Functional Imaging in Nephro Urology - 1st Edition

Visit the link below to download the full version of this book:

https://medipdf.com/product/functional-imaging-in-nephro-urology-1st-edition/

Click Download Now

Functional Imaging in
Nephro-Urology

Edited by

Alain Prigent MD
Professor of Physiology and Biophysics
Faculté de Médecine Paris-Sud
Le Kremlin-Bicêtre
France

Chief of the Department of Biophysics and Nuclear Medicine

CHU Bicêtre
Le Kremlin-Bicêtre
France

Amy Piepsz MD
Professor of Pediatrics and Nuclear Medicine
Free University of Brussels
CHU St Pierre
Department of Radioisotopes
B-1000 Brussels
Belgium
© 2006 Taylor & Francis, an imprint of the Taylor & Francis Group
Taylor & Francis Group is the Academic Division of Informa plc

First published in the United Kingdom in 2006 by Taylor & Francis, an imprint of the Taylor & Francis Group,
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

Tel: +44 (0)20 7017 6000

Fax: +44 (0)20 7017 6699
E-mail: [email protected]
Website: http://www.tandf.co.uk/medicine

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted,
in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior
permission of the publisher or in accordance with the provisions of the Copyright, Designs and Patents Act
1988 or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency,
90 Tottenham Court Road, London W1P 0LP.

Although every effort has been made to ensure that all owners of copyright material have been acknowledged
in this publication, we would be glad to acknowledge in subsequent reprints or editions any omissions brought
to our attention.

Although every effort has been made to ensure that drug doses and other information are presented
accurately in this publication, the ultimate responsibility rests with the prescribing physician. Neither the
publishers nor the authors can be held responsible for errors or for any consequences arising from the use of
information contained herein. For detailed prescribing information or instructions on the use of any product or
procedure discussed herein, please consult the prescribing information or instructional material issued by the
manufacturer.

A CIP record for this book is available from the British Library.

Library of Congress Cataloging-in-Publication Data

Data available on application

ISBN 1 84184 511 6

ISBN 978 1 84184 511 1

Distributed in North and South America by

Taylor & Francis
2000 NW Corporate Blvd
Boca Raton, FL 33431, USA

Within Continental USA

Tel: 800 272 7737; Fax: 800 374 3401
Outside Continental USA
Tel: 561 994 0555; Fax: 561 361 6018
E-mail: [email protected]

Distributed in the rest of the world by

Thomson Publishing Services
Cheriton House
North Way
Andover, Hampshire SP10 5BE, UK
Tel: +44 (0)1264 332424
E-mail: [email protected]

Composition by Scribe Design Ltd, Ashford, Kent, UK

Printed and bound in Great Britain by CPI Bath
Contents

List of contributors vii 11 The radiological approach 91

Preface ix Julie Ferzli, Jean-Nicolas Dacher

12 The nuclear medicine techniques 103

Measurement of renal function in health and
Jørgen Frøkiær, Anni Eskild-Jensen,
disease
Thomas Dissing
1 Introduction 3
13 The point of view of the urologist 117
Alain Prigent
Stephen A Koff
2 Assessment of GFR: chemical techniques
and prediction equations 11 Renovascular disease
Joseph V Nally
14 Present and future developments of
3 Measurements of renal function with radionuclide functional and molecular imaging 127
techniques in adults 19 M Donald Blaufox
Emmanuel Durand
15 Radiological modalities in renovascular disease 133
4 Measurement of GFR in children by means of Nicolas Grenier
radionuclide techniques 31
Amy Piepsz, Hamphrey Ham 16 ACE inhibition renography in the evaluation of
suspected renovascular hypertension 149
5 Assessment of GFR in adults: the point of Andrew Taylor
view of the nephrologist 37
Laurent Juillard, Maurice Laville
Functional and molecular imaging
6 Assessment of GFR in children: the point of
17 Functional imaging of the kidney with
view of the paediatric nephrologist 41
electron beam CT 165
Jean-Pierre Guignard
Lilach O Lerman

Obstruction in adults 18 Functional and cellular imaging of the kidney

by MRI 173
7 Introduction and the role of nuclear medicine 51 Nicolas Grenier, Michael Pedersen, Olivier Hauger
Patrick H O’Reilly
19 Positron emission tomography imaging of
8 The contribution of radiology 57 the kidneys 197
Michel Claudon, Damien Mandry, Zsolt Szabo, William B Mathews
Jean-Nicolas Dacher
Urinary tract infection in children
9 The urologist’s view 67
Stephen CW Brown
20 Introduction 213
Isky Gordon
Antenatally detected hydronephrosis
21 The point of view of the radiologist 217
10. Introduction 83 Jean-Nicolas Dacher
Amy Piepsz
vi Contents

22 The point of view of the paediatric nuclear

medicine physician 223
Monica A Rossleigh

23 The point of view of the paediatric nephrologist 233

Sverker Hansson

Index 241
Contributors

M Donald Blaufox Julie Ferzli

Albert Einstein College of Medicine Départment Central d”Imagerie Médicale
Montefiore Medical Park CHU de Rouen-Hôpital Charles Nicolle
NY 10461 76031 Rouen
USA France

Stephen CW Brown Jørgen Frøkiær

Department of Urology Department of Clinical Physiology and Nuclear Medicine
Stepping Hill Hospital Aarhus University Hospital – Skejby
Stockport DK-8230 Aarhus N
Cheshire Denmark
UK
Isky Gordon
Michel Claudon Great Ormond Street Hospital for Children
Service de Radiologie London
CHU Nancy Brabois-Hôpital d’Enfants UK
54511 Vandoeurve Cedex
France Nicolas Grenier
Service de Radiologie
Jean-Nicolas Dacher Groupe Hospitalier Pellegrin
Departement Central d’Imagerie Médicale 33076 Bordeaux Cedex
CHU de Rouen-Hôpital Charles Nicolle France
76031 Rouen
France Jean-Pierre Guignard
Division of Pediatric Nephrology
Thomas Dissing Centre Hospitalier Universitaire Vaudois
Institute of Clinical Medicine 1011 Lausanne
Aarhus University Hospital – Skejby Switzerland
DK-8200 Aarhus N
Denmark Hamphrey Ham
Department of Nuclear Medicine
Emmanuel Durand Academic Hospital
CHU de Bicêtre University of Ghent
Faculté de Médecine Paris-Sud 9000 Ghent
Department of Biophysics and Nuclear Medicine Belgium
94275 Le Kremlin-Bicêtre Cedex
France Sverker Hansson
The Queen Silvia Children’s Hospital
Anni Eskild-Jensen Pediatric Uronephrologic Centre (PUNC)
Department of Clinical Physiology and Nuclear Medicine Göteborg University
Aarhus University Hospital – Skejby Göteborg
DK-8230 Aarhus N Sweden
Denmark
viii List of contributors

Olivier Hauger Joseph V Nally

ERT CNRS ‘Imagerie Moléculaire et Fonctionnaell Department of Nephrology and Hypertension
Université Victor Ségalen-Bordeaux 2 Cleveland Clinic Foundation
Groupe Hospitalier Pellegrin Cleveland
33076 Bordeaux Cedex Ohio 44195
France USA

Laurent Juillard Amy Piepsz

Department of Nephrology and Hypertension CHU St Pierre
Hôpital Edouart-Herriot Department of Radioisotopes
Lyon Cedex 03 B-1000 Brussels
France Belgium

Stephen A Koff Alain Prigent

Pediatric Urology Faculté de Médecine Paris-Sud
Children’s Hospital Department of Biophysics and Nuclear Medicine
Education Building 94275 Le Kremlin-Bicêtre Cedex
Ohio 43205 France
USA
Patrick H O’Reilly
Maurice Laville Department of Urology
Department of Nephrology and Hypertension Stepping Hill Hospital
Hôpital Edouard Herriot Stockport
Université Claude Bernard Lyon 1 UK
Lyon Cedex 03
France Michael Pedersen
Magnetic Resonance Centre
Lilach O Lerman Institute of Clinical Medicine
Division of Nephrology and Hypertension Aarhus University Hospital – Skejby
Mayo Clinic College of Medicine DK-8200 Aarhus N
Rochester Denmark
Minnesota 55905
USA Monica A Rossleigh
University of New South Wales
Damien Mandry Department of Nuclear Medicine
Service de Radiologie The Prince of Wales and Sydney Children’s Hospitals
CHU Nancy Brabois-Hôpital d’Enfants Randwick NSW 2031
54511 Vandoeuvre Cedex Australia
France
Zsolt Szabo
William B Mathews Johns Hopkins Outpatient Center
Division of Nuclear Medicine Baltimore
Johns Hopkins University MD 21287
Nelson Building USA
Baltimore
MD 21287 Andrew Taylor
USA Division of Nuclear Medicine
Emory University School of Medicine
Atlanta
GA 30322
USA
Preface

In 1967, a first international symposium on radionuclides in long and well-structured discussions with the audience. The
nephro-urology was held in Liège (Belgium). The purpose of most up-to-date developments of the traditional methods
this symposium was to bring together a group of people with were presented by the different speakers, while new
a common interest in the application of radionuclides in techniques, such as functional and molecular imaging with
nephro-urology. Internists, radiologists, urologists, physiolo- MR, CT and PET appeared as promising approaches. What
gists and others representing the basic and clinical sciences came out of these multidisciplinary sessions is remarkably
were brought together for intensive discussions. Since that similar for all topics, namely a critical appraisal of the tradi-
time, similar meetings were organized by the International tional strategies of management and a series of potential new
Scientific Committee of Radionuclides in Nephrourology directions which might, in the near future, significantly change
(ISCORN) in New York (1971), Berlin (1974), Boston the clinical management of the patient.
(1978), London (1981), Lausanne (1986), Williamsburg It appeared therefore that the moment was well chosen
(1989), Chester (1992), Santa Fé (1995), Copenhagen to reassemble this huge amount of information within a book
(1998), Monterey (2001) and La Baule (2004). The final under the general title of ‘The Role of Functional Imaging in
purpose of these meetings was the application of the radionu- Nephro-urology’. The chapters correspond to the five clinical
clide techniques in clinical fields such as hypertension, renal sessions and for each topic, the contributors provided a
transplantation, hydronephrosis and infection. At the same detailed and referenced overview of their expertise,
time, a huge amount of methodological studies had given rise completed by a rich iconography.
to new developments in nuclear medicine. It has been the This book, by its multidisciplinary approach, is a ‘première’
role of ISCORN to chair consensus conferences, which and will provide outstanding information to radiologists
resulted in a better standardization of radionuclide methods working on child and adult, to nuclear medicine physicians, to
in fields such as measurement of renal clearance, evaluation internists and paediatricians, to nephrologists and urologists
of renal transit and drainage, application of captopril renogra- specialized in child and adult.
phy to renovascular disease, cortical scintigraphy in urinary We want to express our sincere thanks for help to the
tract infection in children and management of renal trans- other members of the ISCORN committee: Donald Blaufox
plants. (New York, USA), Keith Britton (London, UK), Eva Dubovsky
The meeting, held in May 2004 in La Baule (France), was (Birmingham, USA), Belkis Erbas (Ankara, Turkey), Jörgen
a kind of achievement. It was the feeling of the Committee Frökiaer (Aarhus, Denmark), Joseph V. Nally (Cleveland,
that the time had come to bring together the different special- USA), Patrick O’Reilly (Stockport, UK), Pilar Orellana
ities involved in the strategy of uro-nephrological diseases and (Santiago, Chile), Monica Rossleigh (Sydney, Australia),
to evaluate the potential place of various techniques in the Michael Rutland (Auckland, New Zealand) and Andrew
management of patients. The basic structure of the sympo- Taylor (Atlanta, USA). Thanks also to the experts of all special-
sium was therefore centred on a series of clinical topics, all of ities who contributed by their outstanding presentations to the
them characterized by a significant number of controversial success of this multidisciplinary event. Their lectures were the
matters: determination of renal function in child and in adult, starting materials for the different chapters of this book. All of
antenatally detected hydronephrosis, renal obstruction in this would not have been possible without the help of the
adults, renovascular hypertension, renal infection in child- organizing committee (Joseph Lecloirec, MD and Mrs Maïté
hood. Radiologists, nuclear medicine physicians, physiolo- Lepelletier) who did a great job in making this conference one
gists, paediatric and adult nephrologists, paediatric and adult of the most exciting meetings ISCORN has ever experienced.
urologists, all eminent experts in their respective fields, devel- Finally sincere thanks to Tyco France and Tyco USA whose
oped the state of the art and constituted then a large panel for role in sponsoring the meeting has been essential.

Alain Prigent, Paris, France

Amy Piepsz, Brussels, Belgium
Editors
Measurement of renal function in
health and disease
1 Introduction
Alain Prigent

Operational definition of renal humans). However, as FF changes occur in certain clinical

circumstances (e.g., proteinuric glomerulopathy, ischaemia,
function postischaemic injury after transplantation, renovascular
hypertension, acute urinary obstruction, …), RPF changes do
The level of the glomerular filtration rate (GFR) is generally not always parallel GFR changes. Moreover, EFoa varies
accepted as the best overall index for the complex functions dramatically and unpredictably in numerous conditions,
of the kidney in health and disease.1 This agreement holds on especially in chronic renal diseases. As an example, the
to functional, pathological, clinical and prognostic arguments. extraction fraction of PAH, considered as the gold standard
The functional coupling between GFR and tubular function molecule for ERPF measurement, is 0.92 ± 0.03
especially relies upon the 'positive' glomerulotubular balance (mean ± SE) in normal volunteers4 but may decrease to an
and the 'negative' tubuloglomerular feed-back, which ensure average of 0.80 in benign essential hypertension,5 0.75 in
an integrative regulation of the whole nephron function. patients treated with cyclosporin,4 0.70 in proteinuric
Similarly, the GFR decrease correlates with the extent of glomerulopathies,6 or to 0.20 in ischaemic acute renal failure
tubulointerstitial fibrosis and/or tubular atrophy in chronic and to 0.10 in the recovery period.7 Moreover, in all these
renal diseases.2 GFR being reduced prior to the onset of cases the standard deviation of the mean EFoa is about 0.10 to
symptoms of renal failure, its assessment enables earlier 0.15, indicating a wide range of variation between individual
diagnosis and therapeutic interventions in patients at risk. data. In renovascular disease, where the renal function is
Thus the level of GFR is a strong predictor of the time of asymmetrical, EFoa of PAH is about 0.55 and 0.75 in the
onset of kidney failure as well as the risk of complications stenotic and contralateral kidney, respectively, and decreases
of chronic kidney disease.1 Many techniques, using either further to 0.35 and 0.65, respectively, after administration of
chemical or radiopharmaceuticals, exist providing either captopril.8 Even with the most sophisticated curve-fitting
estimates or true measurements of the global GFR. procedures most methods are too imprecise for accurate
In case of asymmetrical renal disease the determination of prediction of EFoa in a given individual.4
the individual renal function requires a global GFR measure-
ment to be combined with the assessment of the split renal
function (e.g., expressed in percentages of the global
function) by a noninvasive imaging modality. Although renal
scintigraphy is presently the most often used because of its
Renal function in health
widespread availability, low cost (compared to the alternative
modalities of computed tomography and magnetic resonance Notwithstanding the great variability of GFR even in healthy
imaging), and absence of side effects from the tracers,3 new individuals due to many physiological factors (e.g., body size,
applications of computed tomography (e.g., multidetector gender, age, salt and dietary protein intakes, diurnal varia-
CT, electron beam computerized tomography) and magnetic tions), normal ranges of GFR have been reported. This
resonance imaging appear promising (see Chapters 15 and variability can be reduced by taking into account body surface
17). area ('normalization' to 1.73 m2). When using the 'classical'
The clearances of some tubularly secreted organic anions, gold standard of inulin clearance,9 the mean values of GFR in
such as p-aminohippuric acid (PAH), 131I- or 125I-ortho- young adults are 127 ml/min/1.73 m2 in men and
iodohippurate (OIH), or even 99mTc-mercaptoacetyltriglycine 118 ml/min/1.73 m2 in females with a standard deviation of
(MAG 3), are referred to as the effective renal plasma flow approximately 20 ml/min/1.73 m2, while when using 51Cr
(ERPF). GFR is related to ERPF by the expression: EDTA (ethylenediaminetetraacetic acid) plasma clearance,10
the normal mean (±SD) GFR is 105 (±25) ml/min/1.73 m2
GFR = ERPF.FF/EFoa (no gender difference after correction for BSA). With regards
to transversal studies,10–13 GFR linearly decreases by approxi-
where EFoa is the extraction fraction of the used organic anion mately 1.0 ml/min/1.73 m2 per year with large interindividual
(EFoa = ERPF/RPF, RPF being renal plasma flow), and FF the variation even among 'healthy' individuals. Indeed, regarding
filtration fraction (FF = GFR/RPF, about 0.20 in normal longitudinal studies,14 one-third of the healthy elderly subjects
4 Functional Imaging of Nephro-Urology

has no absolute change, another third has a progressive but definition and classification of stages of chronic kidney disease
small decline, and in the last third of elderly GFR declines to (guideline 1) and another to its evaluation by estimation of
50–70% of the maximum GFR value. In one-month aged GFR (guideline 4).
neonates,15,16 the mean GFR is about half the adult value GFR plays a cornerstone role in the definition of chronic
(55 ml/min/1.73 m2) and increases progressively until 18 kidney disease (CKD), since CKD is defined on two criteria,
months–2 years. Between 2 and 17 years of age, expressed one of which being a decreased GFR:
as ml/min/1.73 m2, the GFR remains constant, with a mean
value of 114 ml/min/1.73 m2 (SD: 24 ml/min/ 1.73 m2), 1. Kidney damage for 3 months at least, as defined by
similar to the value in adults. structural or functional abnormalities of the kidney, with
In normal individuals, the reactive increase in GFR or without decreased GFR, manifest by either patholog-
(120–140% of the baseline value) within the 2 hours follow- ical abnormalities, or markers of kidney damage (includ-
ing an oral protein load (e.g., 300–500 g of cooked beef) is ing abnormalities in the composition of the blood or
defined as 'functional renal reserve'.17 Subsequently, similar urine, or abnormal imaging tests);
increases in GFR were reported18 with either gluconeogenic 2. GFR lower than 60 ml/min/1.73 m2 for 3 months at
amino acids (50–75 g within 3 hours) or dopamine least, with or without kidney damage (as defined in
(1.5–2 µg/kg/min for 2 hours) infusion. Although it was initially criteria 1).
thought that this 'reserve' was lost in the presence of early
renal impairment (i.e., not diagnosed by plasma creatinine Similarly, the GFR level is used for the stage definition of
test), these findings were not confirmed in many later series. CKD:
Expressed as a percentage of baseline GFR value, the
'functional reserve' does not decrease in kidney diseases (see • Stage 1: with normal or increased GFR:
reference 19). Apart from this diurnal variation due to meal GFR ≥ 90 ml/min/1.73 m2
intake, there is a circadian rhythm of GFR20 with a maximum • Stage 2: GFR between 60 and 89 ml/min/1.73 m2 (mild)
around 1 pm, a minimum around 1 am, and a relative ampli- • Stage 3: GFR between 30 and 59 ml/min/1.73 m2
tude ([max – min]/mean) of about 30% for inulin clearance (moderate)
and 20% for creatinine clearance. The nutritional status also • Stage 4: GFR between 15 and 29 ml/min/1.73 m2
affects GFR,21 especially dietary intakes of proteins, calories (severe)
(whatever the nutriments), and sodium (an important deter- • Stage 5: GFR <15 ml/min/1.73 m2 (renal failure).
minant of extracellular fluid volume, ECFV). For an example,
GFR increases to about 140% of its baseline value during For the 'estimation' (not the measurement) of GFR, the
pregnancy in relation to an increase in ECFV. recommendation is to use prediction equations taking into
At the borderline between health and disease, the account serum creatinine concentration and some of the
compensatory hyperfunction of the remnant kidney in variables, which determine the creatinine production, such as
donors restricts, partly the functional lost. Thus GFR (125I- age, gender, body size, ethnicity, etc. Estimating GFR by
iothalamate urinary clearance measurements) is about 60% prediction equation based on serum creatinine is more
(69 ± 4 ml/min/1.73 m2) and 70% (78 ± 5 ml/min/ reliable (i.e., more accurate and more precise) than measur-
1.73 m2) of the predonation value (111 ± 6 ml/min/ ing 24-hour creatinine clearance, mainly because of inter-
1.73 m2) at about one month and 5 years after the nephrec- patient and intrapatient variability in creatinine tubular
tomy, respectively.18 secretion and inability of most patients to accurately collect
timed urine samples.22–24 The day-to-day coefficient of varia-
tion of creatinine clearance has been reported as high as 27%
in a routine clinical setting.25
The two recommended formulae for predicting either
Renal function in disease creatinine clearance or GFR are, for adult patients, the
Cockroft–Gault equation26 and the 'abbreviated' MDRD
The aim of the following chapter is to go through every issue study equation (MDRD for Modification of Diet in Renal
related to the measurement of renal function and to define Disease), respectively and in children, the Schwartz27 and
which methods are adequate for which patients'. However, Counahan–Baratt28 equations, respectively (Table 1.1).
we agree that no single test of GFR is perfectly suited for However, these recommendations do not answer the
every clinical and research application. Thus, the goal should question 'which methods for which patients' since the guide-
be to propose a specific clinical question (screening, confirm- line about estimation of GFR only states that 'all four formulae
ing, following, …) the most accurate, precise, safe, conve- reviewed provide a marked improvement over serum
nient and cost-effective (not only the cheapest) method. creatinine alone' for clinical assessment of kidney disease.
Recently, the Kidney Disease Outcome Quality Initiative Moreover, K/DOQI guidelines acknowledge, firstly that
(K/DOQI) of the National Kidney Foundation (USA) has 'estimation of GFR and creatinine clearance from serum
proposed guidelines,1 among which one is dedicated to the creatinine is critically dependent on calibration of the serum
 Introduction 5

Table 1.1 Equations recommended by the National Kidney Foundation (NKF/DOQI) to predict creatinine
clearance and GFR based on serum creatinine

Adult
(140 – age) × weight
Cockcroft–Gault26 CCR (ml/min) =  × (0.85 if female)
72 × SCR

Abbreviated MDRD78 GFR (ml/min/1.73 m2) = 186 × (SCR)–1.154 × (age)–0.203 × (0.742 if female) × (1.210 if African-American)
= exp(5.228 – 1.154 × Ln [SCR] – 0.203 × Ln (age) – (0.299 if female) + (0.192 if
African-American))

Children
0.55 × length
Schwartz27 CCR (ml/min) = 
SCR
0.43 × length
Counahan–Baratt28 GFR (ml/min/1.73 m2) = 
SCR

SCR, serum creatinine in mg/dl (to convert mg/dl to µmol/l multiply by 88); CCR, creatinine clearance; weight in kg; length in cm; age in years.

creatinine assay', and secondly that, 'in certain clinical situa- nine levels, but conversely higher results for high creatinine
tions, clearance measures may be necessary to estimate GFR'. levels. Due to the lack of a standardized calibration procedure
Numerous methods are used to measure creatinine, using several concentrations (with at least one between 0.1
mainly colorimetric (based on Jaffe' reaction) or enzymatic and 1.7 mg/dl or 90–150 µmol/l), the intra-assay variation is
assays. The more commonly used colorimetric methods too high to allow prediction of creatinine clearance or GFR
systematically overestimate creatinine concentrations by from serum creatinine levels, contrarily to K/DOQI guidelines
about 20% compared to enzymatic measures (lower inter- recommendations. Similar conclusions have been reached by
ference with noncreatine chromogens) and by 20% to 80%, other groups working on the prevalence of low GFR in
when compared to high-performance liquid chromatography nondiabetic Americans31,33 or on the risk factors on renal
(HPLC) and dilution mass spectrometry measures, which function (Prevention of Renal and Vascular End-stage
should approximate 'true creatinine'.1 The College of Diseases study, PREVEND).34 Although the K/DOQI
American Pathologists29 reported that in laboratories working group has chosen an estimated GFR cutoff of less
surveyed in 1994, creatinine was overestimated on average than 60 ml/min/1.73 m2 for diagnosing chronic kidney
by 13% to 17% (0.12–0.17 mg/dl or 11–15 µmol/l). Serum disease in the absence of kidney damage, an improvement in
creatinine assays on the same samples were 0.23 mg/dlL estimating GFR from MDRD formula could be to include
(20.3 µmol/l) higher at the White Sands Laboratory (Third creatinine assay methods as a covariable in the prediction
National Health and Nutrition Examination Survey, equation used30,32 keeping in mind the interlaboratory varia-
NHANES III) than at the Cleveland Clinic (MDRD study), tion in measurement of serum-creatine.
although both laboratories used Jaffe' reaction-based Since creatinine clearance overestimates GFR, the
methods but on different auto-analysers.30 estimates given by the Cockcroft–Gault and Schwartz formu-
Without any correction of this bias by a calibration factor lae are biased too. Thus, in a large sample of more than 500
(0.81), the prevalence of low GFR (30–60 ml/min/1.73 m2) adults with a wide range of GFR (up to approximately
in NHANES III would have been erroneously increased 90 ml/min/1.73 m2), Cockcroft–Gault formula overestimates
fourfold (12.5 versus 3.2%).31 Recently, the French Society GFR, directly measured by 125I-iothalamate urinary clearance,
of Clinical Biology assessed interassay variation and accuracy by 23%.35 With Schwartz formula, the bias increases
of blood creatinine measurements as well as the effect of the markedly in children with low GFR, with overestimation up
standardization of calibration procedures on interassay varia- to 32% and 67% for GFR (125I-iothalamate clearance)
tion. Thirty frozen human sera and three certified reference between 31–50 ml/min/1.73 m2 and lower than
materials were analysed by 17 creatinine assays (12 colori- 30 ml/min/1.73 m2, respectively.36
metric, four enzymatic and one HPLC).32 Most of the Another issue is the reliability of the claimed statement that
commercially available methods had inaccuracy higher than the four formulas recommended provide a clinically useful
10% for serum creatinine lower than 1.7 mg/dl (150 µmol/l). estimate of GFR' in the K/DOQI guidelines. This statement
The median dispersion factor was 14% between 0.5 and relies on a rather optimistic definition of what is an accuracy
1.7 mg/dl (45–150 µmol/l, the range of mild to moderate sufficient enough for good clinical decision-making. Thus, the
renal impairment) and 8% between 2.9 and 4.0 mg/dl accuracy was defined as the percent of GFR estimates within
(250–350 µmol/l). Moreover, the bias was not constant over 30% of measured GFR (i.e.; in the 70–130% range of GFR
the clinical range of serum creatinine, enzymatic assays measured by radionuclide tracer, inulin or iohexol clear-
producing lower results than colorimetric ones for low creati- ances). Results of about ten studies (see reference 1) assess-