Downloaded from http://jcp.bmj.com/ on June 30, 2015 - Published by group.bmj.

com

Original article

Platelet function testing: practice among UK National

External Quality Assessment Scheme for Blood
Coagulation participants, 2006
I Jennings, T A L Woods, S Kitchen, I D Walker

UK National External Quality ABSTRACT evidence largely based on personal experience.3 4

Assessment Scheme (UK Aims: Platelet function testing forms an important part of The Clinical and Laboratory Standards Institute
NEQAS) for Blood Coagulation,
the laboratory investigation of a bleeding tendency; has published guidelines on performance of the
Sheffield, UK
however, little standardisation and quality control is Bleeding Time5 and guidelines on aggregometry are
Correspondence to: available for these tests. A UK National External Quality under construction at the time of writing. The
Dr I Jennings, UK NEQAS for Assessment Scheme (UK NEQAS) for Blood Coagulation most recent UK guidelines were published in 1988,6
Blood Coagulation, Rutledge although more recent guidelines on the manage-
Mews, 3 Southbourne Road, exercise sought to identify current practice among
Sheffield S10 2QN, UK; i. laboratories performing platelet function tests. ment of platelet function disorders give some
jennings@coageqa.org.uk Methods: A questionnaire was circulated in March 2006 recommendations on the investigation of such
to establish the current status of platelet function testing disorders.7
Accepted 14 May 2008 practice among participants of UK NEQAS. Participants
were asked specifically about practice in bleeding time METHODS
testing, PFA-100 analyser use, platelet aggregometry The UK National External Quality Assessment
methodology and additional tests of platelet function. Scheme (UK NEQAS) is a proficiency testing
Results: 169 returned questionnaires revealed that 26 organisation dealing with all aspects of coagulation
centres used bleeding time, the PFA-100 analyser and laboratory testing. Since distribution of test
platelet aggregometry in their investigations; 13 used samples for platelet function testing is not
bleeding time and the PFA-100 only; 33 used bleeding currently possible, in advance of the planned new
time and platelet aggregometry; and 23 used the PFA-100 guidelines we circulated a questionnaire in March
with platelet aggregometry. 58 centres reported that they 2006 to establish the current status of platelet
performed only bleeding times in their investigations, 10 function testing practice among participants.
reported use of the PFA-100 only, and 6 reported use of Participants were asked specifically about practice
aggregometry only. Marked variability was observed in in bleeding time testing, use of the PFA-100 platelet
methodology for each of these tests, and in many cases function analyser (PFA) (Dade Behring, Marburg,
no form of quality control was employed. Germany), platelet aggregometry methodology and
Conclusions: The data confirmed the lack of standardi- additional tests of platelet function (such as
sation in methodology employed in different centres. nucleotide and glycoprotein assessment). All
Updated guidelines and standardisation of platelet responses were coded to maintain participant
function assessment are required to facilitate compar- confidentiality, and data were analysed using
ability between centres. Microsoft excel spreadsheet software.

Platelets form a pivotal role in the haemostatic RESULTS

system, and defects of platelet function (congeni- Responses with respect to platelet function testing
tal, such as Bernard Soulier syndrome and were received from 169 centres, of which 119 were
Glanzmann thrombasthenia, and acquired) are UK centres and 50 were outside the UK.
associated with severe bleeding defects. Diagnosis
of a platelet function disorder is often made by a Tests used in the investigation of platelet function
battery of tests including ex vivo tests (bleeding Table 1 shows the tests used by participating
time) and in vitro tests such as platelet aggrego- centres, and fig 1 shows the combinations of tests
metry. employed. Twenty-six centres used bleeding time,
Despite the importance of platelet function the PFA-100 analyser and platelet aggregometry in
testing in the investigation of a bleeding tendency, their investigations, 13 used bleeding time and the
there is little standardisation and no effective PFA-100 only, 33 used bleeding time and platelet
quality control of these tests. A survey of bleeding aggregometry, and 23 used the PFA-100 and
time methodology in 1984 identified wide varia- platelet aggregometry. Fifty-eight centres reported
tion in methodology used.1 More recently, varia- that they performed only bleeding times in their
bility in testing for platelet aggregometry investigations, 10 reported use of the PFA-100 only,
responses, dense granule content analysis and and six reported use of aggregometry only.
platelet secretion defects has been reported among Ninety-seven centres reported that they
North American laboratories.2 excluded patients from investigation on the basis
Few guidelines have been published on platelet of recent drug history (aspirin, non-steroidal anti-
function testing. There are some reports with inflammatory drugs, clopidogrel—time frame of
recommendations in the absence of high grade within 1–4 weeks prior to testing). Fifty-two

950 J Clin Pathol 2008;61:950–954. doi:10.1136/jcp.2008.057174

 Downloaded from http://jcp.bmj.com/ on June 30, 2015 - Published by group.bmj.com

Original article

Table 1 Number of centres performing each test, and approximate number of tests performed
Number of laboratories
performing tests Approximate number of tests per year
On
On all selected
patients patients Total* ,10 10–20 21–50 51–100 .100 .1000

Bleeding time 48 56 130 34 16 5 2 8 1

PFA-100 analyser 44 28 72 0 8 6 13 20 4
Platelet aggregation 25 57 88 8 18 18 9 8 –
Platelet nucleotides 5 25 30 3 8 3 5 3 –
Platelet glycoproteins 0 23 23 6 4 4 – – –
Serotonin/5HT release 0 2 2 – – – – – –
*Including some centres that did not indicate whether patient selection was carried out.
5HT, 5-hydroxytryptamine.

centres indicated they did not exclude any patients from The source and range of reported reference values for the
investigation on these grounds. bleeding time are shown in table 2. Approximately 15% of
centres had established a locally derived reference range.
Bleeding time
A total of 130 centres reported that they performed bleeding PFA testing
times; these were requested exclusively by haematology The PFA-100 was widely used among UK laboratories in the
clinicians in 105 centres, and by haematology or other investigation of platelet function; 72 centres reported PFA-100
hospital-based clinicians in 23 centres; requests could also be measurement as part of their platelet function screening
generated by general practitioners in four centres. In two procedure. A total of 61% of centres performed PFA-100 analysis
centres bleeding time requests were generated by biomedical or on all patients investigated for platelet function defects, 39% of
clinical scientists, and in two centres by specialist nursing staff. centres performed the test on selected cases only. Fourteen
centres reported that they did not carry out further tests if the
The most widely used devices for measuring bleeding time
PFA-100 analysis was normal, 48 sometimes carried out further
were Simplate (Organon Teknika Corporation, Durham, North
analysis, and nine always proceeded with other investigations.
Carolina, US; n = 54), Surgicutt (ITC, Piscataway, New Jersey,
The source and range of reported reference values employed
US; n = 38) and Triplett (Helena Laboratories, Beaumont,
with the PFA-100 analyser are shown in table 3. Approximately
Texas, US; n = 21). Paediatric Simplate devices (n = 1) and
58% of centres had established a locally derived reference range.
Lancets (n = 2) were also employed. One centre reported use of
a Duke’s (ear-lobe) bleeding time.
A total of 106/111 (95%) of centres reported that venous Platelet aggregometry
pressure was maintained at 40 mm Hg when performing a A total of 88 centres performed platelet aggregometry investiga-
bleeding time, one centre reported 20 mm Hg, one reported an tions.
age-dependent range between 20 and 40 mm Hg, one reported
80 mm Hg, and one reported performing a bleeding time with Sample preparation
venous pressure at 140 mm Hg. All centres performed aggregometry on samples collected into
citrate anticoagulant. Of these, four reported using whole blood
aggregometry. Forty-seven centres reported use of samples
collected into 3.2% or 0.106–0.109 mM citrate, four centres
employed 3.8% citrate; the remainder did not state the citrate
concentration employed. Sixty-two (70%) applied maximum
and minimum time limits between which aggregation must be
performed. The minimum time limit before aggregometry could
be performed ranged from 10 to 90 min (median 30 min) and
the maximum limit applied ranged from 45 to 240 min (median
180 min).
Platelet-rich plasma (PRP) was prepared at speeds ranging
from 75 to 210 g, for between 5 and 15 min. The platelet count
was checked by 79/82 centres and adjusted by 73/79, of which
55 indicated that platelet-poor plasma (PPP) was used to dilute

Table 2 Source and range of reference ranges reported for the bleeding
time
Device Source n Median (min) Range (min)

Simplate Literature 36 9.5 6–10.5

In-house 5 10 8–10
Surgicutt Literature 22 9.5 7–10
Figure 1 Patterns of tests employed in platelet function investigations,
In-house 8 9 7–11
showing numbers of centres performing each combination of tests. PFA,
Triplett Literature 17 9.5 6–10
platelet function analyser.

J Clin Pathol 2008;61:950–954. doi:10.1136/jcp.2008.057174 951

 Downloaded from http://jcp.bmj.com/ on June 30, 2015 - Published by group.bmj.com

Original article

Table 3 Source and range of reference ranges reported for the platelet function analyser
Test performed Test performed Normal range, Normal range Normal range
on all on selected Normal range lower value upper value upper value
patients (n) patients (n) source range (s) range (s) median (s)

Collagen/epinephrine 65 4 In-house 31 59–102 117–185 164

Literature 22 78–96 142–199 164
Collagen/ADP 44 23 In-house 31 42–84 100–146 118
Literature 22 55–71 100–137 113

the PRP. Eighteen centres did not indicate how the dilution was Epinephrine
effected. PRP was most frequently adjusted to a count of Fourteen different sources of epinephrine were reported; only
2006109/l, or to a range of 200–3006109/l. PPP was then the epinephrine supplied by Biodata (n = 17) was used by more
prepared by centrifugation between 1200 and 4000 g, for than 10 centres. Reported concentrations ranged between 1 and
between 5 and .20 min. 1000 mm, with 8/60 centres reporting concentrations in mg/ml.
Four centres stored platelets at 37uC prior to testing, and all Only 11/60 used more than one dilution of epinephrine.
other centres stored platelets at ambient temperature (18–
25uC). Arachidonic acid
Fewer sources of arachidonic acid were reported (7), of which
Agonists employed in platelet aggregometry two are used by .10 centres (Biodata, n = 28; Helena
Table 4 shows the most frequently employed agonists and the Laboratories, n = 18). Greater conformity in concentrations
patterns of use amongst participants. Other agonists employed was also noted, with 31/59 centres using a single concentration
were thrombin (n = 1), tartrate-resistant acid phosphatase of 0.5 mM in their investigation. A further 12 centres used a
(n = 1) and cryoprecipitate (n = 1). Spontaneous aggregation single concentration of 1 mM.
was assessed by two centres.
The range of sources and concentrations of each agonist Collagen
reported by participants was substantial. These are summarised Nine different sources of collagen were reported. Seventeen
below. reported use of bovine source of collagen, 23 reported an equine
source, and four reported a human source. However, some
ADP confusion was evident in the source of collagen, as some centres
Ten different sources of ADP were reported; the most widely reported collagen obtained from different sources but from the
employed were from BioData (Alpha Laboratories, Eastleigh, same manufacturer. Final concentrations varied markedly, but
UK) (n = 23), Helena Laboratories (n = 18), and Sigma, Poole, were related to the source of reagent—centres using equine
UK (n = 14). Within these groups, up to 12 different combina- collagen from Helena Laboratories employed concentrations
tions of ADP concentrations were reported, from a range of from 0.2 to 10 mg/ml; centres using Biodata bovine collagen
0.16–5 mm employed by one participant to 20–40 mm employed reported concentrations between 190 and 1900 mg/ml.
by another. A total of 28 of 76 centres used a single
concentration of ADP, the most frequent of which was 20 mM. Measurement of aggregation response
Participants were asked for the method by which aggregation
Ristocetin responses were reported; fig 2 shows the distribution of
Seventeen different sources of ristocetin were reported; the responses. Normal ranges employed by centres for quantitative
most widely employed were from Biodata (n = 10), Helena aggregometry were established from manufacturers’ informa-
Laboratories (n = 22), and ABP, Marlton, New Jersy, USA tion (n = 14 centres), peer-reviewed literature (n = 8), or in-
(n = 12). Within these groups, up to 15 different combinations house evaluation (n = 23), of which 11 centres reported their
of ristocetin concentrations were reported. A single concentra- observed range of responses and 12 employed the mean (SD) of
tion of ristocetin was used by 40/77 centres, and the most percentage aggregation to establish reference ranges. Fifty-nine
frequent concentration was 1.5 mg/ml. The lowest concentra- centres employed internal quality controls when performing
aggregometry; these controls were obtained predominantly
tion employed was 0.15 mg/ml, and the highest was 3 mg/ml.
from laboratory staff, and sometimes age- and sex-matched to
Twenty-seven centres employed two or more dilutions of
the patient under investigation. Thirteen centres indicated they
ristocetin, the most frequent being a range between 0.5 and
tested a control with every patient, and two reported they only
1.5 mg/ml.
did so if a patient result was abnormal.

Table 4 Pattern of agonists used in platelet aggregometry

investigations Further tests of platelet function
Forty-one centres (24%) reported use of additional tests for
Arachidonic
ADP Ristocetin Epinephrine acid Collagen n platelet function. Thirty centres assessed platelet nucleotide
levels; 14 of these centres reported methodology: 12 performed
! ! ! ! ! 49
chemiluminescence assays, and two used high-performance liquid
! ! ! ! 10
chromatography. The number of tests performed annually in
! ! ! ! 10
different departments ranged from 1 to 300, with five centres
! ! ! 7
measuring nucleotide levels on all investigations of platelet
! ! ! 3
! ! ! ! 1
function. Two centres measured serotonin uptake and release,
and both centres used radiolabelled 5-hydroxytryptamine.

952 J Clin Pathol 2008;61:950–954. doi:10.1136/jcp.2008.057174

 Downloaded from http://jcp.bmj.com/ on June 30, 2015 - Published by group.bmj.com

Original article

investigations. Haubelt et al12 reported a number of pre- and

peranalytical variables contributing to variation in closure times
in normal individuals, and recommended establishment of local
reference ranges and duplicate measurements. In our study, 42%
reported literature as the source of reference range. Interestingly,
upper limits of normal reported by these participants ranged from
142 to 199 s for collagen/epinephrine cartridges. A similar wide
range was reported by centres using locally determined reference
ranges (117–185 s). Although batch-to-batch variation has been
reported, such marked variation is of concern. Lack of available
external quality assessment material and limitation of internal
quality control (one-third of centres did not employ any control,
the remainder performing normal controls only) emphasise the
need for very careful attention to pre- and peranalytical variables
when carrying out this test.
Platelet aggregometry has been employed to investigate
platelet function defects for over 40 years.13 Optimal aggrega-
tion responses to agonists are achieved when pre- and
Figure 2 Methods employed to interpret platelet aggregometry results, peranalytical variables are carefully controlled. All centres in
showing numbers of centres employing each method. AUC, area under
this study employed citrate as an anticoagulant, with the large
curve; ED50, concentration threshold to elicit 50% aggregation; MA,
maximum aggregation response. majority using 3.2% (0.105–0.109 M) trisodium citrate, in
keeping with usage of this concentration in routine haemostatic
investigations in most UK laboratories. Methodology for
Twenty-three centres measured concentration of platelet glyco-
preparation of PRP varied between centres, with a threefold
proteins; all centres reporting methodology employed flow
difference in centrifugation speeds and times reported. All but
cytometry methods. Four centres measured platelet factor 4 and
three centres checked the platelet count and the majority
b-thromboglobulin, and all four used ELISA. One centre reported
adjusted the count with autologous PPP. Some authors do not
assessment of platelet procoagulant activity using flow cytometry
consider it necessary to adjust the platelet count;3 7 Cattaneo et
and annexin V.
al have reported inhibition of aggregation when autologous PPP
is employed to adjust the platelet count,14 and more recently
DISCUSSION Linneman et al15 confirmed decreased aggregation responses
Laboratory tests are of major importance in the evaluation of when platelet counts were adjusted in this way. Aggregometry
platelet function and diagnosis of platelet disorders. However, responses are not much altered between platelet counts of
little progress has been made in the standardisation and 1506109/l to 4506109/l, although the majority of centres in this
performance of these tests, despite comments reported by the study adjust counts to between 2006109/l and 3006109/l.
haemostasis task force of the British Committee for Standards Recommendations on whether adjustment of platelet counts is
in Haematology in 1988.6 The bleeding time has been in use for required will be welcomed.
almost 100 years since its description by Duke8 and remains the Four centres stored samples at 37uC prior to testing; this has
only practical evaluation of ex vivo haemostatic response. been reported to reduce the maximum amplitude of some agonist
However, standardisation of this relatively straightforward test responses, such as that of collagen.3 The time window for testing
seems hard to achieve. Poller1 reported marked variability in is critical for platelet aggregation studies. A reduced aggregation
methodology, and the study reported here confirms a lack of pattern occurs immediately after blood sampling and changes in
standardised methodology today. Sphygmomanometer pres- pH on storage contribute to a significant reduction in agonist
sures between 20 and 140 mm Hg, and reference ranges varying response in platelets stored for .3–4 h, leading to recommenda-
between 6 and 10 min, underline the variability compounded by tion to test within 2 h.3 6 It is of concern that two centres applied
the lack of any available quality control measures, although the no time window, 22 applied an upper time limit only, and of those
possibility of transcription error by the centre reporting 140 mm that did apply minimum and maximum time limits, some
Hg cannot be excluded. The bleeding time is an invasive and reported limits outside those recommended.
subjective investigation that has been reported to be of low There are no recent recommendations for the agonists that
sensitivity and specificity in the detection of mild bleeding should be used in aggregation studies; many authors merely
disorders.9 It is of concern that a large proportion of centres report the agonists commonly used. 2 4 7 Recommendations on a
responding to this questionnaire employed the bleeding time as minimum panel of agonists would be welcome. The large range
their only test of platelet function. Bolton-Maggs et al7 in agonist concentrations reported by participants was in
recommend the bleeding time is ‘‘...considered in special keeping with the findings of Moffat et al2 in the Nascola
circumstances…if other tests have not demonstrated a defect’’. laboratories, indicating a lack of standardisation between
The PFA-100 has been available to clinical laboratories for reagents and between laboratories, and possibly an inability
several years, and among UK NEQAS participants it is now by some laboratories to correctly determine final concentrations
employed in ,42% of centres performing platelet function of reagents. The smallest range of agonist concentrations used
analysis. The PFA-100 is not considered to be of sufficient was for arachidonic acid and for ristocetin-induced aggregation,
sensitivity or specificity to be used as a screening tool in the latter possibly reflecting the depth of literature on von
determining which individuals require further testing for Willebrand factor investigation, though even here there was a
inherited platelet disorders, although some severe defects can 10-fold range of concentrations in reported use. Interpretation
be excluded with a negative screen.10 11 However, 10 centres of results was also variable. Forty-four per cent of centres
reported using this test exclusively in their platelet function reported that they used only a qualitative assessment of

J Clin Pathol 2008;61:950–954. doi:10.1136/jcp.2008.057174 953

 Downloaded from http://jcp.bmj.com/ on June 30, 2015 - Published by group.bmj.com

Original article

aggregation response (compared with only 1/47 in the Nascola Competing interests: None.
study), which is of necessity subjective, and may fail to detect mild
but significant reductions in aggregation response. Fifty-six per REFERENCES
cent of centres determined maximal aggregation, of which almost 1. Poller L. The bleeding time, current practice in the UK. Clin Lab Haematol
1984;6:369–73.
half also quantified further parameters (slope, lag phase, area under
2. Moffat KA, Ledford-Kraemer MR, Nichols WL, et al. Variability in clinical laboratory
curve). Fifty per cent of centres determined a local reference range practice in testing for disorders of platelet function: results of two surveys of the
for maximal aggregation; several authors have recommended this North American Specialized Coagulation Laboratory Association. Thromb Haemost
approach.2 7 10 However, the merits of using single concentrations 2005;93:549–53
3. Breddin HK. Can platelet aggregometry be standardized? Platelets 2005;16:151–8.
of agonists and comparing maximum aggregation to a reference 4. Zhou L, Schmaier AH. Platelet aggregation testing in platelet-rich plasma. Description
range compared with determining agonist dose responses requires of procedures with the aim to develop standards in the field. Am J Clin Pathol
further consideration. Hayward and Eikelboom10 have emphasised 2005;123:172–83.
5. The Clinical and Laboratory Standards Institute. http://www.clsi.org (accessed
the difficulty in adopting reference ranges based on mean (SD) 27 June 2008).
(used by over half of UK NEQAS participants) as responses may 6. The British Society for Haematology BCSH Haemostasis and Thrombosis
not be normally distributed. Task Force. Guidelines on platelet function testing. J Clin Pathol 1988;41:1322–30.
7. Bolton-Maggs PH, Chalmers EA, Collins PW, et al. UKHCDO. A review of inherited
Finally, less than a quarter of centres reported performing platelet disorders with guidelines for their management on behalf of the UKHCDO.
nucleotide measurement, despite reports that aggregation Br J Haematol 2006;135:603–33.
studies may not detect these defects.16 Our survey has indicated 8. Duke WW. The relation of blood platelets to hemorrhagic disease. Description of a
method for determining the bleeding time and coagulation time and report of three
the wide variability of current practice and the need for updated cases of hemorrhagic disease relieved by transfusion JAMA 1910;55:1185.
recommendations and guidelines in this difficult area of 9. Podda GM, Bucciarelli P, Lussana F, et al. Usefulness of PFA-100 testing in the
haemostasis testing. diagnostic screening of patients with suspected abnormalities of hemostasis:
comparison with the bleeding time. J Thromb Haemost 2007;5:2393–8.
10. Hayward CPM, Eikelboom J. Platelet function testing, quality assurance. Semin
Thromb Hemost 2006;33:273–82.
11. Hayward CP, Harrison P, Cattaneo M, et al; The Platelet Physiology Subcommittee of
Take-home messages the Scientific and Standardization Committee of the International Society on Thrombosis
and Haemostasis. Platelet function analyzer (PFA)-100 closure time in the evaluation of
platelet disorders and platelet function. J Thromb Haemost 2006;4:312–9.
c External quality assessment is not available for tests of 12. Haubelt H, Anders C, Vogt A, et al. Variables influencing Platelet Function Analyzer-
platelet function, and internal quality control is not widely 100 closure times in healthy individuals. Br J Haematol 2005;130:759–67.
employed. 13. Born GVR, Cross MJ. The aggregation of blood platelets. J Physiol 1963;168:178–95.
14. Cattaneo M, Lecchi A, Zighetti ML, et al. Platelet aggregation studies: autologous
c Many centres employed screening tests to exclude further platelet-poor plasma inhibits platelet aggregation when added to platelet-rich plasma
investigation despite reports indicating that these methods are to normalize platelet count. Haematologica 2007;92:694–7.
not sensitive to all platelet function defects. 15. Linnemann B, Schwonberg J, Mani H, et al. Standardization of light transmittance
aggregometry for monitoring antiplatelet therapy: an adjustment for platelet count is
c Updated guidelines and recommendations are required to not necessary. J Thromb Haemost 2008;6:677–83.
improve standardisation and between-centre comparability. 16. Nieuwenhuis HK, Akkerman JW, Sixma JJ. Patients with a prolonged bleeding time
and normal aggregation tests may have storage pool deficiency: studies on one
hundred six patients. Blood 1987;70:620–3.

954 J Clin Pathol 2008;61:950–954. doi:10.1136/jcp.2008.057174

 Downloaded from http://jcp.bmj.com/ on June 30, 2015 - Published by group.bmj.com

Platelet function testing: practice among UK

National External Quality Assessment
Scheme for Blood Coagulation participants,
2006
I Jennings, T A L Woods, S Kitchen and I D Walker

J Clin Pathol 2008 61: 950-954

doi: 10.1136/jcp.2008.057174

Updated information and services can be found at:

http://jcp.bmj.com/content/61/8/950

These include:

References This article cites 15 articles, 3 of which you can access for free at:
http://jcp.bmj.com/content/61/8/950#BIBL

Email alerting Receive free email alerts when new articles cite this article. Sign up in the
service box at the top right corner of the online article.

Notes

To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions

To order reprints go to:

http://journals.bmj.com/cgi/reprintform

To subscribe to BMJ go to:

http://group.bmj.com/subscribe/