See discussions, stats, and author profiles for this publication at: [Link]

net/publication/311582095

Pre-analytical issues in the haemostasis laboratory: Guidance for the clinical

laboratories

Article in Thrombosis Journal · December 2016

DOI: 10.1186/s12959-016-0123-z

CITATIONS READS

105 4,381

5 authors, including:

Amandine Magnette Bernard Chatelain

Laboratoire Olivier Université Catholique de Louvain - UCLouvain
5 PUBLICATIONS 202 CITATIONS 282 PUBLICATIONS 4,346 CITATIONS

SEE PROFILE SEE PROFILE

Hugo ten Cate François Mullier

Maastricht University CHU UCL Namur
1,042 PUBLICATIONS 30,715 CITATIONS 416 PUBLICATIONS 17,139 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by François Mullier on 12 December 2016.

The user has requested enhancement of the downloaded file.

Magnette et al. Thrombosis Journal (2016) 14:49
DOI 10.1186/s12959-016-0123-z

REVIEW Open Access

Pre-analytical issues in the haemostasis

laboratory: guidance for the clinical
laboratories
A. Magnette1, M. Chatelain1, B. Chatelain1, H. Ten Cate2 and F. Mullier1*

Abstract
Ensuring quality has become a daily requirement in laboratories. In haemostasis, even more than in other
disciplines of biology, quality is determined by a pre-analytical step that encompasses all procedures, starting with
the formulation of the medical question, and includes patient preparation, sample collection, handling,
transportation, processing, and storage until time of analysis. This step, based on a variety of manual activities, is the
most vulnerable part of the total testing process and is a major component of the reliability and validity of results
in haemostasis and constitutes the most important source of erroneous or un-interpretable results.
Pre-analytical errors may occur throughout the testing process and arise from unsuitable, inappropriate or wrongly
handled procedures. Problems may arise during the collection of blood specimens such as misidentification of the
sample, use of inadequate devices or needles, incorrect order of draw, prolonged tourniquet placing, unsuccessful
attempts to locate the vein, incorrect use of additive tubes, collection of unsuitable samples for quality or quantity,
inappropriate mixing of a sample, etc. Some factors can alter the result of a sample constituent after collection
during transportation, preparation and storage.
Laboratory errors can often have serious adverse consequences. Lack of standardized procedures for sample
collection accounts for most of the errors encountered within the total testing process. They can also have clinical
consequences as well as a significant impact on patient care, especially those related to specialized tests as these
are often considered as “diagnostic”. Controlling pre-analytical variables is critical since this has a direct influence on
the quality of results and on their clinical reliability. The accurate standardization of the pre-analytical phase is of
pivotal importance for achieving reliable results of coagulation tests and should reduce the side effects of the
influence factors. This review is a summary of the most important recommendations regarding the importance of
pre-analytical factors for coagulation testing and should be a tool to increase awareness about the importance of
pre-analytical factors for coagulation testing.
Keywords: Haemostasis assays, Pre-analytical phase, Coagulation assays, Recommendations, Dabigatran,
Idarucizumab, Microparticles, Centrifugation

Background physician requesting the performance of a laboratory

The term “pre-analytical phase” describes all actions and investigation on a patient. It is the most vulnerable
aspects of the medical laboratory diagnostic procedure part of the total testing process, where most laboratory
that occurs prior to the analytical phase [1]. errors occur [2, 3]. It encompasses all procedures, starting
This phase is part of the total laboratory procedure with the formulation of the medical question, and in-
consisting of several stages and beginning with the cludes patient preparation, sample collection, handling,
transportation, processing, and storage until time of
analysis.
* Correspondence: [Link]@[Link]
1
Université catholique de Louvain, CHU UCL Namur, Namur Thrombosis and Pre-analytical laboratory errors can arise throughout the
Hemostasis Center (NTHC), NARILIS, Haematology Laboratory, B-5530 Yvoir, pre-analytical phase, because this phase comprises a lot of
Belgium manual activities and accounts for most of the errors
Full list of author information is available at the end of the article

© The Author(s). 2016 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License ([Link] which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
([Link] applies to the data made available in this article, unless otherwise stated.
Magnette et al. Thrombosis Journal (2016) 14:49 Page 2 of 14

encountered within the testing process as a whole. Unsuit- filling the tubes’, ‘sampling process’, ‘processing of samples’,
able, inappropriate or wrongly handled procedures during ‘transportation of samples’, ‘centrifugation’, ‘storage condi-
collection and handling of specimens are very likely to tions’, ‘freezing’, ‘thawing’. Only articles in English or French
lead to pre-analytical errors [3]. were analysed. The database was searched for studies, clin-
Problems may arise during the collection of blood ical practice recommendations and literature reviews.
specimens such as misidentification of the sample, use The findings of our search have been grouped into dif-
of inadequate devices or needles, incorrect order of ferent subtopics. The first one is devoted to the clinical
draw, prolonged tourniquet placing, unsuccessful at- history, bleeding score and physical examination to dem-
tempts to locate the vein, incorrect use of additive tubes, onstrate the importance of the patient’s history regarding
collection of unsuitable samples for quality (e.g. contam- various diseases such as thrombotic diseases in case of
inated, haemolysed) or quantity (e.g. insufficient amount haemorrhagic diathesis or preoperative assessment. The
of blood or inappropriate blood-to-anticoagulant ratio), following chapters concern the sample collection and se-
inappropriate mixing of a sample, etc. Some factors can quence of drawing blood. The last subtopics concern the
alter the result of a sample constituent after collection sampling process, transportation and storage conditions.
during transportation, preparation and storage. The side
effects of such factors of influence can be reduced by The first step before considering laboratory testing;
standardizing the pre-analytical process [4–6]. Clinical history and physical examination
This review aimed at summarizing recommendations Prior to any type of haemostatic testing a bleeding history
with regard to the pre-analytical phase and provides is a prerequisite for the diagnosis of any bleeding disorder
some guidance to reduce the effects of biological factors and should guide further laboratory investigations [7, 8].
that can have a significant impact on patient care. If this is done diligently, it can be considered as one of
the best screening tests of the risk of bleeding [7, 9, 10]. It
Methods allows the detection of pathology known to cause abnor-
A systematic review was conducted following the recom- mal primary haemostasis (malnutrition, haematologic dis-
mendations of the French Study Group on Haemostasis ease, liver failure, etc.). In case of acquired haemostatic
and Thrombosis (GFHT), the World Health Organization disorders, family history can be seen as a family tree of the
(WHO), the Clinical and Laboratory Standards Institute types of haemorrhage observed as well as their spontan-
(CLSI) guidelines, the International Society on Throm- eous or provoked features. As far as a person’s bleeding
bosis and Haemostasis (ISTH), the European Federation history is concerned, it is important to know the terms of
of Clinical Chemistry and Laboratory Medicine (EFLM) appearance of haemorrhages, their frequency and their
and the British Committee for Standard in Haematology source, and whether bleeding is aggravated after taking as-
(BCSH) about the pre-analytical phase and the pre- pirin and/or other medications like antiplatelet therapy,
analytic variables which can have an impact on the quality NSAIDs, etc. Regarding the surgical and obstetrical his-
of medical laboratory results and about the procedures for tory, it is important to note the number of interventions
sample collection, processing, transportation, and storage that resulted in bleeding complications (new intervention
in haemostasis testing. due to complication, transfusion, admission to intensive
The development of the present literature review was car- care, etc.), but also the number of times that trauma was
ried out using the PubMed database records including ana- not associated with haemorrhagic complications. Anteced-
lysis of references from selected articles from 1991 to 2016 ents of anemia, iron supplementation, hospitalization or
based on the following key-words: ‘pre-analytical phase’, blood transfusion can also guide the clinician [11, 12]. The
‘pre-analytic variables’, ‘pre-analytical quality’, ‘preoperative/ patient’s blood group provides relevant information re-
pre-operative tests’, ‘order of draw’, ‘phlebotomy’, ‘screening lated to haemostatic activity (e.g. activity levels of VWF).
tests’, ‘routine tests’, ‘screening testing’, ‘hemostasis/haemosta- Patients with blood group O are known to have lower
sis’, ‘coagulation tests’, ‘bleeding history’, ‘preoperative bleed- levels of VWF. The presence of the H-antigen (O group)
ing questionnaire’, ‘preoperative evaluation’, ‘bleeding risk’, promotes the cleavage of VWF by ADAMTS13 metallo-
‘haemorrhage/hemorrhage’, ‘surgery’, ‘von Willebrand dis- protease, and, moreover, a higher hepatic clearance of
ease’, ‘von Willebrand factor’, ‘inherited bleeding disorders’, VWF [13, 14].
‘bleeding score’, ‘quality’, standardization’, ‘collection of blood’, In order to prevent disturbing pre-analytical influences,
‘sample collection’, ‘tourniquet’, ‘samples’, ‘transportation’, any interfering drugs should be administered after collect-
‘preparation’, ‘storage’, ‘recommendations’, ‘discard tubes’, ‘acti- ing a blood sample. A record of all the drugs that the sub-
vated partial thromboplastin time, PTT, aPTT’, ‘prothrom- ject took during the week prior to testing should be
bin time, PT’, ‘international normalized ratio, INR’, ‘bleeding collected [15]. Treatment with desmopressin, in treating
time’, ‘platelet count’, ‘platelet function testing’, ‘PFA-100’, or preventing bleeding episodes in patients with von
‘blood sampling’, ‘sample tubes’, ‘anticoagulant’, ‘order of Willebrand disease, haemophilia A and platelet function
Magnette et al. Thrombosis Journal (2016) 14:49 Page 3 of 14

defects, should be noted [16]. It is important to know if coagulopathy. In case of primary haemostatic abnorma-
the patient is pregnant. Pregnancy is associated with in- lities, it is common to seek a cutaneo-mucous type of
crease in fibrinogen, factors VII, VIII, X, VWF, D-dimer haemorrhage. Elements usually seen in haemorrhagic
concentration and with increase in levels of prothrombin syndrome are bruising, purpura, or haemarthrosis. In case
fragments 1 + 2 and thrombin-antithrombin III com- of testing for thrombophilia, signs of previous venous
plexes. There is a decrease in physiological anticoagulants thrombosis including post-thrombotic syndrome (venous
manifested by acquired activated protein C (APC) resist- insufficiency, corona phlebectatica etc.) can be inform-
ance. Free protein S is decreased secondary to increased ative. Moreover, it is very important that the clinician and
levels of its binding protein, the complement component the laboratory exchange all relevant information as much
C4b [17–19]. The overall fibrinolytic activity is impaired as possible [25].
during pregnancy, but returns rapidly to normal following
delivery. This is largely due to placental derived plasmino-
gen activator inhibitor type 2 (PAI-2), which is present in Bleeding score
substantial quantities during pregnancy [19]. Microparti- To collect the bleeding history, in case of haemor-
cles derived from maternal endothelial cells and platelets, rhagic diathesis or preoperative assessment of bleed-
and from placental trophoblasts may contribute to the ing risk, and to obtain the resulting bleeding score
procoagulant effect [19]. (BS), standardized and validated clinical tools are ne-
Hormonal contraceptives can be responsible for inter- cessary [10].
ferences in coagulation testing and may lead to increased a) haemorrhagic diathesis: In an attempt to
concentrations of fibrinogen, prothrombin and factors VII, standardize the diagnostic criteria of von Willebrand dis-
VIII and X, and reduction in coagulation inhibitors, such ease (VWD), the most prevalent congenital bleeding
as antithrombin (AT), protein S and tissue factor pathway disorder which occurs with equal frequency among men
inhibitor (TFPI) [17, 20, 21]. Use of combined oral contra- and women and affecting up to 1% of the general popu-
ceptives is associated with a three-to six-fold increased lation, a BS has been developed [26]. This one is useful
risk of venous thrombosis. This increased risk depends on for the identification of subjects requiring laboratory
the estrogen dose as well as the progestogen type of oral evaluation for VWD and for assessing the clinical severity
contraceptives [22]. These changes tend to me more pro- of the disorder in patients with type 1 VWD [7, 27–29].
nounced in women taking third generation combined oral This BS, related to the number and the severity of bleed-
contraceptives than on second generation [21]. Women ing symptoms, is based on a standardized questionnaire
using combined oral contraceptives with the highest risk and used to evaluate haemorrhagic symptoms [7, 29]. The
of venous thrombosis (e.g. containing desogestrel, cyprot- questionnaire has proven to be useful for diagnostic pur-
erone acetate or drospirenone), have lower levels of PS poses, allowing the establishment of quantitative cut-offs
and TFPI than women using the combined oral contra- discriminating healthy subjects and carriers of VWD [7].
ceptive with the lowest risk of venous thrombosis (i.e. The BS is proposed as a predictor of clinical outcomes of
contraceptives containing levonorgestrel) [22, 23]. inherited VWD, and may also help to identify cases at
Reduced concentrations of protein S can be caused by higher risk of frequent and severe bleeds requiring more
functional impairment of this protein by differences in intensive prophylactic regimens [30].
modulation of its activity by other plasma proteins that Other BS have been developed for the large population
change during the oral contraceptives use or by changes of patients with atrial fibrillation (AF) that require oral
induced in the protein S molecule that impair its anti- anticoagulant therapy (e.g. HAS-BLED, ATRIA, ORBIT
coagulant activity [24]. For platelet function assays, treat- risk scores, etc.) [31].
ment with drugs known to reversibly inhibit platelet The combination of the standardized bleeding question-
function (e.g. NSAIDs) should be stopped at least 3 days naire and an interpretation grid has been referred to as a
before sampling, and treatment with drugs known to irre- Bleeding Assessment Tool (BAT) [7]. Although existing BS
versibly inhibit platelet function (e.g. aspirin, thienopyri- have limitations (e.g. dependence on clinician interpretation
dines) should be stopped at least 10 days before sampling, regarding patient recall, inability to distinguish among
if possible [8, 15]. As it oftentimes impossible to stop all bleeding events occurring at different anatomical sites, etc.)
medication before being sampled for coagulation studies, their use is strongly encouraged, as BATs have proven valid-
drug-induced effects on platelet function should be ity in assessment of symptom severity and help identify pa-
considered when interpreting results [15]. Treatment con- tients needing further investigations [10, 32, 33].
tinuation is also sometimes recommended to assess the ef- To ensure reliability and reduce the time load, the
ficacy and safety of antithrombotic drugs [9]. questionnaire should be performed by a physician or
The referring physician should also have done a other health professional experienced in assessing a
proper physical examination in order to detect signs of (mild) bleeding disorder (MBD), particularly in patients
Magnette et al. Thrombosis Journal (2016) 14:49 Page 4 of 14

who have abnormal clotting screening test results requi- appropriate instructions before blood sampling and
ring invasive procedures or elective surgery [7, 34]. proper training of those persons involved in sampling
The ISTH/SSC Joint Working Group agreed to estab- procedures can reduce or prevent negative influences on
lish a single BAT to standardize the reporting of bleed- laboratory results and their misinterpretation [4, 35].
ing symptoms which would be useful for both paediatric In order to minimize misinterpretation of laboratory re-
and adult populations. More quantitative BATs have sults, blood samples should be collected from fasting sub-
been proposed to improve the diagnostic criteria for jects in the morning between 7 and 9 a.m. and from
MBD because most patients with an MBD do not show subjects who have refrained from smoking for at least
a definitive bleeding history, and are difficult to distin- 30 min [4, 8, 15, 36, 37]. Consumption of caffeine should
guish from normal subjects [7]. be discouraged within the 2 h prior to sampling [4, 15].
This revised BAT could be more generally applicable to Physical activity within the 2 h prior to sampling is not rec-
the investigation of other inherited bleeding disorders with ommended. Blood samples should be collected after the
variable expressivity (e.g. mild platelet function disorders) subject has rested for a short period (at least 5 min) [15].
to avoid the bias of subjective investigator evaluation of Diet and drinking are major factors influencing a num-
haemorrhagic symptoms and to reduce the need for ex- ber of analyses. Before blood sampling, the disturbing in-
pensive laboratory investigations [7, 8, 29]. Even if the fluences of food and drink should be excluded. Patients
ISTH-BAT has not been evaluated sufficiently in inherited should not be studied after meals associated with a high
platelet function disorders to allow a firm recommenda- fat content, so as to avoid the formation of chylomicrons
tion of its use, it could also be potentially useful for the in plasma, which will interfere with light transmission
diagnosis of platelet function disorders [32, 33]. aggregometry [15, 37]. A light meal does not influence
b) pre-operative assessment: For preoperative investiga- the laboratory coagulation tests. However, to completely
tions, some guidelines are available. Recently the French metabolize the lipids intake, sampling after at least 8 h
Society of Anaesthesia and Intensive Care issued recom- (better 12 h) fasting and reduced activity (bed rest) is
mendations for the prescription of routine preoperative recommended [4, 38]. It is also advisable not to eat dark
testing before a surgical or non-surgical procedure, re- chocolate [39].
quiring any type of anaesthesia [9]. The aim of pre- Stress should be avoided under all circumstances (e.g.
anaesthetic screening, based on detailed patient inter- restless child), because stress increases acute phase pro-
views and on physical examination, is to identify patients teins of which von Willebrand factor (VWF), factor VIII
with an increased risk of perioperative bleeding to min- and fibrinogen [40] are most important (in particular in
imise perioperative haemorrhagic complications. the workup for haemophilia or VWD). Smoking around
Routinely haemostasis testing has very little thera- the time of phlebotomy enhances platelets aggregability,
peutic impact and lacks prognostic utility. Hence, and induces a procoagulant state in plasma [41, 42].
haemostasis testing should not be systematically used in Studies have reported an increase in blood coagulability
patients whose history and clinical examination results but impaired fibrinolysis in habitual smokers when com-
do not suggest any haemostatic disorders. Conversely, in pared with nonsmokers [42]. It seems that higher plasma
patients with a positive history of haemorrhagic diath- levels of fibrinogen and viscosity are the main contribu-
esis, haemostasis testing should be requested depending tors to higher coagulability found in smokers, whereas
on the suspected disease [9]. the lower fibrinolytic potential is mainly attributed to an
These recommendations are in line with those issued in increase in PAI-1 activity and possibly also a decrease in
the United States and in the United Kingdom by the Na- tPA activity and lower plasminogen levels [41, 42].
tional Institute for Health and Clinical Excellence (NICE). Smokers may have impaired acute substance P-induced
Only the Italian SISET recommendations (Società Italiana endothelial release of active tPA in vivo related to im-
per lo Studio dell’Emostasi et della Trombosi) suggest per- paired endothelial function, which suggests a possible
forming systematic haemostasis testing with aPTT, PT and direct link among impaired endogenous fibrinolysis,
platelet count prior to any intervention, even in patients endothelial dysfunction, and arterial atherothrombosis in
with no bleeding diathesis history [9]. smokers [42].
Caffeine enhances fibrinolytic potential as whole blood
Variables during sample collection fibrinolysis time is shortened and PAI-1 levels are de-
During the pre-analytical phase, in case of haemostasis creased, whereas tPA activity increases after consump-
testing for patients with a positive history of haemor- tion of coffee and such effects are blunted during
rhagic diathesis, certain steps regarding the preparation caffeine abstinence [42]. About dark chocolate, platelet
of patients and the execution of sampling (specimen col- aggregation is modulated by a flavanol-independent
lection, transportation, sample preparation and storage) mechanism that is likely due to theobromine which is
are of special importance. Providing patients with the containing by cocoa products [39].
Magnette et al. Thrombosis Journal (2016) 14:49 Page 5 of 14

Physical activity may cause an increase in leukocyte biases in the measurement of several haematologic param-
count and activation of coagulation (decrease of pro- eters [47]. Hence, the tourniquet should be placed tightly
thrombin time (PT) and fibrinolysis). It can also activate but less than one minute in order to prevent haemocon-
partial thromboplastin time (aPTT), cause an increase in centration, increased fibrinogen and factors VII, VIII, XII
D-dimer, tissue plasminogen activator (tPA) and plas- as well as activation of endothelial cells and therefore fi-
minogen activator inhibitor (PAI), an exercise-induced brinolysis [47, 48]. Some special coagulation assays, such
adrenaline release on platelet aggregation and an in- as those that measure thrombin generation markers (e.g.,
crease in platelets [4, 15, 36]. Strenuous exercise also thrombin antithrombin complex (TAT) and prothrombin
promotes the release of microparticles by platelets and fragment 1 + 2), should be drawn without the use of a
triggers a transient procoagulant condition, which is tourniquet because that may lead to spurious elevation of
mirrored by increased thrombin generation, platelet hy- these markers, particularly if the tourniquet is left in place
perreactivity, increased activity of clotting factors, com- for more than one minute [5]. A quality laboratory man-
pounded by an increased fibrinolytic activity [41, 43]. ager should verify the length of tourniquet application and
forearm clenching in order to eliminate this source of la-
Training in phlebotomy boratory error and safeguard quality throughout the total
Phlebotomy is the act of puncturing a vein for the testing process [48].
purpose of withdrawing blood and is one of the most Venipuncture may be necessary, on occasion, to obtain
critical parts of the entire pre-analytical phase [44]. blood from an existing vascular access device such as an
Training of phlebotomists is a pre-analytical challenge intravenous (IV) line or a central line [5]. If an IV line is
and requires continuous educational updates as equip- present, the sample site selected is distant from the line.
ment is changed in the healthcare institutions [45]. Butterfly devices and 23 gauge needles can be used on
All staff members should be trained in phlebotomy, to patients with difficult veins (geriatrics, oncology and
prevent unnecessary risk of exposure to blood and to re- paediatrics), provided that the tubing is short (length
duce adverse events for patients [46]. If the phlebotomist <6 cm and air volume <150 μL) [36, 45].
is a member of the laboratory staff, he/she will be aware However, owing to costs and the risk of obtaining un-
of the impact of the quality of sampling on the quality of suitable samples, the use of butterfly needles and IV cathe-
results. If he/she is external to the laboratory, the labora- ters has generally been discouraged, because haemolysis,
tory must provide all information necessary about good activation of the contact system, initiated by activation of
sampling practices [6, 36]. factor XII, can occur when blood comes into contact with
the surface of a biomaterial [2]. Blood samples should be
Blood sampling obtained in a relatively atraumatic fashion, and during col-
According to the recommendations of the EFLM, tube la- lection the blood should flow freely into the collection
belling must be done, before or after venipuncture, in the container [4]. In case of difficult sampling, at least this in-
presence of the patient. The phlebotomist should label formation should be recorded for adequate interpretation
each drawn tube with the patient’s full name, patient’s date of haemostasis assays [50]. Blood sampling should be per-
of birth, identification number [6, 38]. Of course, the use formed without generating foam or bubbles in the sample
of prebarcode tubes may facilitate collecting this informa- and with as little shear stress as possible [4].
tion. The sampling equipment must be sterile and non- The passage of blood through butterfly tubing and IV
pyrogenic. The expiry dates of the needle and the integrity catheters might cause increased haemostatic alterations
of the sterile seal should be checked. The phlebotomy pro- in comparison to blood collection using a conventional
cedure provided by the CLSI requires wearing gloves, straight needle, directly into the tube [2].
cleansing the venipuncture site and drying of skin before The direct transfer of blood specimens from syringes
applying the tourniquet and selecting the venipuncture to blood collection tubes by piercing of the rubber stop-
site and vein [47]. The venipuncture should be ideally col- per of the tube is a practice that should be avoided. This
lected directly from a peripheral vein (antecubital vein). If may cause haemolysis when cells under pressure from
a tourniquet needs to be used, it should immediately be the plunger collide with the tube wall [4].
released when the first tube starts to fill [48]. The diam- The investigation of platelets is highly vulnerable to a
eter of the needle should preferably be comprised between broad series of pre-analytical variables [8]. In particular,
19 and 22 gauge [34, 49]. the vacuum aspiration of blood into primary collection
It is important to take samples so as to reduce platelet tubes could have an influence on platelet function test-
activation in-vitro and to restrict the use of the tourniquet ing. A syringe system permitting slow manual drawing
[15, 36, 49]. Extended tourniquet application might pro- of blood may be superior [4]. For the Platelet Function
duce unnecessary venous stasis or in-vitro haemolysis, Analyser (PFA-100), which replaces the bleeding time in
which could introduce spurious and clinically meaningful many laboratories, the lower shear stress generated by
Magnette et al. Thrombosis Journal (2016) 14:49 Page 6 of 14

manual aspiration of blood into the primary collection generally contain an anticoagulant. The CLSI guidelines
tube would prevent spurious hyper-activation of plate- on the collection of blood specimens for coagulation
lets, thus, preserving the integrity of their function for testing recommend the use of sodium citrate in 105–
subsequent testing on PFA-100 [8, 27, 51]. 109 mmol/L (3.2%) tubes. The use of mmol/L should be
Optimally, blood sampling is done in the laboratory preferred to %. The pH of the anticoagulated plasma
performing the analyses, which allows for verifying many should be comprised between 7.3 and 7.45 [35, 36].
of the above mentioned steps including when needed The vast majority of the samples should be collected
getting additional data on medical history, medication, into trisodium citrate, buffered anticoagulant to help keep
fasting state etcetera. Patient identification and tube la- the pH stable during processing and testing [15, 36]. The
belling are the most critical steps during phlebotomy, anticoagulant effect of sodium citrate is attributed to its
providing an essential safety barrier to prevent patient ability to bind calcium, making the calcium unavailable to
identity mix-up. promote clot formation [4]. However there may also be
Labelling blood tubes in the absence of the patient is a higher citrate concentration (i.e., 3.8% or 129 mmol/L)
potentially life threatening error. According to CLSI, pa- leading to greater calcium binding and longer clotting
tient identification is the responsibility of the phlebot- times. Specimens collected in 3.8% buffered sodium citrate
omist to ensure that blood is drawn from the individual may prolong the PT and aPTT and underestimate fibrino-
designated on the request form [6] (Table 1). gen if the normal range is based on 3.2% citrated samples
[4, 47]. Due to the variation in clotting times and sodium
Sample tubes and anticoagulant citrate concentration, the consensus recommendation
Blood samples for coagulation analyses should be drawn suggests that laboratories should standardize to one citrate
into siliconized glass tubes, or plastic (polypropylene) concentration and develop appropriate reference intervals
tubes [35, 49]. Vacuum tubes must be sealed and CE [4, 5, 35, 36, 47].
marked [15, 36]. Care must be taken to follow manufac- For some analyses, especially platelet function assays,
turer’s expiry dates. Sample tubes for haemostasis analyses buffered citrate solution or other anticoagulants are
used. Citrate, theophylline, adenosine and dipyridamole
Table 1 Summary of key pre-analytical recommendations about (CTAD) is a cocktail of additives that prevent in-vitro
blood sampling platelet activation recommended for measurement of
Blood sampling Use siliconized glass or plastic (polypropylene) tubes. platelet-activation markers such as β-thromboglobulin
Blood samples should be drawn into 105–109 mmol/L or platelet factor 4 (PF4). These tubes allow for more re-
sodium citrate, buffered anticoagulant. liable measurement of unfractionated heparin (UFH)
The pH of the anticoagulated plasma should be and are useful for the study of membrane glycoproteins
comprised between 7.3 and 7.45. platelet by flow cytometry. They should be kept away
Perform blood collection from fasting subjects in the from light [36]. Sampling these tubes allows a delay in
morning (between 7 and 9 a.m.).
analysis until 4 h before centrifugation [52].
The patient should be relaxed. Stress should be It is easy to ensure that the anticoagulant is consistent
avoided.
when working on the primary tube; vigilance is therefore
Label each drawn tube with the patient’s full name, required in case of analysis performed on an aliquot [53].
patient’s date of birth, identification number.
Collect venipuncture directly from a peripheral vein
Sequence of drawing blood
(antecubital vein).
A standardized sequence of blood sampling must be
The diameter of the needle should preferably be
comprised between 19 and 22 gauge. respected in order to avoid carry-over between tubes [3].
National and international (WHO, CLSI) guidelines
Release the tourniquet immediately when the first
tube starts to fill (<1 min). recommended that the order of draw of blood during
The order of drawing blood during phlebotomy
phlebotomy should be blood culture/sterile tubes, then
should be blood culture/sterile tubes, then coagulation coagulation tubes, then plain tubes/gel tubes, then tubes
tubes, then plain tubes/gel tubes, then tubes containing additives [3–5]. According to the CLSI, a first
containing additives.
discard tube (or a non-additive tube) is unnecessary for
Draw a discard tube when citrated plasma is obtained routine coagulation assays [47]. A discard tube (or a cit-
using butterfly systems or other IV catheter devices.
rate tube if the sampling is difficult) must be drawn
Discard tube may be considered to ensure correct when citrated plasma is obtained using butterfly systems
filling of sample tubes for coagulation tests.
or other IV catheter devices because the air volume con-
Ensure correct filling of tubes (>90% filling).
tained in the tubing partially fills the vacuum tube, lead-
Respect the required ratio of sodium citrate to whole ing to insufficient filling of the tube with citrate. A
blood (1:9).
discard tube is also recommended when samples are
Magnette et al. Thrombosis Journal (2016) 14:49 Page 7 of 14

subject to platelet function analysis and for thrombin haematocrit, the citrate volume may be adjusted because
generation measurements [2, 4, 8, 47]. this may also impact the citrate-calcium ratio. It may be
Published studies have demonstrated that for routine necessary to remove part of the citrate solution from the
coagulation testing, the use of a discard tube is not ne- sampling tube prior to drawing blood [35, 36, 44] (Table 2).
cessary because there was no significant difference in the
aPTT and PT results between the first and second tubes
Mixing the samples
drawn [2, 4, 5, 36, 44, 54]. Some studies evaluated the
Following collection, blood should be adequately and
need for discard tubes in a variety of others coagulation
promptly mixed by three to six complete end-over-end
tests (e.g. fibrinogen, D-dimer, factors II, V, VII, VIII, IX,
inversions of the tubes in order to ensure complete dis-
X, XI, proteins C and S and AT, …) and suggested that
tribution of anticoagulant [8, 36, 47, 49, 57].
discard tubes are not necessary when drawing samples
Mixing samples is an important way to prevent in-
for specialized coagulation testing [2, 5, 54, 55]. There
vitro clot formation. Sometimes samples are inappropri-
are no data to support the need for a discard tube for
ately mixed or left unmixed for a long time, thereby
specialized haemostasis assays [5]. However, due to lack
avoiding full contact of the blood with the anticoagulant,
of sufficient evidence, the practice of drawing a discard
which determines partial clotting.
tube should still be recommended [5].
Vigorous shaking, vortexing or agitation of blood sam-
Contamination can occur through contact from micro
ples should be avoided in order to prevent inducing
drops from caps with a compress, or from contact with
haemolysis or spurious platelet and factor activation that
the interior of the tube with the syringe and from the
may result in shortened clotting times or false elevation
anticoagulant of the first tube to the second tube [4].
of clotting factor activity in specimen tests (e.g. factor
Contamination of coagulation assay tubes is possible if
VII) [4, 8, 48, 58].
coagulation tubes are drawn following an additive tube
like certain serum collection tubes containing clot acti-
vators [36]. In case of contamination, the results will be Examination of samples
erroneous and a re-collection will be necessary, leading When obtaining plasma for coagulation testing and prior
to a delay in subsequent medical decisions. to processing, samples must be examined for the pres-
For platelet function assays and coagulation factors, we ence of a clot, precipitates and signs of haemolysis.
suggest numbering each tube according to the sample col- In-vitro clots may develop in samples where the blood
lection order. If a defect is observed on one tube, it could is slow to fill the collection container, where there is
be useful to check this number and verify on another tube. prolonged use a tourniquet, or when considerable ma-
nipulation of the vein by the needle has occurred. These
Sampling process situations must be avoided.
Filling the tubes Several causes may induce in-vitro haemolysis, such as
Sodium citrated tubes must be filled up to 90% of the slow or difficult specimen collection; prolonged tourni-
nominal volume or to the mark noted on the tube if pro- quet placing; use of incorrect devices (e.g., butterfly nee-
vided [41, 47]. The required ratio of sodium citrate to dles or IV catheters) or needles (e.g., small gauge
whole blood is 1:9 [35, 49]. needles); unsuccessful attempts to locate the vein; in-
Under-filling of tubes is another important source of appropriate mixing of the sample; inappropriate centrifu-
error and severely affects laboratory results. An insuffi- gation speed (e.g., > 1500g); inappropriate transportation
cient volume for testing greatly modifies the fixed blood- procedures (pneumatic tube systems, duration, temperature
to-anticoagulant ratio. Under-filling increases the and humidity), … [2, 4, 5].
dilution of the sample due to the volume of liquid anti-
coagulant, and may increase the clotting time due to the Table 2 Summary of key pre-analytical recommendations about
excess calcium-binding citrate present [4, 5, 47]. sample processing
It has been reported that when tubes are drawn at less Sample processing Blood should be adequately and promptly mixed
by 3 to 6 complete end-over-end inversions of the
than 89% of total fill, a clinically significant bias exists in tubes in order to ensure complete distribution of
test results for aPTT, less than 78% for fibrinogen, and less anticoagulant.
than 67% for coagulation factor VIII, whereas PT and acti- Avoid vigorous shaking, vortexing or agitation of
vated protein C resistance remain relatively reliable even in blood samples.
tubes drawn at 67% of the nominal volume [41]. Similar re- Adjust the citrate volume in case of high
sults were obtained in further studies [45, 56]. Instead of haematocrit (remove part of the citrate solution
from the sampling tube prior to drawing blood).
rejecting a sample tube, it may be useful to note the sample
volume and to adapt the result depending on the additional Check tubes for presence of clots, precipitates
or haemolysis.
dilutional (e.g. for factor assays, fibrinogen). In case of high
Magnette et al. Thrombosis Journal (2016) 14:49 Page 8 of 14

Haemolysis can affect some tests of haemostasis, ei- During transportation, samples should be transferred
ther because of the presence of thromboplastic sub- vertically in the shortest possible time [36].
stances or interference of haemoglobin pigment with Delays between sample collection and analysis can
photo-optical systems [2, 49]. cause in-vitro degradation of coagulation proteins [2].
Lysis of the red cell membranes induces the release of Ideally, samples for coagulation assays should be per-
red cells contents (many intracellular enzymes, ADP …) formed in the laboratory that performs the assays. If not,
into plasma and may lead to activation of the plasma specimens should be shipped from peripheral collection
sample altering coagulation parameters and activation of facilities to the core laboratory utilizing current CLSI
other bloodlines (leucocytes, platelets). guidelines (non-refrigerated at ambient temperature in
Haemolysed samples may lead to early flow obstruction as short a time as possible, preferably within the first
in the PFA, presumably due to platelet activation, frag- hour after collection) [5, 57]. Blood samples for coagula-
mentation of platelet and red blood cells and the presence tion analyses should not be shaken. Any sample that has
of micro-thrombi [4]. Haemolysis may lead to statistically been dropped should be discarded. For the transport, a
significant increases in PT and D-dimer. The aPTT can be box maintaining blood tubes in a steady vertical position
falsely prolonged or shortened and AT and fibrinogen de- should be used. Transporting blood tubes in a vertical
creased by in vitro haemolysis [5, 49, 59, 60]. rather than a horizontal position limits the extent of in
Plasmas that are lipemic and icteric may also affect vitro microparticles (MP) generation [62, 63].
analytic results by interfering with optical absorbance or The temperature during transport is of special rele-
impeding light transmittance [49]. Mechanical and/or vance. Extreme temperatures should be avoided in order
electromechanical methods for clot detection should be to maintain sample integrity [5]. Temperatures higher
utilized when possible for these plasma samples [5]. Be- than room temperature can lead to degradation of factor
cause the presence of lipid particles can still bias the V and factor VIII [64].
measurement for biologic interference, the CLSI cur- The use of pneumatic tube systems (PTS) for transport
rently recommends the removal of excess triglycerides of samples is problematic and could have a significant influ-
by ultracentrifugation [61]. This kind of equipment be- ence on platelet function testing [8, 45]. Rapid acceleration
ing unavailable in most routine laboratories due to high and deceleration may induce excessive vibration, denature
costs and incompatibility with daily practice and because proteins and result in haemolysis, platelet activation, and
ultracentrifugation may cause precipitation of the large other effects [4, 35, 63]. Clinical decisions regarding platelet
molecular proteins such as fibrinogen or factor VIII/ function and aspirin responsiveness should not be based on
VWF complex, an alternative approach entails high blood specimens transported by a PTS [65–68] (Table 3).
speed microcentrifugation (e.g., double centrifugation at
>20,000 g for 15 min) or lipid extraction by means of or- Specimen rejection
ganic solvents or lipid-clearing agents [45, 61]. Each laboratory should have guidelines for rejection of
samples; some criteria are obligatory: inappropriate col-
lection tubes and additives, outdated tubes, error in pa-
Transportation of samples tient identification or lack of identification, insufficient
Samples should be transported non-refrigerated at ambi- volume, haemolysed specimens, identification of a clot,
ent temperature (15–25 °C) in as short a time as possible inadequate volume, haematocrit >50% or <30% for the
[8, 49, 57].
Before transport, samples should be tested regarding Table 3 Summary of key pre-analytical recommendations about
identification, safety conditions and stability. Errors in transportation of samples
either sample identification, sample preparation before Transportation of samples Before transport, test samples regarding
or after transport may have an adverse effect on patients identification, safety conditions and stability.
if they are not detected on time [4, 6]. Transport samples at ambient temperature
When possible, samples should be drawn directly in a (15–25 °C) in as short a time as possible.
laboratory. Immediately after drawing, whole blood Draw samples directly in a laboratory.
should remain capped for transport both for safety rea- Immediately after drawing, whole blood
sons and to minimize loss of CO2, which causes pH to should remain capped for transport.
increase, leading to prolongation of PT and/or aPTT Temperature control is recommended in
[44]. Tubes should be stored at ambient temperature rooms where samples are kept for analysis.
until centrifugation [4, 5, 36]. Temperature control is Transfer samples vertically.
recommended in rooms where samples are kept for ana- Do not use pneumatic tube systems (PTS)
lysis [2]. If tubes need to be transported this should be for transport of samples used for platelet
function analysis.
done with care in order to avoid unnecessary agitation.
Magnette et al. Thrombosis Journal (2016) 14:49 Page 9 of 14

PFA, platelet count <100,000/μL for PFA and platelet on PT and fibrinogen concentration [69]. Therefore, it
aggregation … All samples deemed unacceptable due to seems feasible to centrifuge samples for coagulation ana-
pre-analytic handling and unfulfilled transport require- lyses with the rotor brake set to off [4, 62]. However, ac-
ments should be rejected [4–6, 8]. cording to our personal experience, this should be
Haemolysed blood is the most common reason for reject- validated in appropriate studies. Our experience shows it
ing specimens in the laboratory and, therefore, in-depth is important to check the absence of vibration (during
knowledge on how to properly access the vein with the cor- acceleration/deceleration processes) due to lack of cen-
rect blood collection equipment to avoid haemolysis in the trifuge maintenance.
collected specimens is a must for the phlebotomist [45]. Using relative centrifugal forces greater than 1500g is
In case of reception of an aliquot without information not recommended as this may induce platelet activation,
about the anticoagulant used, this one must be rejected haemolysis or other unwanted effects [4, 5, 47]. However,
because serum and all types of plasma have virtually iden- in case of emergency, for general haemostasis parameters
tical visual appearance. Because samples that are rejected performed on fresh plasma, higher centrifugation force
must be recollected, which gives potential delay in patient (greater than 1500g) and shorter time (less than 10 min)
management, phlebotomists must be fully aware of the can be used [36, 47].
common reasons why specimens are rejected (Table 4). Some studies have evaluated the impact of high accel-
eration centrifugation conditions on routine coagulation
Centrifugation testing (including especially PT, aPTT and fibrinogen; no
Whenever a delay in transport is expected, it might be data were found for the thrombin time) and concluded
advisable to perform local centrifugation and separation. that rapid centrifugation does not modify results and
Plasma is generally prepared by centrifugation of a contributes, by decreasing duration of the pre-analytical
whole blood sample. A temperature-controlled centri- variable, to reduce the turnaround time for these tests
fuge is required for processing routine coagulation as- [70–76]. Note that in the REVERSE-AD study conducted
says. Centrifugation should take place at room to evaluate the efficacy and safety of Idarucizumab, the
temperature (15–25 °C) [36]. The effect of centrifugation reversal agent for dabigatran [77], the centrifugation (i.e.
temperature on MP determination is still unknown [37]. 3 min at 1000g) used before performing dilute thrombin
It is recommended to use a centrifuge that has a rotor time and ecarin clotting time, was not validated [78].
with swing-out buckets to facilitate the separation of Both tests were used as primary endpoints and they
plasma from the cellular components and to minimize should thus be interpreted cautiously.
re-mixing of plasma and red cells [5, 36, 44]. It is recom- a) For analysis in platelet-poor plasma For analysis in
mended to centrifuge the primary tube for coagulation PPP, which is the standard material required for PT,
testing at 1500g for no less than 15 min with a centri- aPTT, fibrinogen, single coagulation factor assays, func-
fuge brake set off [8, 36, 47]. The centrifuge should be tional protein S and C assays, activated protein C resist-
validated before use, every 6 months or after modifica- ance, lupus anticoagulant (LA) assays, antiphospholipid
tions, in order to assure that platelet-poor plasma (PPP) antibody testing, thrombin generation, microparticles
is achieved [47]. measurements, homocysteine, VWF assays, tPA, PAI,
Use of the rotor brake may lead to higher residual mi- plasminogen and antiplasmin, or monitoring of unfrac-
croparticles and platelet counts, which also has an effect tioned heparin therapy, and many other tests, sample
tubes require a second centrifugation and should be cen-
Table 4 Summary of key pre-analytical recommendations about trifuged in capped tubes within 4 h after drawing blood
specimen rejection [2, 36, 37, 79]. For thrombin generation measurements,
Specimen rejection All samples deemed unacceptable due to whole blood is best centrifuged immediately, to prevent
pre-analytic handling and unfulfilled transport activation or degradation of coagulation proteins. In
requirements should be rejected.
order to eliminate platelet debris and microparticles
Inappropriate collection tubes and additives. from plasma, which may contribute to the variability in
Outdated tubes. thrombin generation results, a second centrifugation
Error in patient identification or lack of step at 10,000g is recommended [2].
identification. Double centrifugation significantly reduces the residual
Insufficient volume (depending on the assay). An amount of platelets in a sample and can be performed
alternative may be to adapt the result depending to produce PPP such that the post centrifugation
on the additional dilutional (e.g. for factor assays,
fibrinogen). plasma platelet count is less than or equal to 10x109/L.
Haemolysed specimens (depending on the assay).
Residual platelets in plasma have been known to affect
phospholipid-dependent coagulation tests through the ex-
Identification of a clot.
posure from platelet membranes of anionic phospholipids
Magnette et al. Thrombosis Journal (2016) 14:49 Page 10 of 14

that quench LA activity. This effect leads to shorter co- Table 5 Summary of key pre-analytical recommendations about
agulation time and is particularly evident, especially in centrifugation
test plasmas that undergo freezing and thawing before Centrifugation Use a temperature-controlled centrifuge for processing
analysis. Adequate plasma preparation for LA testing is routine coagulation assays.
an essential prerequisite for reliable diagnosis [80]. Validate the centrifuge before use, every 6 months or
after modifications, in order to assure that platelet-poor
The generation of PPP should be prepared for all coagu- plasma (PPP) is achieved.
lation parameters if there is a possibility that measurement
Check the absence of vibration (during acceleration/
is not performed immediately after centrifugation, or if deceleration processes) due to lack of centrifuge
samples are going to be frozen [5, 63]. maintenance.
Following initial centrifugation, the plasma is carefully Centrifuge the primary tube for coagulation testing at
transferred to a nonactivating plastic centrifuge tube 1500g, 15 min
using an automatic pipette, and then centrifuged again In case of emergency, for PT, APTT and fibrinogen
for about 15 min. When pipetting, 1 cm of plasma must performed on fresh plasma, higher centrifugation force
(greater than 1500g) and shorter time (less than 10 min)
remain above the buffy layer.
can be used.
It is important to use an automatic pipette because The preparation of PPP require a double centrifugation
it permits slow and linear suction, unlike plastic pipettes to obtain a residual platelet count lower than 10x109/L.
that pose a risk of getting a high rate of residual Following initial centrifugation, transfer carefully the
platelets. The plasma is aliquoted to a secondary tube, plasma to a nonactivating plastic centrifuge tube using
an automatic pipette, and then centrifuged again for
taking care not to include the residual platelets that about 15 min.
may have precipitated at the bottom of the centrifuge
The preparation of platelet-rich plasma (PRP) for platelet
tube. It has been demonstrated that a high platelet function analysis requires a centrifugation performed at
count (>200 × 109/L) does not affect results of PT, D- 200–250 g for 10 min without application of a rotor
dimer, fibrinogen and aPTT assays, when samples are brake.
tested fresh and analyzed immediately after centrifu-
gation [36, 79]. These samples should not be stored Storage conditions
for later analyses [47]. a) conditions for the interval from sampling to analysis:
b) For analysis in platelet-rich plasma The preparation For routine coagulation testing like the PT and the
of platelet-rich plasma (PRP) for platelet function ana- aPTT, storage of uncentrifuged samples at room
lysis requires that centrifugation is performed at 200– temperature up to 6 h may yield acceptable results.
250 g for 10 min without application of a rotor brake However, a shorter delay is desirable. Whole blood as-
[8, 15]. These centrifugal forces appear to be the best says should be performed within 4 h after blood sam-
condition for preparing PRP for light transmission pling and centrifugation should ideally be taken within
aggregometry (LTA) studies, both in terms of the degree 1 h [4, 41, 47, 49, 79, 81]. During storage, samples
of contamination of PRP by other blood cells and of plate- should remain capped [5]. Extremes of temperature (e.g.,
let reactivity [15]. The recommendation to avoid using a both refrigerated or high) should be avoided.
rotor brake is based on expert opinion and should be Cold storage of citrated whole blood prior to centrifuga-
demonstrated in appropriate studies. tion, by placing samples either in an ice bath or in refriger-
Centrifugation with less than 150 g does not yield a ated (2–8 °C) storage, is no longer recommended.
sufficient volume of PRP. The PRP should then be trans- Improper storage of whole blood at cold temperature may
ferred to a capped polypropylene tube and a platelet cause VWF and factor VIII values to fall below normal ref-
count performed. erence threshold levels, which may potentially lead to a
For LTA, the platelet count of PRP samples should not false suspicion of haemophilia A or VWD due to inappro-
be adjusted to a standardized value by addition of autolo- priate pre-analytical handling of blood [4, 5, 25, 44, 82].
gous PPP, unless the platelet count is >600x109/L [8, 15]. Once the blood sample has been centrifuged, plasma can
However, the results of LTA studies could be inaccurate remain on the cells in a capped primary tube until testing,
when the platelet count in the PRP samples is lower than or it can be aliquoted and stored in a secondary tube.
150x109/L [8, 15]. When aliquoting the plasma, care must be taken so that
All assays using whole blood or PRP need to be performed the buffy coat (the layer of cells between the red cells and
within a maximum of 4 h after blood sampling [4, 15]. plasma) is not disturbed or that this cellular component is
If the pre-analytical phase of the laboratory is supported not introduced back into the plasma. Aliquoted plasma is
by an automated system, vigilance must be in order to test stable for 4 h when refrigerated, except for plasma for PT-
and validate the centrifugation conditions meeting the re- INR and PT-sec, which should not be refrigerated [2, 5, 47].
quirements of haemostasis but also the requirements of This is due to the potential for cold activation of a sample.
chemistry and immunology analysis [36] (Table 5). This may strongly influence some of the coagulation assays
Magnette et al. Thrombosis Journal (2016) 14:49 Page 11 of 14

and may lead to platelet activation, activation of factor VII, factor VIII and possibly factor V [83]. aPTT from patients
which in turn can give shorter clotting times and hence receiving unfractionated heparin (UFH) therapy has
lower PT-INR and PT-sec results. shorter stability. For these samples, due to the variable
Delays in transport may affect in particular the labile heparin neutralization by platelet factor 4 (PF4), the delay
factors (FV, FVIII), leading to prolonged clotting times before centrifugation should not exceed 1 h for a sample
and in vitro loss of factor activity [2, 83]. collected in citrate and 4 h in CTAD. If the whole blood
The time frame for analysis depends upon the stability sample is centrifuged within 1 h of collection, the plasma
of the analysis, which depends on storage temperatures can be left on top of the cells at room temperature up to
and conditions. As far as the stability of citrated whole 4 h prior to testing [83]. This applies also to fibrinolysis
blood, plasma on cells and aliquoted plasma is con- parameters [35, 47, 52, 81, 83].
cerned, more studies are needed, because there are large PT, aPTT and factor VIII tests from frozen samples
differences between the conclusions of different studies cannot be performed [4]. Freezing has an inconstant and
and partially missing studies. unpredictable effect on the results and may cause signifi-
b) short term plasma storage: PPP should be stored at cant elevations of aPTT, but also PT [4, 84–86]. Poorly
room temperature (20–25 °C) or at −80 °C until analysis. handled frozen material may also causes shortening of
External influences (like temperature, light …) may be of aPTT or PT [5]. If double centrifugation is not done
major impact (Table 6). when preparing a sample to be frozen this can lead to a
Samples for PT/INR have longer stability (24 h) at lysis of residual platelets upon freezing of plasma sample
room temperature. Samples can be stored as whole and lead to shortened APTT results in heparinized pa-
blood or stored following centrifugation [81]. tients [80].
Samples for aPTT testing should be stored at room Freezing leads to a marked decrease especially in FVIII
temperature and be performed using fresh plasma activity [4, 87]. In case of a factor defect observed on a
within 4 h. The limited stability is largely due to time- frozen sample, it is suggested to repeat the analysis on a
dependent generation or loss of labile factors, particulary fresh sample.
For platelet function assays, samples should rest at
room temperature for at least 15 min before analysis.
Table 6 Summary of key pre-analytical recommendations about Testing should be completed within 3 to 4 h of collec-
storage conditions tion [4, 5, 8, 15].
Storage conditions Store samples at room temperature (15–25 °C) Stability of plasma samples for special coagulation as-
until analysis. says (except FVIII, anti-FXa for UFH) is largely un-
Perform whole blood assays <4 h after blood known. The shortest stability is observed for factors V
sampling and centrifugation <1 h.
and VIII. Analyses should be performed within 3 h.
Extremes of temperature (e.g. both refrigerated or For unknown reason, protein S is labile. It has been
high) should be avoided.
demonstrated that protein S activity is unstable, with a
Store PPP at room temperature (15–25 °C) or statistically significant loss of activity demonstrated at
at −80 °C until analysis.
8 h while fibrinogen, protein C and antithrombin activity
If the whole blood sample is centrifuged within
1 h of collection, the plasma can be left on top
appear to remain relatively constant when stored at
of the cells at room temperature up to 4 h room temperature for up to 7 days [83]. VWF appears
prior to testing. to be stable at room temperature for 48 h [4].
Time from sampling to analysis depends on analyte: Stability of the vitamin K-dependant factors has been
Samples for PT/INR have longer stability (24 h) at reported to be 24 h at room temperature [4, 5, 83, 88].
room temperature.
Samples for aPTT should be performed using fresh For thrombin generation, direct plasma preparation is
plasma <4 h (<1 h in patients treated with preferred to whole blood storage. When immediate ana-
unfractionated heparin). lysis is not possible, plasma is most stable when incu-
For platelet function assays, samples should rest at
room temperature for at least 15 min before bated at room temperature, instead of 4 °C or 37 °C,
analysis. Testing should be completed <3–4 h most likely because of more activation and degradation
of collection. of coagulation proteins at these temperatures [2].
For factors V and VIII analyses should be performed
<3 h. For samples coming from external laboratories, it is im-
Fibrinogen, protein C and antithrombin activity portant to check prior to analysis whether the aliquot ex-
appear to remain relatively constant when stored ternal samples (fresh or frozen) really consist of citrated
at room temperature for up to 7 days.
Protein S activity is unstable, with a statistically plasma. For specialized coagulation testing, most samples
significant loss of activity at 8 h. are sent as frozen samples, whilst samples for routine
VWF appears to be stable at room temperature coagulation are more often sent as citrated blood (primary
for 48 h.
tube). Ideally, the primary tube should be received together
Magnette et al. Thrombosis Journal (2016) 14:49 Page 12 of 14

with the aliquot tube [39]. It is not suggested to recen- volume of 500–1200 μL) should be prepared as a backup.
trifuge the aliquots coming from external laboratories. Frozen aliquots should be transported on dry ice [36].
c) long term plasma storage: Samples that cannot be
tested within 4 h should be centrifuged and the plasma Conclusion
aliquot frozen [79] (Table 7). Continuous monitoring and management of pre-analytical
For analyses that can be performed on frozen plasma, errors is crucial in order to improve the quality of the pre-
freezing should be fast (using rapid freezing technique analytical phase, which is essential for patient care.
like liquid nitrogen), and samples should be preferably Standardization efforts are essential to control and prevent
stored at −70 °C (or below) rather than −20 °C. Plasma errors and to ensure the quality of exploration in haemosta-
samples frozen at minus 20 °C remain stable for 2 weeks sis. They are also necessary for all clinical laboratories
for most coagulation parameters [36, 49]. However, it is accredited by International Organization for Standardization
imperative that a frost-free freezer is not used [36, 44, (ISO) document 15189. The accurate standardization of
49, 83]. Plasma frozen at minus 80 °C remains stable for the pre-analytical phase is of pivotal importance to achieve
6 – 18 months dependent on the parameter [47, 63, 64]. reliable results of coagulation tests.
Prior to testing, frozen plasmas should be thawed rap- Owing to the development of large laboratory networks
idly at 37 °C (to prevent denaturing fibrinogen) and and of decentralized phlebotomy services and analytical
tested immediately. This usually takes at least 3–5 min laboratories, standardized and unequivocal procedures
for a 1–2 mL sample. The sample must be mixed gently and protocols are essential for sample collection, including
to resuspend any cryoprecipitate [49]. patient preparation, specimen acquisition, handling and
The samples have to be thawed at least 5 min in a water storage. These procedures are intended to prevent these
bath at 37 °C and not at room temperature, on a bench or problems and to protect against complications and patient
in a microwave oven. After thawing they should be gently mismanagement that could otherwise arise when speci-
stirred [2, 36]. After thawing, and for thrombin generation mens are not collected properly in order to achieve accur-
measurements, plasma is best analysed immediately [2]. ate and reliable coagulation measurements.
Freeze-thawing may produce phospholipid rich mem- The effects of pre-analytical variables on the reliability
brane microvesicles from platelet damage which may then and consistency of screening tests is often forgotten due
mask the presence of a lupus anti-coagulant [89]. to a lack of understanding and awareness. This can be
If a pathological parameter is obtained on a frozen sample, improved by educating healthcare professionals who are
this parameter is suggested to be re-tested on fresh plasma. involved in drawing blood for testing.
Samples may not be re-frozen. Several aliquots (suggested
Acknowledgements
Not applicable.
Table 7 Summary of key pre-analytical recommendations about
freezing and thawing Funding
Not applicable.
Freezing and thawing Do not perform PT, aPTT and factor VIII tests
from frozen samples. Availability of data and materials
Centrifuge samples that cannot be tested within Not applicable.
4 h and frozen the plasma aliquot.
Authors’ contributions
Use rapid freezing technique (liquid nitrogen). AM and FM performed the review of the literature and designed the
Store samples at −70 °C (or below) rather manuscript. AM drafted the manuscript. All authors read and approved the
than −20 °C. final manuscript.

Plasma samples frozen at minus 20 °C remain Competing interests

stable for 2 weeks. The authors declare that they have no competing interests.
Plasma frozen at minus 80 °C remains stable for
6 – 18 months dependent on the parameter. Consent for publication
Not applicable.
Do not re-freeze samples (but prepare a
sufficient number of aliquots). Ethics approval and consent to participate
Thaw samples rapidly at 37 °C (to prevent Not applicable.
denaturing fibrinogen) at least 5 min in a water
bath at 37 °C and not at room temperature, Author details
1
on a bench or in a microwave oven. Université catholique de Louvain, CHU UCL Namur, Namur Thrombosis and
Test immediately. Hemostasis Center (NTHC), NARILIS, Haematology Laboratory, B-5530 Yvoir,
Belgium. 2Maastricht University Medical Centre and Cardiovascular Research
After thawing, mix the sample gently to Institute (CARIM), Department of Internal Medicine, Maastricht, The
resuspend any cryoprecipitate. Do not vortex Netherlands.
or shake.
Received: 3 September 2016 Accepted: 1 December 2016
Do not re-frozen samples.
Magnette et al. Thrombosis Journal (2016) 14:49 Page 13 of 14

References progestogens on protein S and tissue factor pathway inhibitor. J Thromb

1. Guder WG. History of the preanalytical phase: a personal view. Biochem Haemost. 2008;6:346–51.
Med (Zagreb). 2014;24:25–30. 24. Koenen RR, Christella M, Thomassen LG, Tans G, Rosing J, Hackeng TM.
2. Loeffen R, Kleinegris MC, Loubele ST, Pluijmen PH, Fens D, van Oerle R, ten Effect of oral contraceptives on the anticoagulant activity of protein S in
Cate H, Spronk HM. Preanalytic variables of thrombin generation: towards a plasma. Thromb Haemost. 2005;93:853–9.
standard procedure and validation of the method. J Thromb Haemost. 25. Armstrong E, Joutsi-Korhonen L, Lassila R. Interaction between clinic and
2012;10:2544–54. laboratory. Thromb Res. 2011;127 Suppl 2:S2–4.
3. Cornes M, van Dongen-Lases E, Grankvist K, Ibarz M, Kristensen G, Lippi G, 26. Centers for Disease Control and Prevention: [Link] Accessed
Nybo M, Simundic AM, Working Group for Preanalytical Phase EFoCC, 17 Nov 2016.
Laboratory M. Order of blood draw: Opinion Paper by the European 27. Castaman G, Tosetto A, Goodeve A, Federici AB, Lethagen S, Budde U, Batlle
Federation for Clinical Chemistry and Laboratory Medicine (EFLM) J, Meyer D, Mazurier C, Goudemand J, Eikenboom J, Schneppenheim R,
Working Group for the Preanalytical Phase (WG-PRE). Clin Chem Ingerslev J, Habart D, Hill F, Peake I, Rodeghiero F. The impact of bleeding
Lab Med. 2016;55:27–31. history, von Willebrand factor and PFA-100 ((R)) on the diagnosis of type 1
4. Guder WG, Narayanan S. Pre-Examination Procedures in Laboratory von Willebrand disease: results from the European study MCMDM-1VWD.
Diagnostics: Preanalytical Aspects and Their Impact on the Quality of Br J Haematol. 2010;151:245–51.
Medical Laboratory Results. De Gruyter; 2015. 28. Tosetto A, Rodeghiero F, Castaman G, Goodeve A, Federici AB, Batlle J,
5. Kitchen S, Olson JD, Preston FE. Quality in Laboratory Hemostasis and Meyer D, Fressinaud E, Mazurier C, Goudemand J, Eikenboom J,
Thrombosis. Bognor Regis: Wiley-Blackwell, 2nd Edition. 2013; 22–44. Schneppenheim R, Budde U, Ingerslev J, Vorlova Z, Habart D, Holmberg L,
6. van Dongen-Lases EC, Cornes MP, Grankvist K, Ibarz M, Kristensen GB, Lippi Lethagen S, Pasi J, Hill F, Peake I. A quantitative analysis of bleeding
G, Nybo M, Simundic AM. Working group for preanalytical phase EFoCC, symptoms in type 1 von willebrand disease: results from a multicenter
laboratory M. Patient identification and tube labelling - a call for European study (MCMDM-1 VWD). J Thromb Haemost. 2006;4:766–73.
harmonisation. Clin Chem Lab Med. 2016;54:1141–5. 29. Rodeghiero F, Castaman G, Tosetto A, Batlle J, Baudo F, Cappelletti A,
7. Rodeghiero F, Tosetto A, Abshire T, Arnold DM, Coller B, James P, Neunert Casana P, De Bosch N, Eikenboom JC, Federici AB, Lethagen S, Linari S,
C, Lillicrap D, ISj VWF. Perinatal/pediatric hemostasis subcommittees Srivastava A. The discriminant power of bleeding history for the diagnosis
working G. ISTH/SSC bleeding assessment tool: a standardized of type 1 von Willebrand disease: an international, multicenter study.
questionnaire and a proposal for a new bleeding score for inherited J Thromb Haemost. 2005;3:2619–26.
bleeding disorders. J Thromb Haemost. 2010;8:2063–5. 30. Federici AB, Bucciarelli P, Castaman G, Mazzucconi MG, Morfini M, Rocino A,
8. Harrison P, Mackie I, Mumford A, Briggs C, Liesner R, Winter M, Machin S. Schiavoni M, Peyvandi F, Rodeghiero F, Mannucci PM. The bleeding score
British committee for standards in H. Guidelines for the laboratory predicts clinical outcomes and replacement therapy in adults with von
investigation of heritable disorders of platelet function. Br J Haematol. Willebrand disease. Blood. 2014;123:4037–44.
2011;155:30–44. 31. O′Brien EC, Simon DN, Thomas LE, Hylek EM, Gersh BJ, Ansell JE, Kowey PR,
9. Bonhomme F, Ajzenberg N, Schved JF, Molliex S, Samama CM, French A. Mahaffey KW, Chang P, Fonarow GC, Pencina MJ, Piccini JP, Peterson ED.
Intensive care committee on evaluation of routine preoperative T, French The ORBIT bleeding score: a simple bedside score to assess bleeding risk in
society of a, intensive C. Pre-interventional haemostatic assessment: atrial fibrillation. Eur Heart J. 2015;36:3258–64.
guidelines from the French society of anaesthesia and intensive care. 32. Gresele P. Subcommittee on platelet physiology of the international society
Eur J Anaesthesiol. 2013;30:142–62. on T, Hemostasis Diagnosis of inherited platelet function disorders:
10. Tosetto A, Castaman G, Rodeghiero F. Bleeders, bleeding rates, and guidance from the SSC of the ISTH. J Thromb Haemost. 2015;13:314–22.
bleeding score. J Thromb Haemost. 2013;11 Suppl 1:142–50. 33. Gresele P, Harrison P, Bury L, Falcinelli E, Gachet C, Hayward CP, Kenny D,
11. Stepanian A, Biron-Andreani C. Primary hemostasis exploration. Ann Biol Mezzano D, Mumford AD, Nugent D, Nurden AT, Orsini S, Cattaneo M.
Clin (Paris). 2001;59:725–35. Diagnosis of suspected inherited platelet function disorders: results of a
12. Haute Autorité de Santé. Biologie des anomalies de l’hémostase. Service worldwide survey. J Thromb Haemost. 2014;12:1562–9.
évaluation des actes professionnels. 2010. V n°5 du 20/10/10. 34. Tosetto A, Castaman G, Plug I, Rodeghiero F, Eikenboom J. Prospective
13. O′Donnell J, Laffan MA. The relationship between ABO histo-blood group, evaluation of the clinical utility of quantitative bleeding severity
factor VIII and von Willebrand factor. Transfus Med. 2001;11:343–51. assessment in patients referred for hemostatic evaluation. J Thromb
14. Franchini M, Capra F, Targher G, Montagnana M, Lippi G. Relationship Haemost. 2011;9:1143–8.
between ABO blood group and von Willebrand factor levels: from biology 35. Polack B, Schved JF, Boneu B. Groupe d’Etude sur l’Hemostase et la T.
to clinical implications. Thromb J. 2007;5:14. Preanalytical recommendations of the ‘Groupe d'Etude sur l'Hemostase et la
15. Cattaneo M, Cerletti C, Harrison P, Hayward CP, Kenny D, Nugent D, Nurden Thrombose’ (GEHT) for venous blood testing in hemostasis laboratories.
P, Rao AK, Schmaier AH, Watson SP, Lussana F, Pugliano MT, Michelson AD. Haemostasis. 2001;31:61–8.
Recommendations for the standardization of light transmission 36. GFHT. (French Study Group on Hemostasis and Thrombosis). 2015. http://
aggregometry: a consensus of the working party from the platelet [Link]/site/Pratiques-Professionnelles/Documents-GEHT/Variables-
physiology subcommittee of SSC/ISTH. J Thromb Haemost. 2013;11:1183–9. Preanalytiques/Recommandations-Variables-preanalytiques_69_722.html.
16. Ozgonenel B, Rajpurkar M, Lusher JM. How do you treat bleeding disorders Accessed 3 Nov 2016.
with desmopressin? Postgrad Med J. 2007;83:159–63. 37. Mullier F, Bailly N, Chatelain C, Chatelain B, Dogne JM. Pre-analytical issues
17. Lipets EN, Ataullakhanov FI. Global assays of hemostasis in the diagnostics in the measurement of circulating microparticles: current recommendations
of hypercoagulation and evaluation of thrombosis risk. Thromb J. 2015;13:4. and pending questions. J Thromb Haemost. 2013;11:693–6.
18. Levine AB, Teppa J, McGough B, Cowchock FS. Evaluation of the 38. Lima-Oliveira G, Salvagno GL, Lippi G, Danese E, Gelati M, Montagnana M,
prethrombotic state in pregnancy and in women using oral contraceptives. Picheth G, Guidi GC. Could light meal jeopardize laboratory coagulation
Contraception. 1996;53:255–7. tests? Biochem Med (Zagreb). 2014;24:343–9.
19. Brenner B. Haemostatic changes in pregnancy. Thromb Res. 2004;114:409–14. 39. Rull G, Mohd-Zain ZN, Shiel J, Lundberg MH, Collier DJ, Johnston A, Warner
20. Franchi F, Biguzzi E, Martinelli I, Bucciarelli P, Palmucci C, D′Agostino S, TD, Corder R. Effects of high flavanol dark chocolate on cardiovascular
Peyvandi F. Normal reference ranges of antithrombin, protein C and protein function and platelet aggregation. Vascul Pharmacol. 2015;71:70–8.
S: effect of sex, age and hormonal status. Thromb Res. 2013;132:e152–7. 40. O'Donnell J, Tuddenham EG, Manning R, Kemball-Cook G, Johnson D, Laffan
21. Sandset PM. Mechanisms of hormonal therapy related thrombosis. Thromb M. High prevalence of elevated factor VIII levels in patients referred for
Res. 2013;131 Suppl 1:S4–7. thrombophilia screening: role of increased synthesis and relationship to the
22. Raps M, Helmerhorst FM, Fleischer K, Dahm AE, Rosendaal FR, Rosing J, acute phase reaction. Thromb Haemost. 1997;77:825–8.
Reitsma P, Sandset PM, van Vliet HA. The effect of different hormonal 41. Lippi G, Salvagno GL, Montagnana M, Lima-Oliveira G, Guidi GC, Favaloro EJ.
contraceptives on plasma levels of free protein S and free TFPI. Thromb Quality standards for sample collection in coagulation testing. Semin
Haemost. 2013;109:606–13. Thromb Hemost. 2012;38:565–75.
23. van Vliet HA, Bertina RM, Dahm AE, Rosendaal FR, Rosing J, Sandset PM, 42. Lee KW, Lip GY. Effects of lifestyle on hemostasis, fibrinolysis, and platelet
Helmerhorst FM. Different effects of oral contraceptives containing different reactivity: a systematic review. Arch Intern Med. 2003;163:2368–92.
 Magnette et al. Thrombosis Journal (2016) 14:49 Page 14 of 14

43. Chen YW, Chen JK, Wang JS. Strenuous exercise promotes shear-induced 67. Lance MD, Kuiper GJ, Sloep M, Spronk HM, van Oerle R, ten Cate H, Marcus
thrombin generation by increasing the shedding of procoagulant MA, Henskens YM. The effects of pneumatic tube system transport on
microparticles from platelets. Thromb Haemost. 2010;104:293–301. ROTEM analysis and contact activation assessed by thrombin generation
44. McCraw A, Hillarp A, Echenagucia M. Considerations in the laboratory test. Thromb Res. 2012;130:e147–50.
assessment of haemostasis. Haemophilia. 2010;16 Suppl 5:74–8. 68. Lance MD, Marcus MA, van Oerle R, Theunissen HM, Henskens YM. Platelet
45. Lippi G, Becan-McBride K, Behulova D, Bowen RA, Church S, Delanghe J, concentrate transport in pneumatic tube systems–does it work? Vox Sang.
Grankvist K, Kitchen S, Nybo M, Nauck M, Nikolac N, Palicka V, Plebani M, 2012;103:79–82.
Sandberg S, Simundic AM. Preanalytical quality improvement: in quality we 69. Daves M, Giacomuzzi K, Tagnin E, Jani E, Adcock Funk DM, Favaloro EJ,
trust. Clin Chem Lab Med. 2013;51:229–41. Lippi G. Influence of centrifuge brake on residual platelet count and
46. WHO Guidelines on Drawing Blood: Best Practices in Phlebotomy. Geneva; routine coagulation tests in citrated plasma. Blood Coagul Fibrinolysis.
2010. [Link] 2014;25:292–5.
[Link]?ua=1. 70. Nelson S, Pritt A, Marlar RA. Rapid preparation of plasma for ‘Stat’
47. CLSI GP41-A6 (replaces H03-A6). Procedures for the Collection of Diagnostic coagulation testing. Arch Pathol Lab Med. 1994;118:175–6.
Blood Specimens by Venipuncture; Approved Standard—Sixth Edition. Vol. 71. Pappas AA, Palmer SK, Meece D, Fink LM. Rapid preparation of plasma for
27 No. 26. coagulation testing. Arch Pathol Lab Med. 1991;115:816–7.
48. Lima-Oliveira G, Lippi G, Salvagno GL, Montagnana M, Picheth G, Guidi GC. 72. Lippi G, Franchini M, Montagnana M, Salvagno GL, Poli G, Guidi GC. Quality
The effective reduction of tourniquet application time after minor and reliability of routine coagulation testing: can we trust that sample?
modification of the CLSI H03-A6 blood collection procedure. Biochem Med Blood Coagul Fibrinolysis. 2006;17:513–9.
(Zagreb). 2013;23:308–15. 73. Lippi G, Rossi R, Ippolito L, Zobbi V, Azzi D, Pipitone S, Favaloro EJ, Funk
49. Mackie I, Cooper P, Lawrie A, Kitchen S, Gray E, Laffan M, British Committee DM. Influence of residual platelet count on routine coagulation, factor VIII,
for Standards in H. Guidelines on the laboratory aspects of assays used in and factor IX testing in postfreeze-thaw samples. Semin Thromb Hemost.
haemostasis and thrombosis. Int J Lab Hematol. 2013;35:1–13. 2013;39:834–9.
50. Riley RS, Tidwell AR, Williams D, Bode AP, Carr ME. Laboratory Evaluation of 74. Lippi G, Salvagno GL, Montagnana M, Manzato F, Guidi GC. Influence of the
Hemostasis. [Link] centrifuge time of primary plasma tubes on routine coagulation testing.
[Link]. Blood Coagul Fibrinolysis. 2007;18:525–8.
51. Lippi G, Ippolito L, Zobbi V, Sandei F, Favaloro EJ. Sample collection and 75. Boudaoud L, Divaret G, Marie P, Bezeaud A. Rapid centrifugation for routine
platelet function testing: influence of vacuum or aspiration principle on coagulation testing. Ann Biol Clin (Paris). 2006;64:315–7.
PFA-100 test results. Blood Coagul Fibrinolysis. 2013;24:666–9. 76. Sultan A. Five-minute preparation of platelet-poor plasma for routine
52. Becton, Dickinson and Company (BD) Diagnostics : [Link] coagulation testing. East Mediterr Health J. 2010;16:233–6.
Accessed 17 Nov 2016. 77. Ageno W, Buller HR, Falanga A, Hacke W, Hendriks J, Lobban T, Merino J,
53. Lippi G, Salvagno GL, Adcock DM, Gelati M, Guidi GC, Favaloro EJ. Right or Milojevic IS, Moya F, van der Worp HB, Randall G, Tsioufis K, Verhamme P,
wrong sample received for coagulation testing? tentative algorithms for Camm AJ. Managing reversal of direct oral anticoagulants in emergency
detection of an incorrect type of sample. Int J Lab Hematol. 2010;32:132–8. situations. Anticoagulation Education Task Force White Paper. Thromb
54. Smock KJ, Crist RA, Hansen SJ, Rodgers GM, Lehman CM. Discard tubes are Haemost. 2016;116:1003–10.
not necessary when drawing samples for specialized coagulation testing. 78. Pollack Jr CV, Reilly PA, Eikelboom J, Glund S, Verhamme P, Bernstein RA,
Blood Coagul Fibrinolysis. 2010;21:279–82. Dubiel R, Huisman MV, Hylek EM, Kamphuisen PW, Kreuzer J, Levy JH,
55. Raijmakers MT, Menting CH, Vader HL, van der Graaf F. Collection of blood Sellke FW, Stangier J, Steiner T, Wang B, Kam CW, Weitz JI. Idarucizumab
specimens by venipuncture for plasma-based coagulation assays: necessity for dabigatran reversal. N Engl J Med. 2015;373:511–20.
of a discard tube. Am J Clin Pathol. 2010;133:331–5. 79. Suchsland J, Friedrich N, Grotevendt A, Kallner A, Ludemann J, Nauck M,
56. Adcock DM, Kressin DC, Marlar RA. Minimum specimen volume Petersmann A. Optimizing centrifugation of coagulation samples in
requirements for routine coagulation testing: dependence on citrate laboratory automation. Clin Chem Lab Med. 2014;52:1187–91.
concentration. Am J Clin Pathol. 1998;109:595–9. 80. Mayo Medical Laboratories. Ensuring Specimen Integrity: Proper Processing
57. CLSI H21-A5: Collection, Transport, and Processing of Blood Specimens for and Handling of Specimens for Coagulation Testing. [Link]
Testing Plasma-Based Coagulation Assays and Molecular Hemostasis Assays; [Link]/articles/communique/2008/[Link].
Approved Guideline—Fifth Edition. Vol. 28 No. 5. 81. Salvagno GL, Lippi G, Montagnana M, Franchini M, Poli G, Guidi GC.
Influence of temperature and time before centrifugation of specimens for
58. Ernst DJ, Ernst C. Phlebotomy tools of the trade: part 4: proper handling
routine coagulation testing. Int J Lab Hematol. 2009;31:462–7.
and storage of blood specimens. Home Healthc Nurse. 2003;21:266–70.
82. Bohm M, Taschner S, Kretzschmar E, Gerlach R, Favaloro EJ, Scharrer I. Cold
59. Arora S, Kolte S, Dhupia J. Hemolyzed samples should be processed for
storage of citrated whole blood induces drastic time-dependent losses in
coagulation studies: the study of hemolysis effects on coagulation
factor VIII and von Willebrand factor: potential for misdiagnosis of
parameters. Ann Med Health Sci Res. 2014;4:233–7.
haemophilia and von Willebrand disease. Blood Coagul Fibrinolysis.
60. Favaloro EJ, Adcock Funk DM, Lippi G. Pre-analytical variables in coagulation
2006;17:39–45.
testing associated with diagnostic errors in hemostasis. Lab Med. 2012;43:1–10.
83. Kottke-Marchant K, David B. Laboratory hematology practice. Wiley-
61. Lippi G, Plebani M, Favaloro EJ. Interference in coagulation testing: focus on
Blackwell. 2012;31:416–7.
spurious hemolysis, icterus, and lipemia. Semin Thromb Hemost. 2013;39:258–66.
84. Zurcher M, Sulzer I, Barizzi G, Lammle B, Alberio L. Stability of coagulation
62. Lacroix R, Judicone C, Poncelet P, Robert S, Arnaud L, Sampol J, Dignat-
assays performed in plasma from citrated whole blood transported at
George F. Impact of pre-analytical parameters on the measurement of
ambient temperature. Thromb Haemost. 2008;99:416–26.
circulating microparticles: towards standardization of protocol. J Thromb
85. Rao LV, Okorodudu AO, Petersen JR, Elghetany MT. Stability of prothrombin
Haemost. 2012;10:437–46.
time and activated partial thromboplastin time tests under different storage
63. Lacroix R, Judicone C, Mooberry M, Boucekine M, Key NS, Dignat-George F,
conditions. Clin Chim Acta. 2000;300:13–21.
The ISSCW. Standardization of pre-analytical variables in plasma
86. Alesci S, Borggrefe M, Dempfle CE. Effect of freezing method and storage at
microparticle determination: results of the international society on −20° C and −70° C on prothrombin time, aPTT and plasma fibrinogen
thrombosis and haemostasis SSC collaborative workshop. J Thromb
levels. Thromb Res. 2009;124:121–6.
Haemost. 2013;11:1190–3.
87. Bach J, Haubelt H, Hellstern P. Sources of variation in factor VIII, von
64. Adcock Funk DM, Lippi G, Favaloro EJ. Quality standards for sample Willebrand factor and fibrinogen measurements: implications for detecting
processing, transportation, and storage in hemostasis testing. Semin deficiencies and increased plasma levels. Thromb Res. 2010;126:e188–95.
Thromb Hemost. 2012;38:576–85. 88. Marlar RA, Gausman JN. Laboratory testing issues for protein C, protein S,
65. Bolliger D, Seeberger MD, Tanaka KA, Dell-Kuster S, Gregor M, Zenklusen U, and antithrombin. Int J Lab Hematol. 2014;36:289–95.
Grapow M, Tsakiris DA, Filipovic M. Pre-analytical effects of pneumatic tube 89. A Practical Guide to Laboratory Haemostasis: [Link]
transport on impedance platelet aggregometry. Platelets. 2009;20:458–65. [Link]/[Link]. Accessed 17 Nov 2016.
66. Lance MD, Henskens YM. Effect of pneumatic tube transport on rotational
thromboelastometry. Br J Anaesth. 2013;110:142.

View publication stats