, the resulting plasmin activity was

determined with S-22S1 immediately and after 30 min incubation with purified
<l2-antiplasmin at 37 °e. The differences in plasmin activity represent the
amount of PPI complexes formed (Figure 29.1).

Specificity
The described test is specific for PPI complexes and allows full plasma recovery,
provided the standard was diluted in PPI-depleted plasma in the same con-
centration as the plasma samples to be tested.

Difficulties, source of error, trouble-shooting

The major source of error in the PPI ELISA described lies in the preparation
of samples and a possible PPI complex formation ex vivo. Therefore, care has
to be taken to collect blood samples into a suitable inhibitor whenever
plasminaemia or plasmin formation after blood collection is expected. See
section 'Sample preparation' under Procedure.

2.0

1.8

1.6

1.4

1.2

Plasmin Plasmin
0.6 + +
Buffer Plasmin inhibitor

0.4
... PPI-Complex prepared in Plasma
0.2 • PPI-Complexes from purified
Componen1s

0.1 0.2
IAQhnl Complexes
Figure 29.1 Standard curves obtained with the described test. Comparison of PPI complexes
prepared in plasma with PPI complexes from purified components. In the insert the arnidolytic
determination of plasmin activity from freshly prepared plasmin incubated with or without excess
of u2"antiplasmin is shown. From the differences in plasmin activity (plasmin inhibited) the
concentration of formed complexes is calculated

271
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

Sensitivity and reference values

In this assay the lower detection limit is 10 ng/ml in purified systems as well
as in plasma.
1. Reference values: in a normal control population PPI values with a median
of 131 ng/ml and a range from 10 to 514 ng/ml are found; in serum samples
PPI values are mainly increased (see also Table 29.1).
2. Pathological values: thrombolytic therapy results in PPI plasma values of
> 800 ng/ml. In patients with coronary heart disease treated successfully by
percutaneous transluminal coronary angiography (PTCA) PPI values were
found to be increased 12 months thereafter, whereas in male survivors of
myocardial infarction there was a low but significant decrease of PPI
complexes in plasma (Figure 29.2)41.

Control and calibration procedures

There is no internationally recognized PPI standard available at the moment.
Therefore, standardization should be done by comparison with one of the
commercially available reference plasmas with given PPI content, or in a similar
manner as described in the 'Standardization' section, under Evaluation of
Results.

Comparison with data from other studies

As can be seen from Table 29.l, and from the introduction to the 'Methods'
section, levels of PPI complexes obtained with commercially available methods

p<O.05

•
• •

• •
••
~
•• • •

I·• ..
is.
E • I ••
8 • •
• ••• • - • •
•• •••
-I!••• I,

0
-HIP.1
II
Mlmale
••
Ii
:.
Normal male
••
PTCA
Figure 29.2 PPI complex values in male survivors of myocardial infarction and in matched
controls. In comparison values from patients 12 months after successful PTCA are shown (median
values indicated)

272
 PLASMIN-a 2 -ANTIPLASMIN COMPLEXES

correlate in principle with each other. The differences in the levels of PPI
complexes might, however, be due to different standards used, as well as to
PPI complex formation ex vivo.

References
1. Wiman B. Human alpha-2 antiplasmin. Methods Enzymo11981; 80: 395-408.
2. Sakata Y, Aoki N. Significance of cross-linking of alpha 2-plasmin inhibitor to fibrin in
inhibition of fibrinolysis and in hemostasis. J Clin Invest 1982; 69: 536-42.
3. Aoki N, Harpel PC. Inhibitors of the fibrinolytic enzyme system. Semin Thromb Hemostas
1984;10: 24--41.
4. Hoylaerts M, Rijken DC, Lijnen HR, Collen D. Kinetics of the activation of plasminogen
by human tissue plasminogen activator. Role of fibrin. J BioI Chern 1982; 257: 2912-19.
5. Beckmann R, Geiger M, Binder BR. Plasminogen activation by tissue plasminogen activator
in the presence of stimulating CNBr fragment FCB-2 of fibrinogen is a two-phase reaction.
Kinetic analysis of the initial phase of slow plasmin formation. J BioI Chern 1988; 263:
7176--80.
6. Urano T, Kamiya T, Sakaguchi S, Takada Y, Takada A. Fibrinogenolysis and fibrinolysis in
normal volunteers and patients with thrombosis after infusion of urokinase. Thromb Res
1985; 39: 145-55.
7. Levi M, Roem D, Kamp AM, de Boer JP, Hack CE, ten Cate JW. Assessment of the relative
contribution of different protease inhibitors to the inhibition of plasmin in vivo. Thromb
Haemostas 1993; 69: 141-6.
8. Gabazza EC, Taguchi 0, Yamakami T, Machishi M, Ibata H, Suzuki S. Evaluating
prethrombotic state in lung cancer using molecular markers. Chest 1993; 103: 196-200.
9. Okajima K, Kohno I, Soe G, Okabe H, Takatsuki K, Binder BR. Direct evidence for systemic
fibrinogenolysis in patients with acquired alpha 2-plasmin inhibitor deficiency. Am J Hematol
1994; 45: 16-24.
10. Gabazza EC, Taguchi 0, Yamakami T et al. Alteration of coagulation and fibrinolysis systems
after multidrug anticancer therapy for lung cancer. Eur J Cancer 1994; 30A: 1276-81.
11. Nanninga PB, van Teunenbroek A, Veenhof CH, Buller HR, ten Cate JW. Low prevalence
of coagulation and fibrinolytic activation in patients with primary untreated cancer. Thromb
Haemostas 1990; 64: 361-4.
12. Mari D, Mannucci PM, Coppola R, Bottasso B, Bauer KA, Rosenberg RD. Hypercoagulability
in centenarians: the paradox of successful aging. Blood 1995; 85: 3144-9.
13. Uenomachi H, Sonoda M, Miyauchi T et al. Relationship between intracoronary thrombolysis
and fibrino-coagulation - special reference to TATIPIC and FPAlPIC. Jpn Circ J 1996; 60:
149-56.
14. Dietrich W Reducing thrombin formation during cardiopulmonary bypass: is there a benefit
of the additional anticoagulant action of aprotinin? J Cardiovasc Pharmacol1996; 27 (Suppl.
1): S50-7.
15. Kawasuji M, Ueyama K, Sakakibara N et al. Effect of low-dose aprotinin on coagulation
and fibrinolysis in cardiopulmonary bypass. Ann Thorac Surg 1993; 55: 1205-9.
16. Malyszko J, Malyszko JS, Pawlak D, Pawlak K, Buczko W, Mysliwiec M. Hemostasis, platelet
function and serotonin in acute and chronic renal failure. Thromb Res 1996; 83: 351-61.
17. Jansen PM, Boermeester MA, Fischer E et al. Contribution of interleukin-l to activation of
coagulation and fibrinolysis, neutrophil degranulation, and the release of secretory-type
phospholipase A2 in sepsis: studies in nonhuman primates after interleukin-l alpha
administration and during lethal bacteremia. Blood 1995; 86: 1027-34.
18. Pedersen OD, Gram J, Jespersen 1. Plasminogen activator inhibitor type-I determines plasmin
formation in patients with ischaemic heart disease. Thromb Haemostas 1995; 73: 835-40.
19. Agnelli G, Iorio A, Parise P, Goldhaber SZ, Levine MN. Fibrinogenolysis and thrombin
generation after reduced dose bolus or conventional rtPA for pulmonary embolism. Blood
Coagul Fibrinolysis 1997; 8: 216-22.
20. Nielsen EW, Johansen HT, Hogasen K, Wuillemin W, Hack CE, Mollnes TE. Activation of
the complement, coagulation, fibrinolytic and kallikrein-kinin systems during attacks of
hereditary angioedema. Scand J Immunol1996; 44: 185-92.

273
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
21. Handa K, Takao M, Nomoto J et al. Evaluation of the coagulation and fibrinolytic systems
in men with intermittent claudication. Angiology 1996; 47: 543-8.
22. Handa K, Kimoto K, Kawaguchi H et al. Plasmin and thrombin inhibitors in essentially
untreated patients with coronary artery spasm. Int J Cardiol 1993; 42: 263-7.
23. Cugno M, Hack CE, de Boer JP, Eerenberg AJ, Agostoni A, Cicardi M. Generation of plasmin
during acute attacks of hereditary angioedema. J Lab Clin Med 1993; 121: 38-43.
24. Quehenberger RP, Loner U, Kapiotis S et al. Effects of the third generation oral contracep-
tives containing newly developed progestagens on fibrinolytic parameters. Fibrinol Proteol
1997; 11: 97-101.
25. Quehenberger P, Kapiotis S, Partan C et al. Studies on oral contraceptive-induced changes
in blood coagulation and fibrinolysis and the estrogen effect on endothelial cells. Ann Hematol
1993; 67: 33-6.
26. Takahashi H, Tatewaki W, Wada K, Hanano M, Shibata A. Thrombin vs. plasmin generation
in disseminated intravascular coagulation associated with various underlying disorders [see
comments]. Am J Hemato11990; 33: 90-5.
27. Kario K, Matsuo T. Lipid-related hemostatic abnormalities in the elderly: imbalance between
coagulation and fibrinolysis. Atherosclerosis 1993; 103: 131-8.
28. Kario K, Matsuo T, Kodama K, Matsuo M, Yamamoto K, Kobayashi H. Imbalance between
thrombin and plasmin activity in disseminated intravascular coagulation. Assessment by the
thrombin-antithrombin-III complexlplasmin-alpha-2-antiplasmin complex ratio. Haemostasis
1992; 22: 179-86.
29. Merryman P, Tannenbaum SH, Gralnick HR et al. Fibrinolytic and coagulant responses to
regional limb perfusions of tumor necrosis factor, interferon-gamma, andlor melphalan.
Thromb Haemostas 1997; 77: 53-6.
30. Vanderpoll T, Coyle SM, Levi M et al. Effect of a recombinant dimeric tumor necrosis factor
receptor on inflammatory responses to intravenous endotoxin in normal humans. Blood 1997;
89: 3727-34.
31. Pajkrt D, Vanderpoll T, Levi M et al. Interleukin 10 inhibits activation of coagulation and
fibrinolysis during endotoxinemia. Blood 1997; 89: 2701-5.
32. Corssmit EP, Levi M, Hack CE, ten Cate Jw, Sauerwein HP, Romijn JA. Fibrinolytic response
to interferon-alpha in healthy human subjects. Thromb Haemostas 1996; 75: 113-17.
33. van Deventer SJ, Buller HR, ten Cate JW, Aarden LA, Hack CE, Sturk A. Experimental
endotoxemia in humans: analysis of cytokine release and coagulation, fibrinolytic, and
complement pathways. Blood 1990; 76: 2520-6.
34. Levi M, de Boer JP, Roem D, ten Cate Jw, Hack CEo Plasminogen activation in vivo upon
intravenous infusion of DDAVP. Quantitative assessment of plasmin-alpha 2-antiplasmin
complex with a novel monoclonal antibody based radioimmunoassay. Thromb Haemostas
1992; 67: 111-16.
35. Montes R, Paramo JA, Angles Cano E, Rocha E. Development and clinical application of a
new ELISA assay to determine plasmin-alpha 2-antiplasmin complexes in plasma. Br J
Haematol1996; 92: 979-85.
36. de Boer JP, Creasy AA, Chang A et al. Activation patterns of coagulation and fibrinolysis in
baboons following infusion with lethal or sublethal dose of Escherichia coli. Circ Shock 1993;
39: 59-67.
37. Meijer P, Kamerling SWA, van de Ham FJ et al. Baseline levels of alpha-2-antiplasmin-
plasmin complex in human plasma. Fibrinolysis 1994; 8 (Suppl. 2): 124-5.
38. Hattey E, Wojta J, Huber K, Binder BR. Development and evaluation of ELISA systems for
determination of plasmin-alpha-2-antiplasmin (PIP) complexes in plasma. Fibrinolysis 1986;
(Suppl.) 1: A152.
39. Hattey E, Beckmann R, Krutisch G, Huber K. Evaluation of limited activation of the plasma
fibrinolytic system during thrombolysis by rt-PA. Fibrinolysis 1988; 2 (Suppl.l): 2.
40. Hattey E, Wojta J, Binder BR. Monoclonal antibodies against plasminogen and alpha-2-
antiplasmin: binding to native and modified antigens. Thromb Res 1987; 45 : 485-95.
41. Carroll VA, Griffiths MR, Geiger Met al. Plasma protein C inhibitor is elevated in survivors
of myocardial infarction. Arterioscler Thromb Vasc Bio11997; 17: 114-18.
42. Pelzer H, Pilgrim A, Schwarz A, Merte D, Keuper H, Hock H. Determination of alpha2
antiplasmin complex in human plasma with an enzyme-linked immunosorbent assay.
Fibrinolysis 1993; 7: 69-74.

274
30
Soluble fibrin and degradation
products of fibrinogen (FgOP), fibrin
(FbOP; O-dimer) and total of FgOP
and FbOP (TOP)
W. NIEUWENHUIZEN and R. BOS

INTRODUCTION

Under normal physiological conditions the (low) activities of coagulation and

fibrinolysis are balanced. It has long been recognized that the assessment of
the enzymatically formed products of these two opposing processes, i.e. fibrin
and fibrin degradation products, may be of clinical and diagnostic value for
the detection of possible disturbances in the haemostatic balance.
Both the conversion of soluble fibrinogen to insoluble fibrin and fibrin
dissolution are multi-step processes that proceed via a series of intermediate
products. In this chapter some immunoassays, in particular enzyme immuno-
assays, will be described which can be used for the determination of soluble
fibrin defined as complexes between fibrin and fibrinogen molecules, and for
the determination of fibrin(ogen) degradation products in plasma samples.

SOLUBLE FIBRIN

When the coagulation system is activated active thrombin may be formed,

which then acts on fibrinogen by cleavage of the fibrinopeptides A, and
simultaneously (but more slowly) of fibrinopeptides B from the amino-termini
of the Au and Bf3 chains, respectively. This exposes new amino-terminal
sequences, which bind to sites which already pre-exist in fibrinogen. At an early
stage of fibrin formation complexes will be formed between fibrin molecules
275
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
and fibrinogen molecules. These complexes are soluble, and are designated as
soluble fibrin, which term is preferred over fibrin monomers. When thrombin
remains active more and more soluble fibrin will be formed, and eventually the
fibrin moieties will segregate from the complexes, and form a fibrin gel in which
the fibrin subunits are crosslinked by factor XIIIa. Increased levels of soluble
fibrin have long been considered as indicators of an impending thrombotic
event.

FIBRINOGEN DEGRADATION PRODUCTS (FgDP)

Under extreme disease conditions, but especially during thrombolytic therapy

with streptokinase, the plasmin activity, generated in the circulation, may
temporarily be incompletely inhibited by the natural plasmin inhibitors. Plasmin,
the most important fibrinolytic enzyme, is in itself not fibrin-specific. It may
also degrade other plasma proteins, including coagulation factors, e.g.
fibrinogen.
The first product which is formed when plasmin digestion of fibrinogen
(fibrinogenolysis) occurs, is designated as fragment X. Fragments X are slowly
clottable and can be described as fibrinogen molecules from which about 60%
of the length of both A a-chains has been removed. From a single fragment
X, one fragment Y and one fragment D are subsequently formed. Fragment
Y is composed of one D- and one E-domain. Fragment Y will eventually be
cleaved into one E- and one D-fragment. Thus one fibrinogen molecule yields
two D-fragments and one E-fragment.

FIBRIN DEGRADATION PRODUCTS (FbDP)

Plasmin digestion of the different forms of fibrin (fibrinolysis) proceeds via

intermediate products, analogous to those described above for fibrinogen, since
plasmin follows the same interdomainal cleavage pattern in fibrin as in fibrinogen.
Non-crosslinked desAA fibrin (fibrin I) will subsequently yield fragments
XI> VI' D, E I and desAABB fibrin (fibrin II) will yield XII' Y II , D, En. The
suffixes I and II denote the absence of FpA and of both FpA and FpB,
respectively.
Crosslinked fibrins I and II consist of very long polymers of fibrin I and II,
in which the subunits are covalently linked by isopeptide bonds. Plasmin will
attack the fibrin subunits in these polymeric structures in a random order. This
will result in smaller, soluble fragments of the original polymers, with a range
of molecular weights 1,2. These are collectively designated as X-oligomers 2-4.

D-dlmer
Eventually, these X-oligomers can be digested to fragments D-dimer, i.e. two
covalently bound D-domains 1,2, and fragments E. It should be emphasized
that so-called D-dimer assays detect not only D-dimer, but also all fibrin
276
 SOLUBLE FIBRIN AND FIBRINOGEN
degradation products comprising one or more D-dimer motifs (e.g. X-oligomers).
D-dimer as such rarely occurs in patient plasmas.

PRINCIPLES AND CHARACTERISTICS

Serum samples should not be used
Even in disease states the blood concentrations of fibrin( ogen) derivatives will
be low, i.e. in the ng to ",g/ml range, as compared with the fibrinogen con-
centration which is in the mg/ml range. For that reason, sensitive immunological
methods such as enzyme immunoassays (EIA) or latex agglutination assays
are required. Until relatively recently, only serum could be used as a sample,
since the available polyclonal antibodies crossreact virtually completely with
fibrinogen. Therefore, the latter had to be removed, e.g. by serum preparation.
Well-known serum assays are the Thrombo-Wellcotest assay and the tanned
red-cell haemagglutination inhibition assays. Serum, however, is a notorious
source of artifactual results, for the following reasons:
1. Incomplete clotting of the crossreacting fibrinogen may occur in cases of
dysfibrinogenaemia; when a patient has been exposed to he~arin6 or when
anticoagulating fibrinogen degradation products are presene- . In polyclonal
antibody-based serum assays this will inevitably lead to spuriously high or
false-positive results. These can also result from partial lysis of the clot
during serum preparation, not only in hyperfibrinolytic patients, but even
in normal individuals lO•
2. Some deftradation products will coagulate8,1l or become adsorbed to the
cloeO,12, 3. For that reason they will not be recovered in the serum, and
false-negative or spuriously low results may be obtained.
3. During serum preparation, fibrinogen degradation products will lose their
FpA. As a result they can no longer be discriminated from non-crosslinked
fibrin degradation products, and primary fibrinogenolysis cannot be detected.
4. Serum can obviously not be used for the assessment of the products of
ongoing coagulation in a patient, i.e. is not suitable for soluble fibrin assess-
ment.

Type of assay
With the advent of monoclonal antibody (mAb) technology it became possible
to develop assays that can be performed with plasma samples, thus avoiding
the serum artifacts mentioned above.
This chapter includes new assays based on mAb which determine in plasma
samples s~ecifically soluble fibrin21, degradation products of fibrinogen
(FgDP)I4-' 6, degradation products of fibrin I4,IS,17 (FbDP including D-dimer)
and the total of FgDP plus FbDP, designated as TDp 10,14,IS.
For soluble fibrin and D-dimer an increasing number of test kits has become
commercially available, including qualitative and semi-quantitative latex slide
tests, and quantitative assays (EIA and latex immunoassays, LIA). In tables
277
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
30.3 and 30.4 commercially available kits of soluble fibrin and D-dimer are
listed which have been frequently referred to in the literature up to now.
Note: The Chromogenix assay 'Coatest Soluble Fibrin' is based on the ability
of fibrin to potentiate the t-PA - catalysed conversion of plasminogen to
plasmin, which is analysed using a chromogenic substrate (S-2403). Other
assays detect antigen levels.
The details of how to perform the tests mentioned above can be found in
the kits inserts. As an example we describe below in more detail the quantitative,
sandwich-type enzyme immuno-assays (EIA) manufactured by Organon Teknika
and sold under the names Fibrinostika Soluble Fibrin, Fibrinostika FgDP,
Fibrinostika FbDP and Fibrinostika TDP (Table 30.1).

Fibrinostika Soluble Fibrin (SF)

The capture antibody in this EIA is a monoclonal antibody directed against a

fibrin-specific epitope, i.e. y 312-32423 • The tagging antibody is directed against
an epitope in the carboxy-terminal region of the fibrin(ogen) (A)a. chains2o •
On the basis of the combined wecificities of the two antibodies, the assay is
specific for intact soluble fibrin .
In a stud~ of 81 healthy individuals values ranging from 13 to 105 ng/ml
were found 1. It is recommended that each laboratory should determine its
own reference range based on the population tested.

Flbrlnostika FgDP

The capture antibody in this EIA is a mAb (FDP-14) specific for degraded
forms of both fibrinogen and fibrin, i.e. it does not react with intact fibrin and

Table 30.1 EIA for soluble fibrin and degradation products of fibrinogen and fibrin, and the
total of the latter two, manufactured by Organon Teknika as Fibrinostika Soluble Fibrin, FgDP,
FbDP and TDP, respectively

EIA specific for Specificity of

Capture antibody Tagging antibody

Soluble fibrin Fibrin-specific epitopes, Carboxyl-terminal area of

e.g. y 312-314 Au-chains (G8)
Fibrinogen degradation Altered conformation Amino terminus of Au-chains
products· (FgDP) in fibrin( ogen) degradation including FpA (mAb Y 18)
products (mAb FDP-14)
Fibrin degradation products Altered conformation D-domain of both crosslinked
(FbDP)* in fibrin( ogen) degradation and non-crosslinked FbDP
products (mAb FDP-14) (mAbDD 13)
Total of degradation products Altered conformation Mixture of two mAbs (mAb Y
(TDP)· of fibrin (FbDP)plus in fibrin(ogen) degradation 18 and mAb DD 13), see
those of fibrinogen (FgDP) products (mAb FDP-14) above

*Duration of assay only 45 min; uses dried mAb precoated plates.

278
 SOLUBLE FIBRIN AND FIBRINOGEN

fibrinogen 14 • When the test sample is incubated in the well of a microtitre plate
coated with FDP-14 both FgDP and FbDP will be bound. The tagging antibody
(YI8) in this EIA is specific for fibrinopeptide A-containing material 15; it does
not react with free fibrinopeptide A.
On the basis of the specificities of FDP-14 and Yl8 the assa~ is specific for
degradation products comprising fibrinopeptide A, i.e. FgDp 1 •
Analysis of 42 healthy individuals in an outside laboratory showed an upper
=
limit for FgDP of 250 ng FE/ml (FE fibrinogen equivalent unit). This is an
indication only, and each laboratory should determine its own reference range.

Fibrinostika FbOP

The capture of mAb of this EIA is FDP-14, i.e. the same as in Fibrinostika
FgDP. However, the tagging mAb is specific for FbDP. It is important to note
that it does not discriminate between crosslinked (including D-dimer) and
non-crosslinked FbDP. In that respect it is not fully comparable with the existing
EIA for D-dimer. The combination of the specificities of the capture and
tagging antibodies in Fibrinostika FbDP make this EIA specific for FbDp 17•
Analysis of 42 healthy individuals in an outside laboratory showed an upper
limit for FbDP of 310 ng FE/mI. This is an indication only, and each laboratory
should determine its own reference range.

Fibrinostika TOP

The capture antibody, in this EIA also, is FDP-14. Thus, in the first step of
this EIA both FgDP and FbDP are bound. The tagging antibody consists of
a mixture of mAbs used in Fibrinostika FgDP and Fibrinostika FbDP. This
means that in Fibrinostika TDP both FgDP and FbDP (= TDP) are detected.
The EIA has been published in a version in which polyc1onal tagging antibodies
were used (10).
The reproducibilities are given in Table 30.2. Analysis of 42 healthy individuals
in an outside laboratory showed an upper limit for TDP of 650 ng FE/mt. This
is an indication only, and each laboratory should determine its own reference
range.

PATHOPHYSIOLOGY
Soluble Fibrin

Relatively little is known about the relationship between SF levels as marker

of hypercoagulation and various diseases 19- 22,24. It has been shown that soluble
Fibrin levels are significantly increased in patients with pulmonary embolism,
deep venous thrombosis and disseminated intravascular coagulation DIC 21 •
The SF assay of Chromogenix (Table 30.3) was found to ~redict organ system
failure and outcome in patients in the intensive care unit 5,26.
279
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Table 30.2 Reproducibility of Fibrinostika SF, FgDP, FbDP and TDP (Organon Teknika)

Sample· CV (%) within-run CV (%) between-run

SF FgDP FbDP TDP SF FgDP FbDP TDP

1 5.2 15.7 4.2 11.0 9.6 19.3 10.8 27.2

2 5.4 7.0 5.0 9.3 4.3 12.5 14.8 30.8
3 8.1 6.7 8.8 14.0 18.3 4.2
4 8.3 3.7 1.9 8.9 21.5 15.5

• Clinical samples, three runs in triplicate

Concentration samples 1 and 2 used in Fibrinostika Soluble Fibrin 440 and 1205 nglml; Fibrin-
ostika FgDP 6785, 3176, 1341 and 257 ng fibrinogen equivalent unit (FE) mI in samples 1, 2, 3
and 4; Fibrinostika FbDP 8132, 5128, 1961 and 713 ng FElmi in samples 1, 2, 3 and 4;
Fibrinostika TDP 9978,5273, 1842 and 256 ng FElmI, in samples 1,2,3 and 4, respectively

D-dlmer
D-dimer assays from different sources have been compared with respect to
efficiency, and clinical characteristics27- 33 • A wide range of sensitivities and
specificities has been reported because of inter- and intra-assay variations and
differences in patient populations32 • Latex agglutination assays seem to be
relatively insensitive as screening tests28•32 • ELISA methods have previously
been proven to be reliable27 in the exclusion of thrombosis as occurring in
pulmonary embolism. The latest rapid methods have also been claimed to

Table 30.3 Commercially available assays for soluble fibrin

Supplier Product Method Minimal Relative Limit oj

assay specificity detection
time (Ilg/ml)
(min)

Chromogenix Coatest Soluble Quantitative; 15 + 0.23

Fibrin chromogenic assay
(t-PA stimulation)
Diagnostica-Stago; FSTest Semi-quantitative 16 +/- 15-20
Boehringer erythrocyte
Mannheim agglutination
Dade-Behring Berichrom FM Quantitative; 6 + 1.0
chromogenic assay
(t-PA stimulation)
Organon Teknika Fibrinostika Quantitative; 180 ++ 0.01
Soluble Fibirin ELISA
Boehringer Enzymun-Test Quantitative; 120 + 0.12
Mannheim FM ELISA
American Thrombus Quantitative; 95 ++ 0.13
Biogenetic Sciences Precursor ELISA
Protein-TpP

280
 SOLUBLE FIBRIN AND FIBRINOGEN
Table 30.4 Commercially available assays for D-dimer

Supplier Product Manual Automatic

LX fF EfA Other MLX EfA Other

Agena,b Dimertest Latex x

SimpliRED D-dimer x
Dimertest Gold x
Auto Dimertest x

BioMerieux FDPSlidex x
Vidas D-dimer x
Biopool TintElize D-dimer x
Minute D-dimer x
Nephelotex D-dimer x
Boehringer Tinaquant D-Dimer" x
Mannheim
Dade-Behring Enzygnost D-dimer x
BCD-dimer x
Turbiquant D-dimerd x
Data-Fi Dimertest x
Diagnostica- D-di Test x
Stago
Asserachrom D-dimer x
STA Liatest D-di x
Instant I.A. x
Hemoliance Ortho Dimertest x
Iatron LPIA ACE D-dimer x
IATRO DDIE Test x
IL IL D-dimer Test x
Nycomed Nycocard D-dimer x
Organon Fibrinosticon x
Teknika
Fibrinostika FbDP* x
Sigma Accuclot D-dimer x

• Products from Agen are distributed by a variety of companies including American Diagnostica,
Ortho, Dade-Behring, Fujirebio and Chromogenix.
b SimpliRED from Agen is based on agglutination of the patients' own red cells.
C Tinaquant is used by Boehringer Mannheim for application on the Hitachi analysers; in addition

Boehringer Mannheim is the distributor of Stago's Liatest, manual latex and ELISA.
d Turbiquant is an automatic turbidimetric test for Behring's Turbitimer system.
= = = =
Abbreviations: LX late; IF immune filtration; EIA enzyme immunoassay; MLX microlatex.
• Assay also measures non-crosslinked FbDP

present a good sensitivity to thrombosis 33 . A serious problem remains: the

results of different D-dimer assays vary widely depending on the assay used34•
It makes comparison of, for example, cut-off values for diagnosis difficult.
There are at least two reasons for the discrepancies. The D-dimer assays are
based on different mAb which may detect the various D-dimer comprising
281
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
fragments to a different extent; moreover the assays have different calibra-
tors34. The next step required is to prepare an international reference for use
by manufacturers and others34 . Finally the value and the utility of each
individual D-dimer assay should be established in clinical material.

FgDP and FbDP

Clinical experience with the new assays is rapidly growing, and fibrin(ogen)
degradation products have been measured in plasma samples of patients with
a variety of diseases, such as DIC, deep venous thrombosis (DVT) and
pulmonary embolism (PE), in pregnancy and (pre)eclampsia, in coronary artery
disease and thrombolytic therapy, in liver disease, in liver transplantation and
in malignancies 18 .
Fibrinolysis appears to be associated with (low grades of) fibrinogenolysis.
The process of secondary fibrinolysis, as derived from FbDP levels, appears
to predominate in deep venous thrombosis, pulmonary embolism, myocardial
infarction and unstable angina. Thrombolytic therapy, especially with strep-
tokinase, results in a pronounced fibrinogenolysis (primary fibrinolysis) as
assessed by the levels of FgDP. The occurrence of an extensive fibrinogenolysis
was also observed during orthotopic liver transplantation. Several diseases,
such as DIC, malignancies and liver disease, exhibit both fibrinogenolysis and
fibrinolysis secondary to an activated state of coagulation.
Plasma FbDP levels, as detected by EIA, reflect decreases in thrombus size
and may be used to monitor efficacy of heparin treatment. Low initial values
of FbDP are predictive of poor thrombus dissolution in patients with DVT
following streptokinase or urokinase therapy. Elevated plasma levels of cross-
linked FbDP have been measured during thrombolytic therapy of acute
myocardial infarction. They are derived mainly from extracoronary sites and
are non-predictive of recanalization following thrombolytic therapy. Plasma
fibrin degradation products levels are related to the severity of liver cirrhosis.
The observed increased fibrin(ogen) levels during normal and complicated
pregnancy suggest that fibrinolysis is not necessarily depressed. The fibrinolytic
system remains active in patients with malignancies, as demonstrated by elevated
levels of D-dimer and FbDP.
As mentioned in the introduction, a disturbance in the haemostatic balance
will presumably be reflected in the products of both coagulation and fibrinolysis.
Detection of fibrin( ogen) derivatives by the monoclonal antibody-based assays
provides valuable information on haemostasis in several diseases. Future clinical
studies are required to establish the clinical relevance of measuring simultane-
ously the products of the two opposing processes, coagulation and fibrinolysis,
i.e. soluble fibrin and FbDP.

References
1. Doolittle RF. Fibrinogen and fibrin. Sci Am 1981; 245: 92-101.
2. GraeffH, Hafter R. Detection and relevance of cross-linked fibrin derivatives in blood. Semin
Thromb Haemostas 1982; 8: 57-68.
282
 SOLUBLE FIBRIN AND FIBRINOGEN
3. Gaffney PJ, Perry MJ. 'Giant' fibrin fragments and thrombosis. Thromb Haemost 1985; 54:
931 (abstract).
4. Gaffney PJ, Creighton LC, Harris R, Perry MJ. Monoclonal antibodies (MABS) to crosslinked
fibrin fragments. Their characterization and potential clinical use. In: Miiller-Berghaus G. et
ai., eds. Fibrinogen and its derivatives: biochemistry, physiology and pathophysiology.
Amsterdam: Excerpta Medica, 1986; 273-84.
5. Merskey C, Lalezazi P, Johnson Al A rapid simple sensitive method for measuring fibrinolytic
split products in human serum. Proc Soc Exp BioI Med NY 1969; 131: 871-5.
6. Connaghan DG, Francis DW, Ryan DH, Marder VI Prevalence and clinical implications of
heparin-associated false positive tests for serum fibrin(ogen) degradation products. Am J Clin
Patho11986; 86: 304-10.
7. Haverkate F, Timan G, Nieuwenhuizen W Anticlotting properties of fragments D from human
fibrinogen and fibrin. Eur J Clin Invest 1979; 9: 253-5.
8. Nieuwenhuizen W, Gravesen M. Anticoagulant and calcium-binding properties of high
molecular weight derivatives of human fibrinogen, produced by plasmin (fragments X).
Biochim Biophys Acta 1981; 668: 81-8.
9. Nieuwenhuizen W, Voskuilen M, Hermans 1. Anticoagulant and calcium-binding properties
of high molecular weight derivatives of human fibrinogen (plasmin fragments Y). Biochim
Biophys Acta 1982; 708: 313-16.
10. Koopman J, Haverkate F, Koppert PW, Nieuwenhuizen W, Brommer EJP, van der Werf WGC.
New enzyme immunoassay of fibrin-fibrinogen degradation products in plasma using a
monoclonal antibody. J Lab Clin Med 1987; 109: 75-84.
11. Marder VJ, Shulman NR. High molecular weight derivatives of human fibrinogen produced
by plasmin II. Mechanism of their anticoagulant activity. J Bioi Chern 1969; 244: 2120--4.
12. Gaffney PJ, Perry MI Unreliability of current serum fibrin degradation products (FDP)
assays. Thromb Haemost 1985; 53: 301-2.
13. Niewiarowski S, Stewart GJ, Marder VI Formation of highly ordered polymers fibrinogen
and fibrin degradation products. Biochim Biophys Acta 1970; 221: 326--41.
14. Koppert PW, Koopman J, Haverkate F, Nieuwenhuizen W Production and characterization
of a monoclonal antibody reactive with a specific neoantigenic determinant (comprising B~
54-118) in degradation products of fibrin and of fibrinogen. Blood 1986; 68: 437-41.
15. Koppert PW, Huijsmans CMG, Nieuwenhuizen W A monoclonal antibody, specific for human
fibrinogen, fibrinopeptide A-containing fragments and not reacting with free fibrinopeptide
A. Blood 1985; 66: 503-7
16. Koppert PW, Kuipers W, Hoegee-de Nobel E, Brommer EJP, Koopman J, Nieuwenhuizen W
A quantitative enzyme immunoassay for primary fibrinogeno1ysis products in plasma. Thromb
Haemostas 1987; 57: 25-8.
17. Koppert Pw, Hoegee-de Nobel E, Nieuwenhuizen W A monoclonal antibody-based enzyme
immunoassay for fibrin degradation products in plasma. Thromb Haemostas 1988; 59: 310-15.
18. Kroneman H, Nieuwenhuizen W, Knot EAR. Monoclonal antibody-based plasma assays for
fibrin(ogen) and derivatives, and their clinical relevance. Blood Coagul Fibrinol1990; 1:
91-111.
19. Scheefers-Borchel U, Miiller-Berghaus G, Fuhge P, Eberle R, Heimburger N. Discrimination
between fibrin and fibrinogen by a monoclonal antibody against a synthetic peptide. Proc
Nat! Acad Sci USA 1985; 82: 7091-5.
20. Nieuwenhuizen W, Hoegee-de Nobel E, Laterveer R. A rapid monoclonal antibody-based
enzyme immunoassay (EIA) for the quantitative determination of Soluble Fibrin in plasma.
Thromb Haemostas 1992; 68; 273-7.
21. Bos R, Laterveer GH, Lockwood D, Szewczyk K, Nieuwenhuizen W A new enzyme
immunoassay for Soluble fibrin, with a high discriminating power for thrombotic disorders.
Thromb Haemostas 1998, in press.
22. Nieuwenhuizen W Soluble fibrin as a molecular marker for a pre-thrombotic state; a mini
review. Blood Coagul Fibrinol1993; 4: 93-6.
23. Schielen WJG, Adams HPHM, van Leuven K, Voskuilen M, Tesser GI, Nieuwenhuizen W
The sequence y-[312-324] is a fibrin-specific epitope. Blood 1991; 77: 2169-73.
24. Dempfle CE, Pfitzner SA, DoHman M, Huch K, Stehle G, Heene DL. Comparison of
immunological and functional assays for measurement of Soluble fibrin. Thromb Haemostas
1995; 74: 673-9.

283
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
25. Wiman B, RAnby M. Determination of Soluble fibrin in plasma by a rapid and quantitative
spectrophotometric assay. Thromb Haemostas 1986; 55: 189-93.
26. Bredbacka, S, Blombiick M, Wiman B. Soluble Fibrin - a predictor for the development and
outcome of multiple organ failure. Am J Hemato11994; 46: 289-94.
27. Bounameaux H, de Moerloose P, Perrier A, Reber G. Plasma measurement of D-dimer as
diagnostic aid in suspected venous thromboembolism - an overview. Thromb Haemost 1994;
71: 1--6.
28. Elias A, ApteI I, Huc B, Chale JJ, Nguyen, F, Cambus JP, Boccalon H, Boneu B. D-dimer
test and diagnosis of deep vein thrombosis: a comparative study of 7 assays. Thromb Haemostas
1996; 76: 518-22.
29. Veitl M, Hamwi A, Kurturan A, Virgolini I, Vukovich Th. Comparison of four rapid D-dimer
tests for diagnosis of pulmonary embolism. Thromb Res 1996; 82: 3~07.
30. Janssen MCH, Heebels AE, de Metz M, Verbruggen H, Wollersheim H, Janssen S, Schuurmans
MMJ, Novakova IRO. Reliability of five rapid D-dimer assays compared to ELISA in the
exclusion of deep venous thrombosis. Thromb Haemostas 1997; 72: 262--6.
31. Legnani C, Pancani C, Palareti G, Guazzaloca G, Fortunato G, Grauso F, Golfieri R,
Gianpalma E, Coccheri S. Comparison of new rapid methods for D-dimer measurement to
exclude deep vein thrombosis in symptomatic outpatients. Blood Coagul Fibrinol1997; 8:
296-302.
32. Lee AY, Ginsberg JS. The role of D-dimer in the diagnosis of venous thromboembolism. Curr
Opin Pulmon Med 1997; 3: 275-9.
33. Freyburger G, Trillaud H, Labrouche S, Gauthier P, Javorschi S, Bernard P, Grenier N. D-dimer
strategy in thrombosis exclusion. A gold standard study in 100 patients suspected of deep
venous thrombosis on pulmonary embolism: 8 DD methods compared. Thromb Haemo-
stasl998; 79: 32-7.
34. Nieuwenhuizen W. A reference material for harmonization of D-dimer assays; SSC
Communication. Thromb Haemostas 1997; 77: 1031-3.

284
31
Venous occlusion test in fibrinolysis
assays
J.JESPERSEN

INTRODUCTION

Since its introduction more than 30 years ago 1,2 the venous occlusion test (VO
test) has been widely used to assess systemic fibrinolytic capacity after venous
occlusion. In the 1980s the YO test was included in the EeAT procedures and
in the EeAT clinical follow-up studies in order to have a response after a
standardized stimulus (low basal systemic fibrinolytic activity).
Despite its apparent imperfection (see section on imprecision and stability
over time) with a relatively low correlation between repeated measurements of
the YO test, as reported in a subset of the EeAT prospective Angina Pectoris
Study 3, the plasma plasminogen activator tissue-type (t-PA) activity after the
YO test was decreased in patients at risk of a coronary event. Furthermore, all
the results of the EeAT Angina Pectoris Study of approximately 3000
individuals from suspected angina pectoris, and 106 definite coronary events
over a 2-year follow-up, pointed to a lowered fibrinolytic capacity as a risk
marker of a cardiac event, but its predictive power was smaller than that of
plasma protein concentration (baseline value)4.

Pathophysiology
In a number of observational studies a poor fibrinolytic response after venous
occlusion has frequently been observed in patients with recurrent venous
thrombosis5-8. At least five families have been identified with recurrent venous
thrombosis and an impaired fibrinolytic response to venous occlusion9-13. The
incidence of abnormal fibrinolysis in patients with venous thrombosis is about
30%6,8,14, and the YO test result also appears to have a predictive value for the
recurrence of venous thrombosis 15.
285
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
A reduced fibrinolytic response after a VO test characterizes patients with
ischaemic heart disease (lHD), as revealed from case-control studies l 6--l9.
Decreased fibrinolytic capacity after venous occlusion has been shown to have
a predictive value as a risk marker for reinfarction and cardiac death among
young survivors of acute myocardial infarction 20 . As in the ECAT Angina
Pectoris Study the latter study used venous occlusion for 10 min; this relatively
short occlusion time was chosen in order to achieve a high patient acceptance
rate, i.e. high compliance.

Assessment of impaired fibrinolytic response

Determination of the fibrinolytic capacity following a VO test was, until the
introduction of specific assays, performed by a global screening test, based for
instance on the determination of the fibrinolytic activity in plasma euglobulin
fractions. Results of such global assays were thought to reflect mainly the
systemic t-PA activity. However, significant contributions of both the factor
XII-dependent pathway and the prourokinase/urokinase have been observed,
probably through the generation of plasmin via the t-PA system and subsequent
activation of the intrinsic system2l - 26 .
Specific assays now make it possible to determine the plasma concentra-
tions of t-PA, measured by enzymatic and immunological procedures27- 3o (see
also Chapters 24 and 25) following a VO test. Various approaches have been
described in order to categorize patients as 'good' or 'poor' responders to the
VO tese ' , among other factors according to the presence or absence of residual
plasminogen activator inhibitor (PAI)-activity 32, the net t-PA capacity to
exhaust PAl activity, i.e. net increased t-PA protein concentration over pre VO
test PAl activity 33.

Imprecision and stability over time

So far the distinction between 'good' and 'poor' responders to the VO test and
the specific determination of components within the t-PAlPAI-1 system have
revealed wide inter-laboratory variation 30. An extensive investigation by Stegnar
and Pentek34 has shown that the response to venous occlusion in otherwise
healthy subjects is influenced by age, gender, body weight, blood lipids, and
insulin levels. These factors must be considered in the design of observational
studies, i.e. cohort study or cross-sectional study, or crossover study, or be
adjusted for when the VO test is applied.
Few studies have dealt with the reproducibility and stability over time of the
response to the VO test in high-risk patients and healthy individuals.
Within the group of patients with thromboembolism persistent low fibrinolysis
(global fibrinolysis assays) was found in the range of 69-82% in different studies,
and altogether in 82 patients5 ,l5,3l. In patients suffering from angina pectoris
the VO test (t-PA activity) performed twice, with a 2-year interval, the correlation
coefficient was 0.26, indicating, although statistically significant, relatively low
stability over time 3 •
286
 VENOUS OCCLUSION TEST
Marckmann et al. 35 have provided substantial knowledge on the variability
over time from a I-year monitoring study of 17 healthy young men with constant
lifestyles. Table 31.1 shows the total variability (repeated measurements), intra-
personal variation, and seasonal variation of the VO test and measurements
of plasma t-PA protein concentration. For comparison the estimates for baseline
values of t-PA and PAI-l protein plasma concentrations are also given. For all
three variables the total variability was substantial (range of CV (%)46-34)
and intra-personal variability (range of CV (%)35-22) (excluding intra-serial
analytical variation).
In a short study also in healthy volunteers, the intra-personal and intra-
analytical variations expressed as CV (%) for plasma PAI-l protein and t-PA
protein were 33% and 14%, res~ectively36. These figures are comparable with
the study of Marckmann et al. 5, who found for PAI-l protein a CV of 35%,
and for t-PA protein, a CV of 22%. Intermediate values were found for the VO
test and the plasma t-PA protein concentration.
Thus, the hitherto available data revealed that the VO test response to change
in t-PA protein concentration is only moderately stable over time. The same,
however, holds for other established cardiovascular risk factors 3,35,37. Maybe
the relatively good stability of baseline t-PA protein may, for this reason, be
the best scoring representative of the t-PAlPAI-1 system, when only one blood
sample is available for determination. At least in the ECAT Angina Pectoris
Study this seems to be the case4 .
Descriptive statistics of t-PA and PAI-l are given in Table 31.2.

PRINCIPLE OF THE TEST

An impaired response following the VO test has been attributed to either an

increased baseline level of plasmin~en activator inhibitors or to a deficient
release of t-PA from the vessel wall ,32,38.
When external pressure is applied onto the arm by a blood cuff several
factors cause a rise in the fibrinolytic activity, e.g. accumulation, extravasation
of fluid, clearance, and metabolic changes. Although Keber 39 provided strong
evidence that, during venous occlusion, the accumulation of t-PA can be
explained merely by continuously produced t-PA of the endothelium and
change of clearance via the liver, an acute release of t-PA cannot be excluded40 •

EQUIPMENT
1. Blood pressure cuff, balloon and sphygmomanometer.
2. At least two blood test tubes placed in an ice-water bath.

PERFORMANCE OF THE VO TEST

1. The participant is asked to sit down in an easy chair or to lie down on a bed
for at least 15 min.
287
 ~
o

Table 31.1 Estimates of components of variation in plasma t-PA before and after venous occlusion (VO), and PAI-I of healthy young men with steady o~
lifestyles followed over I year (percentage of total variance shown in parentheses) (modified from ref. 35) ~
-I
Variation
m
o
:::c
Variable nind nabs Mean Total Inter-personal Intra-personal"- Seasonal z
oc
t-PA protein (ng/ml) 16 79 4.1 SD2 1.88 0.95 (51%) 0.93 (49%) m
I\) SD 1.37 0.98 0.96 en
<Xl
<Xl 15.92 (51%) 1.47 (5%) Z
t-PA protein + VO (ng/ml) 16 79 11.9 SD2 31.14 13.75 (44%)
SD 5.58 3.71 3.99 1.21 ~
JJ
log (PAI-I protein) (ng/ml) 16 79 0.73 b S~ 0.105 0.037 (35%) 0.068 (65%) o
SD 0.32 0.19 0.26 s:::
OJ
alnclude intraserial analytical variation - t-PA, CV 8%; PAl-I, CV 7%.
oen
bpAI_I protein 5.37 ng/ml. en
nind, number of individuals; nob., number of observations I
»
~
z
c
»r-
 VENOUS OCCLUSION TEST
Table 31.2 Descriptive statistics of plasma t-PA, and PAl-I, measured in healthy young adults
(n = 74);+ VO, after venous occlusion (n = 44) (modified from ref. 35)
Non-parametric Parametric

Variables Median (5-95 Mean (range of

percentile) central 90"/0)

t-PA protein (nglml)*(n = 74) 3.3 (1.4-6.9) 3.5 (0.8-6.2)

t-PA protein + VO (ngfml)(n = 44) 12.8 (S.0-24.2) 13.8 (-O.8-28.S)
PAI-l protein (nglml)(n = 74) 4.9 (1.3-IS.8) 6.5 (-2.9-1S.9)

*Medians: 3.S nglml (males) vs 2.6 nglml (females),p= 0.01 (Mann-Whitney U-test)

2. Blood is collected from one arm by venipuncture.

3. The blood pressure cuff is then wrapped around the other arm just above
the elbow.
4. The cuff is inflated to a pressure 10--15 mmHg above diastolic pressure, i.e.
the diastolic pressure is read from the manometer at the point where, during
gradual inflation, Korotkow sounds disappear.
5. By squeezing the balloon occasionally the pressure is maintained for 10 min.
6. Exactly 10 min after the inflation of the cuff to the desired level, venipuncture
is performed in the congested arm vein.
7. Just before withdrawing the needle the cuff should be deflated.
During the procedure the colour of the forearm skin turns purple, and lighter
patches may become visible. In the first few minutes some aching and tingling
maybe felt.

COMMENTS

Recently the VO test has been criticized because of an excessively high variability
of the venous occlusion itseW 1• This is not a problem only related to the VO
test, but also to the measurement of a large number of haemostatic factors
without venous occlusion. Both analytical imprecision and within-person (intra-
individual). variability can be substantial compared with variability amo~
persons3,3 ,42,43 (see also the section on imprecision and stability over time)37, •
Despite such a variability, even baseline PAI-l protein level and activity were
positively associated with, for example, coronary risk when adjustments were
made for centre, age, and sex4 • This indicates that the underlying relationship
to thrombotic risk must be greater than anticipated, since even a single
measurement was found to be an important risk marker, despite only moderate
reproducibility. This point also seems to occur with the VO test, but underlines
the need for improved measurement procedure, harmonization, and
standardization of haemostatic assays, including fibrinolytic assays and the
VO test. Thus, until a consensus is established on how to perform the VO test,
type of assays, degree of deviation from established on reference range, and
whether correction for haemoconcentration is needed or not, its main application
is in well-defined clinical studies.
289
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
This laboratory manual can be an important platform in the needed
standardization process. In the meantime each laboratory must describe any
deviation from the suggested standardized procedure, and must establish its
own reference material. Examples of such procedures are available (e.g. refs.
6, 8, 14, 32,40, 45,and 46).

References
1. Nilsson 1M, Robertson B. Effect of venous occlusion on coagulation and fibrinolytic
components in normal subjects. Thromb Diathes Haemorrh 1968; 20: 397--408.
2. Robertson BR, Pandolfi M, Nilsson 1M. 'Fibrinolytic capacity' in healthy volunteers as
estimated from effect of venous occlusion of arms. Acta Chir Scand 1972; 138: 429-36.
3. Pyke SDM, Thompson SG, Buchwalsky R, Kienast J, on behalf of the ECAT Angina Pectoris
Study Group. Variability over time of haemostatic and other cardiovascular risk factors in
patients suffering from angina pectoris. Thromb Haemostas 1993; 70: 743-6.
4. Thompson SG, Kienast J, Pyke SDM, Haverkate F, van de Loo JCW, for the European
Concerted Action on Thrombosis and Disabilities Angina Pectoris Study Group. Hemostatic
factors and the risk of myocardial infarction or sudden death in patients with angina pectoris.
N EnglJ Med 1995; 332: 635--41.
5. Isacson S, Nilsson 1M. Defective fibrinolysis in blood and vein walls in recurrent 'idiopathic'
venous thrombosis. Acta Chir Scand 1972; 138: 313-19.
6. Nilsson 1M, Ljungner H, Tengborn L. Two different mechanisms in patients with venous
thrombosis and defective fibrinolysis: low concentration of plasminogen activator or increased
concentration of plasminogen activator inhibitor. Br Med J 1985; 290: 1453-6.
7. Stalder M, Hauert J, Kruithof EKO, Bachmann F. Release of vascular plasminogen activator
(v-PA) after venous stasis: electrophoretic-zymographic analysis of free and complexed v-PA.
Br J Haemato11985; 61: 169-76.
8. Juhan-Vague I, Valadier J, Alessi MC et al. Deficient t-PA release and elevated PA inhibitor
levels in patients with spontaneous or recurrent deep venous thrombosis. Thromb Haemostas
1987; 57: 62-72.
9. Johansson L, Hedner U, Nilsson 1M. A family with thromboembolic disease associated with
deficient fibrinolytic activity in vessel wall. Acta Med Scand 1978; 203: 477-80.
10. Alexandre P, Larcan A, Briquel ME. Recurring thrombo-embolic accidents caused by family-
related deficiency of the fibrinolysis system. Blut 1980; 41: 437--44.
II. Jorgensen M, Mortensen JZ, Madsen AG, Thorsen S, Jacobsen B. A family with reduced
plasminogen activator activity in blood associated with recurrent venous thrombosis. Scand
J Haeinatol1982: 29: 217-23.
12. Stead Nw, Bauer KA, Kinney TR et al. Venous thrombosis in a family with defective release
of vascular plasminogen activator and elevated plasma factor VIII/von Willebrand's factor.
Am J Med 1983; 74: 33-9.
13. Boyko OB, Pizzo SY. Mesenteric vein thrombosis and vascular plasminogen activator. Arch
Pathol Lab Med 1983; 107: 541-2.
14. Alessi MC, Juhan-Vague I, Valadier J, Philip-Joet C, Holvoet P, Collen D. Relevance of free
tPA assay following venous occlusion in patients with venous thromboembolic disease. Thromb
Haemostas 1988; 59: 346-7.
15. Korninger C, Lechner K, Niessner H, GOssinger H, Kundi M. Impaired fibrinolytic capacity
predisposes for recurrence of venous thrombosis. Thromb Haemost 1984; 52: 127-30.
16. Walker ID, Davidson JF, Hutton I, Lawrie TOY. Disordered 'fibrinolytic potential' in coronary
heart disease. Thromb Res 1977; 10: 509-20.
17. Gritsyuk AI, Schogelsky VI. Evaluation of blood coagulation and prethrombotic state in
patients with coronary atherosclerosis by application of controlled local venous blockade.
Circulation 1979; 60: 220A.
18. Rawles JM, Warlow C, Ogston D. Fibrinolytic capacity of arm and leg veins after femoral
shaft fracture and acute myocardial infarction. Br Med J 1975; 2: 61-2.
19. Hamsten A, Blombiick M, Wiman B. Svensson J, Szamosi A, de Faire U, Mettinger L.
Haemostatic function in myocardial infarction. Br Heart J 1986; 55: 58-66.

290
 VENOUS OCCLUSION TEST
20. Hamsten A, de Faire U, Walldius G et al. Plasminogen activator inhibitor in plasma: risk
factor for recurrent myocardial infarction. Lancet 1987; 2: 3-9.
21. Jespersen 1. The diurnal increase in euglobulin fibrinolytic activity in women using oral
contraceptives and in normal women, and the generation of intrinsic fibrinolytic activity.
Thromb Haemostas 1986; 56: 183-8.
22. Rijken DC, Wijngaards G, Welbergen J. Relationship between tissue plasminogen activator
and the activators in blood and vascular wall. Thromb Res 1980; 18: 815-30.
23. Kluft C. Wijngaards G, Jie AFH. The factor XII-independent plasminogen proactivator
system of plasma includes urokinase-related activity. Thromb Haemostas 1981; 46: 343.
24. Kluft C, Jie AFH. Interaction between the extrinsic and intrinsic system of fibrinolysis. In:
Davidson JF et al. eds. Progress in clinical fibrinolysis and thrombolysis. Edinburgh: Churchill
Livingstone, 1979: 25-31.
25. Kluft C, Dooijewaard G, Emeis JJ. Role of the contact system in fibrinolysis. Semin Thromb
Hemostas 1987; 13: 50-68.
26. Jespersen 1. Pathophysiology and clinical aspects of fibrinolysis and inhibition of coagulation.
Experimental and clinical studies with special reference to women on oral contraceptives and
selected groups of thrombosis prone patients. Thesis. Dan Med Bull 1988; 35: 1-33.
27. Rijken DC, Juhan-Vague I, De Cock F, Collen D. Measurement of human tissue-type
plasminogen activator by two-site immuno radiometric assay. J Lab Clin Med 1983; 101:
274-84.
28. Verheijen JH, Mullaart E, Chang GTG, Kluft C, Wijngaards G. A simple, sensitive
spectrophotometric assay for extrinsic (tissue-type) plasminogen activator applicable to
measurements in plasma. Thromb Haemostas 1982; 48: 266-9.
29. Chmielewska IN, Wiman B. Determination of tissue plasminogen activator and its 'fast'
inhibitor in plasma. Clin Chem 1986; 32: 482-5.
30. Jespersen 1. How to detect defects of tissue plasminogen activator in thrombosis-prone patients.
An introduction to discussion. Fibrinolysis 1988; 2 (Suppl. 2): 104-11.
31. Tengborn L. Laboratory investigation of fibrinolytic and other haemostatic risk factors in
patients with venous thromboembolism. In: Nilsson TK, Boman K, Jansson JH, eds. Clinical
aspects of fibrinolysis. Stockholm: Almqvist & Wiksell International, 1991; 67-84.
32. Nguyen G, Horellou MH, Kruithof EKO, Conard J, Samama MM. Residual plasminogen
activator inhibitor activity after venous stasis as a criterion for hypofibrinolysis: a study in
83 patients with confirmed deep vein thrombosis. Blood 1988; 72: 601-5.
33. SyrjiUii MT, Krusius T, Petiijii J, Vahtera E, Rasi V. Venous occlusion test: Assessment of
fibrinolytic capacity by D-dimer latex agglutination test. Fibrinolysis 1993; 7: 41-5.
34. Stegnar M, Pentek M. Fibrinolytic response to venous occlusion in healthy subjects:
Relationship to age, gender, body weight, blood lipids and insulin. Thromb Res 1993; 69:
81-92.
35. Marckmann P, Sandstrom B, Jespersen 1. The variability of and associations between measures
of blood coagulation, fibrinolysis and blood lipids. Atherosclerosis 1992; 96: 235-44.
36. Kluft C. Constitutive synthesis of tissue-type plasminogen activator (t-PA) and plasminogen
activator inhibitor type 1 (PAl -1): conditions and therapeutic targets. Fibrinolysis 1994; 8
(Suppl. 2): 1-7.
37. Thompson SG, Martin JC, Meade TW Sources of variability in coagulation factor assays.
Thromb Haemostas 1987; 58: 1073-7.
38. Jennings I, Luddington RJ, Harper PL. Changes in endothelial-related coagulation proteins
in response to venous occlusion. Thromb Haemostas 1991; 65: 374-6.
39. Keber D. Mechanism of tissue plasminogen activator release during venous occlusion.
Fibrinolysis 1988; 2 (Suppl. 2): 96-103.
40. Petiijii 1. Fibrinolytic response to venous occlusion for 10 and 20 minutes in healthy subjects
and in patients with deep vein thrombosis. Thromb Res 1989; 56: 251-63.
41. Stegnar M, Mavri A. Reproducibility of fibrinolytic response to venous occlusion in healthy
subjects. Thromb Haemostas 1995; 73: 453-7.
42. Gram J, Declerck PJ, Sidelmann J, Jespersen J, Kluft C. Multicentre evaluation of commercial
kit methods: plasminogen activator inhibitor activity. Thromb Haemostas 1993; 70: 852-7.
43. Declerck PJ, Moreau H, Jespersen J, Gram J, Kluft C. Multicenter evaluation of commercially
available methods for the immunological determination of plasminogen activator inhibitor-l
(PAl-I). Thromb Haemostas 1993; 70: 858-63.

291
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
44. Jansson J-H, Norberg B, Nilsson TK. Impact of acute phase on concentrations of tissue
plasminogen activator and plasminogen activator inhibitor in plasma after deep-vein thrombosis
or open-heart surgery. Clin Chern 1989; 35: 1544-5.
45. Nicoloso G, Hauert J, Kruithof EKO, Van Melle G, Bachmann F. Fibrinolysis in normal
subjects - comparison between plasminogen activator inhibitor and other components of the
fibrinolytic system. Thromb Haemostas 1988; 59: 299-303.
46. Sultan Y, Harris A, Strauch G, Venot A, De Lauture D. A dynamic test to investigate potential
tissue plasminogen activator activity. Comparison of deamino-8-D-argininevasopressin with
venous occlusion in normal subjects and patients. J Lab Clin Med 1988; 111: 645-53.

292
32
List of manufacturers

Materials and reagents for the procedures and the assays below correspond with
the order and the content of the assay chapters and are followed by a list of
addresses.

Chapter 3: Blood collection and sample preparation: pre-analytical variation

Diatube H (CTAD) Diagnostica Stago
Stabilyte tubes (citrate at reduced pH) Biopool
Vacutainer tubes (plain, citrate, EDTA, Becton Dickinson
CTAD)
Venoject tubes (plain, citrate, EDTA) Terumo

Chapter 7: Endogenous thrombin potential

Actin FS Dade-Behring
Arvin Knoll
Malonyl-a-aminoisobutyryl-arginine Serbio
paranitroanilide methyl ester hydrochloride
PPACK Calbiochem
Relipidated recombinant human tissue factor Dade-Behring
(Innovin)
Reptilase Boehringer~annheim
Substrate S2165
Substrate S2238 Chromogenix

Chapter 8: Fibrinogen
Standard 89/644 NIBSC

Chapter 9: Activated factor vn

Bovine brain phosphatidylserine
Bovine liver phosphatidylethanolamine
Bovine serum albumin, fatty acid-free Calbiochem
Egg phosphatidylcholine Avanti Polar Lipids
293
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

Factor VII-deficient plasma George King

Bio-Medical
Plasma-derived factor VIla
Recombinant factor VIla American Diagnostica
Recombinant soluble tissue factor Diagnostica Stago
Capture antibody of factor VIla is not
commercially available
Commercially available kits
Staclot VIla-rTF Diagnostica Stago
Standard 89/688 NIBSC

Chapter 10: Factor vn activity and antigen

Commercially available factor VII antigen kits
Asserachrome VII:Ag Diagnostica Stago

Chapter 11: Factor VIn clotting activity

Automated APTT Organon Teknika
Factor VIII-depleted plasma Diagnostica Stago
Ortho
Commercially available kits Dade-Behring
Chromogenix
Diagnostica Stago
Standard 911666 NIBSC

Chapter 12: Von Willebrand factor

Horseradish peroxidase Sigma
o-phenylenediamine Sigma
Commercially available kits
Asserachrom vWF (ELISA) Diagnostica Stago
VonWF:Ag (ELISA) American Diagnostica
Standard 911666 NIBSC

Chapter 13: Antithrombin activity and antigen

Reagents for antithrombin activity
CBS 3447 (chromogenic substrate) Diagnostica Stago
Control plasmas (normal or pathological
range) Chromogenix
Diagnostica Stago
S-2238 (chromogenic substrate)
S-2765 (chromogenic substrate) Chromogenix
Reagents for antithrombin antigen
Antithrombin antiserum
Normal protein standard Dade-Behring
Standard 93/768 NIBSC

294
 LIST OF MANUFACTURERS

Chapter 14: Protein C activity and antigen

Commercially available protein C activity kits
Bioclot Protein C Biopool
Coamatic Protein C Chromogenix
MDA Protein C Organon Teknika
Spectrolyse Protein C Biopool
Stachrom Protein C
Staclot Protein C Diagnostica Stago
Reagents for protein C antigen
Conjugated antiprotein C immunoglobulins DAKO
Coomassie Brillant Blue R 250 Serva
Gelbond Film FMC Bioproducts
Laurell plates American Diagnostica
Diagnostica Stago
Tricine Sigma
Seakem LE Agarose FMC Bioproducts
Commercially available protein C antigen kits
Asserachrom Protein C Diagnostica Stago
Technoclone Protein C Technoclone
Thrombonostica Protein C Organon Teknika
Standard 93/590 NIBSC

Chapter 15: Protein S antigen

Antibodies American Diagnostica
DAKO
Enzyme Research
Laboratories
Bovine serum albumin
Ortho-phenylene-diamine (dihydochloride) Sigma
Commercially available kits
Asserachrom free protein S (ELISA)
Asserachrom total protein S (ELISA) Diagnostica Stago
Thromborostika protein S (ELISA) Organon Teknika
Assera-plate protein S (Laurell) Diagnostica Stago
Biopool protein S EID kit (Laurell) Biopool
Liatest protein S (nephelometric) Diagnostica Stago
Standard 93/590 NIBSC

Chapter 16: Protein S activity

Diluent Instrumentation
Laboratory

295
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
Commercially available kits
IL Test Protein S assay kit Instrumentation
Laboratory
Protein S Reagent Dade-Behring
Staclot protein S Diagnostica Stago
Standard 93/590 NIBSC

Chapter 17: Activated protein C (APC) resistance

APTT reagent Organon Teknika
Activated protein C Enzyme Research
Laboratories
Factor V-deficient plasma Diagnostica Stago
Commercially available kits
Coatest APC resistanee Chromogenix

Chapter 18: Tissue factor pathway inhibitor (TFPI)

Chromogenic substrate Chromogenix
Factor VIla (human) Diagnostica Stago
Factor VIla (human recombinant) Novo-Nordisk
Factor X Chromogenix
Factor Xa Chromogenix
Polybrene Sigma
Tissue factor (human recombinant) Dade-Behring
Commercially available kits
Imubind free TFPI American Diagnostica

Chapter 20: Heparin cofactor U

Capture antibody (product no. 2320) American Diagnostica
Instant non-fat dried milk Sigma
Substrate (Chromozyme TH) Boehringer Mannheim
TMB Peroxidase EIA Substrate Kit Biorad

Chapter 21: Fibrinopeptide A (FPA)

Anti-FPA antiserum Imco
Aprotinin (Trasylol) Bayer
FPA-standard Imco
Heparin (Liquemin) Roche
Immobilized second antibody (Sac-eel.
Wellcome Diagnostics) IDS
Monovette (blood collection system) Sarstedt
PPACK Calbiochem
Tyr-FPA Imco

296
 LIST OF MANUFACTURERS
Suppliers of reagents for the FPA
radioimmunoassay Bachem
Imco
Kordia
Wherl

Chapter 22: Thrombin-antithrombin (TAT) complexes

Commercially available kits
ELISA Enzygnost TAT Dade-Behring

Chapter 23: Prothrombin fragment FI +2

PPACK Calbiochem
Commercially available kits
ELISA Enzygnost Fl +2 micro Dade-Behring

Chapter 24: Tissue-type plasminogen activator (t-PA) activity

Chromogenic substrate (S-2403) Chromogenix
giu-plasminogen Chromogenix
Biopool
Monoclonal t-PA antibody (B-lO) Imco
Stabilyte tubes Biopool
Commercially available kits
Chromolize tPA Biopool
Immuno Free t-PA Corvas
Standard 86/670 NIBSC

Chapter 25: Tissue-type plasminogen activator antigen (t-PA Ag)

Commercially available kits
Asserachrom t-PA (ELISA) Diagnostica Stago
Coaliza t-PA (ELISA) Chromogenix
Imulyse t-PA (ELISA) Biopool
Innotest t-PA (ELISA) Innogenetics
TintElize t-PA (ELISA) Biopool
t-PA assay kit (ELISA) Cabru
Imco
Monozyme
Standard 86/670 NIBSC

Chapter 26: Plasminogen activator inhibitor-l (PAl-I) antigen

Commercially available kits
Asserachrom PAI-l Diagnostica Stago
Coaliza PAI-l Chromogenix
Imulyse PAI-l (ELISA) Biopool
PAI:Ag (ELISA) American Diagnostica
PAI-l ELISA kit Monozyme
297
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
PAI-l antigen Technoclone
TintElize PAI-l (ELISA) Biopool
Standard 92/654 NIBSC

Chapter 27: Plasminogen activity

Plasminogen free fibrinogen Imco
Streptase Dade-Behring
Streptokinase Chromogenix
Substrate Chromozym PL Pentapharm
Substrate S-225l Chromogenix
Standard (gIu-plasminogen) 78/646 NIBSC

Chapter 28: Plasmin inhibitor activity

Abnormal control plasma Chromogenix
Dade-Behring
Methylamine hydrochloride Sigma
Normal control plasma Dade-Behring
Chromogenix
Diagnostica Stago
Organon Teknika
Diagnostica Stago
Normal reference plasma Biopool
Chromogenix
Plasmin
Substrate S2403 Chromogenix
Tween 80 Merck
Commercially available kits
Actichrome <l2-antiplasmin American Diagnostica
<l2-antiplasmin Instrumentation
Laboratory
Coamatic Plasmin Inhibitor Chromogenix
Berichrome <lrantiplasmin Dade-Behring
Chromo strate Antiplasmin Organon Teknika
Orthokrome Antiplasmin Ortho
Spectrolyse <l2-antiplasmin Biopool
Stachrome Antiplasmin Diagnostica Stago
Unitest <l2-antiplasmin Unicorn

Chapter 29: Plasmin-arantiplasmin complexes

ABTS Boehringer Mannheim
Antibodies Technoclone
Aprotinin (Trasylol) Bayer
Benzamidiniumchloride Merck
Bovine serum albumin Dade-Behring
PAP complex depleted plasma Technoclone
PAP complex standard Technoclone

298
 LIST OF MANUFACTURERS

PPACK Calbiochem
Thimerosal Merck
Tween 20 Sigma
Commercially available kits
PAP kits (ELISA) Dade Behring
Technoc1one
Teijin

Chapter 30: Soluble fibrin and degradation products of fibrinogen

(Fg, fibrin (FbDP; D-dimer) and total of FgDP and FbDP (TDP)
Commercially available kits
Coaset Fibrin Monomer Chromogenix
D-dimer kits Agen Biomedical
American Diagnostica
BioMerieux
Diagnostica Stago
latron
Nycomed
Fibrinostica soluble fibrin
Fibrinostica FbDP
Fibrinostica FgDP
Fibrinostica TDP Organon Teknika
Soluble fibrin kit American Biogenetic
Sciences

ADDRESSES
Agen Biomedical Ltd American Diagnostica Inc.
11 Durbell Street 222 Railroad Avenue
PO Box 391 PO Box 1165
Acacia Ridge 4110 Greenwich CT 06836-1165
Brisbane USA
Australia Phone: + 1 2036610000
Phone: +61 733706300 Fax: + 1 203661 7784
Fax: + 61 733706370 E-mail: [email protected]
Internet: http://www.agen.com.au
E-mail: [email protected] Avanti Polar Lipids Inc.
700 Industrial Park Drive
Alabaster
AL 35007
American Biogenetic Sciences USA
1375 Akron Street Copiague Phone: + 1 80022706511 + 1205663
NY 11726 2494
USA Fax: + 1 800 229 10041 + 1 205 663
Phone: + 1 516 789 2600 0756
Fax: + 1 516789 1661 Internet: http://www.avantilipids.com
Internet: http://www.mabxa.com E-mail: [email protected]

299
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

BachemAG Fax: +15107415817

Hauptstrasse 144 Internet: http://www.biorad.com
CH-4416 Bubendorf
Switzerland
Phone: +4161 93123 33 Boehringer Mannheim GmbH
Fax: +41 61 931 2549 Sandhofer Strasse 116
Internet: http://www.bachem.com Postfach 310120
E-mail: [email protected] 0-68305 Mannheim
Germany
BayerAG Phone: +49 621 759 85 68
0-51368 Leverkusen Fax: +49 621 75940 83
Bayerwerk Internet: http://www.boehringer-
Germany mannheim.com
Phone: +49214301
Fax: +492143066411 Cabru
Internet: http://www.bayer.com Via Caduti per la Patria 47
20050 Lesmo (MI)
Becton Dickinson Italy
1 Becton Drive Phone: +39 39 698 1589
Franklin Lakes Fax: +39396065174
New Jersey 07417-1885
USA
Phone: + 1 201 8474500 Calbiochem
Fax: +12018474886 POBox 12087
Internet: http://www.bd.com San Diego
CA 92112-4180
BioMerieux USA
376 Chem de L'orme Phone: + 1 800 854 3417
69280 Marcy L'etoile Fax: + 1 800 776 0999
France Internet: http://www.ca1biochem.com
Phone: +33 478 87 2000 E-mail: [email protected]

Biopool International Chromogenix AB

6025 Nicole Street Taljegardsgatan 3
Ventura S-431 53 Molndal
CA93003 Sweden
USA Phone: +46 31 706 2000
Phone: + 1 805 654 0643 Fax: +46 31 864626
Fax: + 1 805 654 0681 Internet: http://www.chromogenix.se
Internet: http://www.biopool.se E-mail: [email protected]
E-mail: [email protected]

Bio-Rad Laboratories CorvasInc.

1000 Alfred Nobel Drive 3030 Science Park Rd.
Hercules La Jolla
California 94547 CA92037
USA USA
Phone: + 1 510 724 7000 Phone: + 1 6194559800
300
 LIST OF MANUFACTURERS
Dade-Behring USA
POBox 1149 Phone: + 1 913 469 5464
D-35001 Marburg 1 Fax: + 1 913 469 0871
Germany Internet: http://kingbiomed.com
Phone: +49 642 39 8880 E-mail: [email protected]
Fax: +49 642 131388
Internet: http:// Iatron Laboratories Inc.
www.dadebehring.com 1-1, Higashi-Kanda 2-Chrome,
Chiyoda-ku
DAKOAIS Tokyo 101
Produktionsvej 42 Japan
DK-2600 Glostrup Phone: 03 38621761
Denmark Fax: 03 38621760
Phone: +45 44 85 95 ()() Internet: http://www.sosei.coml
Fax: +45 44 85 95 95 companies/iatron.html
Internet: http://www.dako.dk
Immuno Diagnostic Systems Ltd
Diagnostica Stago (IDS)
9, Rue des Freres Chausson Boldon Business Park
F-92600 Asnieres sur Seine Boldon
France Tyne&Wear
Phone: +33146882020 NE359PD
Fax: +33 1479 10891 United Kingdom
Internet: http://www.stago.fr Phone: +44 191 5190660
E-mail: stago@stagoJr Fax: +44 191 5190760

Enzyme Research Laboratories !moo

300 N. Michigan Street, Suite 103 Hudiksvallsgatan 4B
South Bend S-113 30 Stockholm
IN 46601 Sweden
USA Phone: +46 833 53 09
Phone: + 1 2192882268 Fax: +46872 84 776
Fax: + 12192882272
Innogenetics NY
FMC BioProducts Industriepark 7
191 Thomaston Street Box 4
Rockland B-9052 Zwijnaarde
Maine 04841 Belgium
USA Phone: +3292410711
Phone: + 1 207 584 3400 Fax: +329241 799
Fax: + 1 207 594 3426 Internet: http://www.innogenetics.be
Internet: http://www.bioproducts.com
E-mail: [email protected] Instrumentation Laboratory SpA
Viale Monza 338
George King Bio-Medical Inc. 20128 Milan
11771 West 112th St. Italy
Overland Park Phone: +39 225 221
KS 66210-4192 Fax: +3922575250
301
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL

KnollAG Novo Nordisk A/S

Knollstrasse NovoAlle
67061 Ludwigshafen DK-2880 Bagsvaerd
Postfach 210805 Denmark
67008 Ludwigshafen Phone: +45 4444 8888
Germany Fax: +45 4449 0555
Phone: +49 621 5890 Internet: http://www.novo.dk
Fax: +496215892896
Internet: http://www.knoll.de Nycomed Pharma A/S
E-mail: [email protected] Sandakerveien 78
PO Box 4220 Torshov
N-04010s10
Kordia bv Norway
Oude Singel158 Phone: +4723185050
NL-2312 RG Leiden Fax: +4723186000
The Netherlands Internet: http://www.nycomed-
Phone: +31 71 523 1050 amersham.com
Fax: +31 71 5221700
Internet: http://www.kordia.n1 Organon Teknika
E-mail: [email protected] Boseind 15
5281 RM Boxtel
MerckKGaA The Netherlands
D-64271 Darmstadt Phone: +31 411 6549111
Germany Fax: +31 411 654201
Phone: +496151 72 0 Internet: http://www.akzonobel.coml
Fax: +496151 72 2000 home.htm
Internet: http://wingate.merck.de Ortho Diagnostic Systems Inc.
E-mail: [email protected] 1001 US Highway 202
Raritan 08869-1487 N.J.
Monozyme USA
Agemalle3 Phone: + 1 908218 1300
DK-2970 Hersholm
Denmark Pentapharm Ltd
Phone: +4545572014 Enge1gasse 109
Fax:+45 45 76 57 08 POBox
CH-4002 Basel
Switzerland
National Institute for Biological Phone: +41 61 706 4848
Standards and Control (NIBSC) Fax: +41 61 3199619
Blanche Lane Internet: http://www.pentapharm.ch
South Mimms
Potters Bar Roche Pharmaceuticals
Hertfordshire EN6 3QG F. Hoffmann-La Roche Ltd
United Kingdom CH-4070 Basel
Phone: +44 1707 654 753 Switzerland
Fax: +44 1707 646 730 Phone: +41 61 688 11 11
Internet: http://www.nibsc.ac.uk Fax: +41 61 691 93 91
E-mail: [email protected] Internet: http://www.roche.com
302
 LIST OF MANUFACTURERS

Sarstedt Technoc1one GmbH

Postfach 12 20 Millinergasse 23
D-51582 Niimbrecht Postfach 158
Germany A-1092 Vienna
Phone: +49 2293 3050 Austria
Fax: +492293 305 122 Phone: +43 1 3422 80
Internet: http://www.sarstedt.com Fax: +43 1 3195278

Serbio Teijin Ltd

PAE Parispace 3 Uchisaiwai-cho 2-1-1
125 Avenue Louis Roche Chiyoda-ku
92230 Gennevilliers Tokyo 100
France Japan
Phone: +33141471500 Phone: +81 335064892
Fax: +33 147987337
Fax: +81 335082767
Serva
Terumo Europe NY
Boehringer Inge1heim Bioproducts
Interleuvenlaan 40
Partnership
3001 Leuven
Branch Germany/Headquarters
Belgium
Czernyring 22/11
D-69115 Heidelberg Phone: +32 16381211
Germany Fax: +32 16229104
Phone: +49 6221 59 8300
Fax: +496221 598313 Unicorn Diagnostics Ltd.
Internet: http://www.bi- London
bioproducts.de United Kingdom
E-mail: [email protected] TeVFax: +44 181 559 1006

Sigma Chemical Company WherlGmbH

PO Box 14508 Halchtersche Strasse 49
St. Louis D-38304 Wo1fenbuettel
MO 63178 Germany
USA Phone: +49 533195610
Phone: +13147715765 Fax: +49 5331956113
Fax: +13147715757 Internet: http://www.wherl.com
Internet: http://www.sigma.sial.com E-mail: [email protected]
E-mail: [email protected]

303
Index

accuracy 4 biliary obstruction 46

activated partial thromboplastin time [AP1TJ bleeding
APe resistance 163 ANT 38
ANT 37 factor VII activity and antigen 99
blood collection 24 good medical laboratory services 9
lupus anticoagulant 183 introduction 3
protein C activity and antigen 129 PAI-1 antigen 239
quality assessment 32, 35 plasmin-a-antiplasmin complexes 266
activated protein C (APC) quality assessment 31
protein C activity and antigen 129 t-PA antigen 232
protein S activity 153 blood collection 21
protein S antigen 141 blood pressure 227
activated protein C (APC) resistance blood sampling 21
APe resistance 163
prothrombin fragment Fl +2 218 C4b binding protein (C4bBP) 142, 153
quality assessment 36 cancer 159
acute pancreatitis 191 cardiovascular disease
amyloidosis 260 factor VII activity and antigen 99
angina pectoris fibrinogen 80,81
FPA 199 introduction 3
protein C activity and antigen 131 cathepsin G 189, 196
prothrombin fragment F1 +2 218 central nervous system thrombosis 130
t-PA activity 227 central ophthalmic thrombosis 130
venous occlusion test 285, 286 cerebral infarction 199
annexin V 183 cerebral vein thrombosis 124
a.-antiproteinase 130 cerebrovascular diseases 164
anticardiolipin antibodies 183, 184 chemotherapy 213
antiphospholipid antibodies 159, 183 cholesterol 177
a2-antiplasmin 257 chronic renal failure 191
antithrombin chymotrypsin 189,196
antithrombin activity and antigen 121 cold activation 95
quality assessement 32, 36 coronary artery disease
TAT complexes 209 endogenous thrombin potential 77
arterial ischaemic events 231 fibrinogen 80
arterial thrombosis 124, 142 plasmin-a-antiplasmin complexes 272
atherosclerosis 118 soluble fibrin and degradation products of
autoimmune disease 183, 184 fibrinogen 282

305
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
TAT complexes 212 factor VIII inhibitors 185
CTAJ) 25,26,234 factor XII 3
familial thrombophilia 130
D-dimer 36, 275 fibrin monomer 276
DDAVP 232 fibrin(ogen) degradation products 36,276
deep vein thrombosis fibrinogen
antithrombin activity and antigen 124 fibrinogen 79
APTT 40 introduction 1
endogenous thrombin potential 77 quality assessment 31, 36
FPA 199 soluble fibrin and degradation products of
protein S antigen 143 fibrinogen 276
prothrombin fragment Fl +2 218 fibrinopeptide A (FPA) 199,217
soluble fibrin and degradation products of
fibrinogen 279,282 good laboratory practice 3, 9
TAT complexes 211,213 good medical laboratory services (GMLC) 9
t-PA antigen 232
diabetes mellitus haematocrit 22
FPA 200 haemophilia A
heparin cofactor II 191 activated factor VII 91
protein C activity and antigen 131 factor VIII clotting activity 112
protein S antigen 143 von Willebrand factor 118
prothrombin fragment Fl +2 218 HDL cholesterol 227
t-PA activity 227 heart disease 90
von Willebrand factor 118 heparin cofactor
disseminated intravascular coagulation (DIC) antithrombin activity and antigen 121
antithrombin activity and antigen 126 APC resistance 165
APTT 40 APTT 37,38,39,43
factor VIII clotting activity 112 blood collection 23
fibrinogen 80, 85 endogenous thrombin potential 67
FPA 199 fibrinogen 85
heparin cofactor II 191 heparin cofactor II 189
plasmin inhibitor activity 260 lupus anticoagulant 186
plasmin---a-antiplasmin complexes 265, PT 47
266 quality assessment 36
protein C activity and antigen 131 soluble fibrin and degradation products of
prothrombin fragment Fl+2 218 fibrinogen 277,282
TAT complexes 213 TFPI 172
TFPI 173 hepaticfailure 191
DNA analysis 27 hirudin 190
dysfibrinogenaenlia 80,277 histidine-rich glycoprotein 247
hormone replacement 159
eclampsia 282 human immunodeficiency 191
elastase---al-antiproteinase complex 209 human leuserpin-2 189
elastase 196, 260 hypercholesteroiaemia 173, 213
EN 45001 10 hyperiipaemia 85
EN 46001 14 hyperlipoproteinaemia 260
EN 46002 14 hypertension 118, 131
evacuated tubes 24, 42, 226
infections 184
factor V Leiden 25,130,213 inflammatory diseases 112, 118
factor VII 26, 31 INR 24, 35, 36, 133
factor VIII insulin 286
APC resistance 163,164 insulin resistance 227
APTT 43 ischaenlic heart disease 90
blood col1ection 22, 23, 25, 26 international sensitivity index (lSI) 54, 59
factor VIII clotting activity 107 ISO 25 11,12
quality assessment 32, 33, 36 ISO 9001 14
von Willebrand factor 117 IS09002 14

306
 INDEX
kallikrein 231 APTT 39
endogenous thrombin potential 77
L-asparaginase 126,131,260 FPA 200
leukaemia 213, 260 heparin cofactor II 191
lifestyle variables 22 protein C activity and antigen 131
lipids 286 protein S antigen 143
~-lipoproteinaemia 173 prothrombin fragment FI +2 218
liver cirrhosis prothrombin time 47
antithrombin activity and antigen 125
FPA 200 plasmin 276,286
soluble fibrin and degradation products of plasmin inhibitor 231,247,257
fibrinogen 282 plasmin~-antiplasmin complexes 265
t-PA activity 227 plasminogen 2,25,36,247
von Willebrand factor 118 plasminogen activator (t-PA)
liver disease blood collection 25
APTI 40 introduction 3
protein C activity and antigen 131 PAl-l antigen 239
protein S activity 159 plasmin inhibitor activity 260
protein S antigen 143 plasmin-a-antiplasmin complexes 265
prothrombin time 47,60 t-PA activity 223
soluble fibrin and degradation products of t-PA antigen 231
fibrinogen 282 venous occlusion test 285, 288, 289
TAT complexes 213 plasminogen activator inhibitor-I (PAl-I)
liver function 46 blood collection 25
liver transplantation 282 PAl-I antigen 239
lupus anticoagulant t-PA activity 223
APC resistance 165 venous occlusion test 286, 288, 289
APTT 37,38 platelet factor 3 26
quality assessment 36 platelet factor 4 26
polybrene 47
Ilrmacroglobulin 65, 130, 260 post-infectious 159
malignancy 213,282 post-surgical patients
menopausal status 124 antithrombin activity and antigen 126,
mesenteric vein thrombosis 124 131
myocardial infarction factor VIII clotting activity 112
FPA 199 plasmin inhibitor activity 260
plasmin-a-antiplasmin complexes 266, TAT complexes 213
272 t-PA activity 227
prothrombin fragment Fl +2 218 von Willebrand factor 118
soluble fibrin and degradation products of pre-analytical factors 4
fibrinogen 282 pre-analytical variation 21
TAT complexes 213 precision 4, 6
t-PA activity 227 pro-urokinase system 3
venous occlusion test 286 proficiency testing 32
protein C 32, 36, 129
nephropathy 227 protein S 36, 141, 153
nephrotic syndrome prothrombin fragment Fl +2 209,217
antithrombin activity and antigen 126 prothrombin time (PT)
heparin cofactor II 191 APTT 37
plasmin inhibitor activity 260 factor VII activity and antigen 100, 103
protein C activity and antigen 131 fibrinogen 83
protein S activity 159 good medical laboratory services 24
protein S antigen 143 introduction 2, 4
lupus anticoagulant 183
oral anticoagulants prothrombin time 45
antithrombin activity and antigen 124, quality assessment 32, 35
126 TFPI 172
APC resistance 164 PTCA 272

307
 LABORATORY TECHNIQUES IN THROMBOSIS - A MANUAL
pulmonary embolism AP1T 37
antithrombin activity and antigen 124 blood collection 21
soluble fibrin and degradation products of introduction 2, 4
fibrinogen 279,282 protein S activity 157
TAT complexes 211,213 prothrombin time 45, 56
pulmonary thrombosis 199 quality assessment 36
purpura fulminans 130, 142 thrombosis
antithrombin activity and antigen 125
quality assessment 1, 10, 14, 18,29,30,32, APC resistance 167
39 soluble fibrin and degradation products of
quality management 11,12,14,17 fibrinogen 280
TFPI 173
renal failure 191,260 t-PA antigen 232
respiratory distress syndrome 218 Thrombotest 51
ristocetin cofactor 115 thrombotic diseases 90,239,276
tissue factor pathway inhibitor (TFPI) 65,
sample preparation 21 99,171
sepsis 212 total quality management (TQM) 10
Shewhart plots 10 triglycerides 227
sickle-cell disease 131
soluble fibrin 275 urokinase-type PA 239,260
spontaneous abortions 184 unstable angina
Stabilyte 25, 226 TAT complexes 213
streptokinase TFPI 182
plasmin-u-antiplasmin complexes 266 uraemia
plasminogen activity 247 plasmin-u-antiplasmin complexes 266,
soluble fibrin and degradation products of 270
fibrinogen 276,282 plasminogen activity 247
stroke 183, 266 soluble fibrin and degradation products of
surgery fibrinogen 282
antithrombin activity and antigen 124 venous occlusion test 286
APTf 40 von Willebrand factor 118
heparin cofactor II 191
prothrombin fragment Fl +2 211 venous occlusion test 285
prothrombin time 47 venous thromboembolism 164
systemic lupus erythematosus (SLE) 183 venous thrombosis
antithrombin activity and antigen 121
thrombin-antithrombin (TAT) lupus anticoagulant 183
complex 209, 217 protein S antigen 142
thrombin 1 venous occlusion test 285
thrombocytopenic purpura 118 von Willebrand factor 130
thromboembolic disease 183 vitronectin 239
thromboembolism 286 von Willebrand disease 118,232
thrombomodulin 65, 129 von Willebrand factor 22, 36, 115
thrombophilia 3
thromboplastin warfarin 47,91
activated factor VII 89 Westgard rules 10

308