_q Association of Clinical Pathologists Broadsheet 122: August 1989 Platelet function testing Blood platelets contribute to normal haemostasis in two main ways. They adhere to subendothelial microfibrils and collagen in the blood vessel wall, after which they change shape, undergo a specific release reaction, and aggregate with each other to form a primary haemostatic plug. As a consequence of these events, particularly during release and aggregation, procoagulant activities are generated, chiefly involv- ing the platelet membrane phospholipids (platelet factor 3) so that blood coagulation is initiated at, and is largely localised to, the areas of platelet aggregation. Quantitative or qualitative platelet disorders may result in a pronounced bleeding tendency, mainly due to failure of platelet plug formation, but also to a lesser extent to the suboptimal activation of blood coagula- tion. This Broadsheet sets out to describe the methods which can be used reliably to detect and classify such bleeding disorders. Indications for platelet function tests Defective platelet function or thrombocytopenia may present clinically with a variety of symptoms indicative of primary haemostatic failure, such as bruises or ecchymoses, epistaxis, gastrointestinal haemorrhage or menorrhagia. Platelet defects usually give rise to a fairly mild bleeding disorder, and patients may present with excessive bleeding only after surgery or dental extractions. Some, however, may have severe haemorrhagic symptoms early in life. In patients with mild symptoms it is usually impossible to be certain from the history alone whether the patient has a platelet or a coagulation defect. The initial investiga- tion should include the following screening tests: blood film, platelet count, bleeding time, prothrombin time, activated partial thromboplastin time and assays for fibrinogen, factor VIII:C, von Willebrand factor antigen (vWF:Ag) and ristocetin cofactor activity. Only when a coagulation abnormality and von Willebrand's disease have been excluded should platelet function be studied further. A suggested scheme for the investigation of platelet function is shown in the figure. Accepted for publication 31 March 1989 Because many commonly used drugs also affect platelet function, these should also be excluded, or, if this is impossible for clinical reasons, their influence should be taken into account when interpreting the results. BLOOD FILM Inspection of a well spread, well stained film is often most helpful and will allow an estimate of the platelet count and will show any gross morphological abnor- malities in the platelets themselves. The presence of large platelets may be indicative of a congenital thrombopathy, such as Bernard-Soulier syndrome. The blood film also enables abnormalities in other cell lines to be observed, such as Dohle bodies in the granulocytes in the May-Hegglin anomaly. Perhaps the most important reason to inspect the blood film is to exclude another underlying haematological disor- der-leukaemia, for example, in which platelet func- tion may be impaired. It also allows other medical conditions, which may result in an acquired bleeding disorder, to be detected-liver, renal, or myeloproliferative diseases. THE BLEEDING TIME Of the many methods described for performing this useful in vivo test, the most sensitive, reproducible, and clinically valid is the template technique' in which the bleeding time of one or more small cuts is determined under highly standardised conditions. Technique (a) Taking care to avoid superficial veins, scars, or bruises, a small area of skin on the lateral aspect of the volar surface of the forearm about two inches below the elbow joint, is cleansed with alcohol (and, if necessary, shaved), and allowed to air dry. (b) A sphygmomanometer cuff is placed around the upper arm and kept inflated to a pressure of 40 mm Hg throughout the test, to standardise the intra-capillary pressure. (c) A template device, preferably a semiautomatic, spring-loaded blade device such as the Simplate II (General Diagnostics), is placed firmly, but without excessive pressure, on the skin in the longitudinal 858 copyright. on 17 August 2018 by guest. Protected by http://jcp.bmj.com/ J Clin Pathol: first published as 10.1136/jcp.42.8.858 on 1 August 1989. Downloaded from
Transcript
_q Association of Clinical Pathologists
Broadsheet 122: August 1989
Platelet function testing
Blood platelets contribute to normal haemostasis intwo main ways. They adhere to subendothelialmicrofibrils and collagen in the blood vessel wall, afterwhich they change shape, undergo a specific releasereaction, and aggregate with each other to form aprimary haemostatic plug. As a consequence of theseevents, particularly during release and aggregation,procoagulant activities are generated, chiefly involv-ing the platelet membrane phospholipids (plateletfactor 3) so that blood coagulation is initiated at, andis largely localised to, the areas ofplatelet aggregation.
Quantitative or qualitative platelet disorders mayresult in a pronounced bleeding tendency, mainly dueto failure of platelet plug formation, but also to a lesserextent to the suboptimal activation of blood coagula-tion. This Broadsheet sets out to describe the methodswhich can be used reliably to detect and classify suchbleeding disorders.
Indications for platelet function tests
Defective platelet function or thrombocytopenia maypresent clinically with a variety of symptomsindicative of primary haemostatic failure, such asbruises or ecchymoses, epistaxis, gastrointestinalhaemorrhage or menorrhagia. Platelet defects usuallygive rise to a fairly mild bleeding disorder, and patientsmay present with excessive bleeding only after surgeryor dental extractions. Some, however, may have severehaemorrhagic symptoms early in life. In patients withmild symptoms it is usually impossible to be certainfrom the history alone whether the patient has aplatelet or a coagulation defect. The initial investiga-tion should include the following screening tests:blood film, platelet count, bleeding time, prothrombintime, activated partial thromboplastin time and assaysfor fibrinogen, factor VIII:C, von Willebrand factorantigen (vWF:Ag) and ristocetin cofactor activity.Only when a coagulation abnormality and vonWillebrand's disease have been excluded shouldplatelet function be studied further. A suggestedscheme for the investigation of platelet function isshown in the figure.
Accepted for publication 31 March 1989
Because many commonly used drugs also affectplatelet function, these should also be excluded, or, ifthis is impossible for clinical reasons, their influenceshould be taken into account when interpreting theresults.
BLOOD FILMInspection of a well spread, well stained film is oftenmost helpful and will allow an estimate of the plateletcount and will show any gross morphological abnor-malities in the platelets themselves. The presence oflarge platelets may be indicative of a congenitalthrombopathy, such as Bernard-Soulier syndrome.The blood film also enables abnormalities in other celllines to be observed, such as Dohle bodies in thegranulocytes in the May-Hegglin anomaly. Perhapsthe most important reason to inspect the blood film isto exclude another underlying haematological disor-der-leukaemia, for example, in which platelet func-tion may be impaired. It also allows other medicalconditions, which may result in an acquired bleedingdisorder, to be detected-liver, renal, ormyeloproliferative diseases.
THE BLEEDING TIMEOf the many methods described for performing thisuseful in vivo test, the most sensitive, reproducible,and clinically valid is the template technique' in whichthe bleeding time of one or more small cuts isdetermined under highly standardised conditions.
Technique(a) Taking care to avoid superficial veins, scars, orbruises, a small area of skin on the lateral aspect of thevolar surface of the forearm about two inches belowthe elbow joint, is cleansed with alcohol (and, ifnecessary, shaved), and allowed to air dry.
(b) A sphygmomanometer cuff is placed around theupper arm and kept inflated to a pressure of40 mm Hgthroughout the test, to standardise the intra-capillarypressure.
(c) A template device, preferably a semiautomatic,spring-loaded blade device such as the Simplate II(General Diagnostics), is placed firmly, but withoutexcessive pressure, on the skin in the longitudinal
858
copyright. on 17 A
ugust 2018 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.42.8.858 on 1 A
plane and the cut(s) made 5 mm long and 1 mm deep.After one second the device is removed and a stop-watch started.
(d) All blood emerging from the wounds is blotted offat 30 second intervals using Whatman No I filterpaper. The wounds should not be touched so as toavoid dislodging any platelet aggregates formingwithin them.
(e) The time at which each wound stops bleeding isnoted and individually reported.
(f) After completing the test the cuff is removed anddried blood around the wounds is cleaned away. Theedges ofthe wound are drawn together and firmly held
with sterile butterfly plasters to minimise scar forma-tion.
InterpretationThe normal template bleeding time is up to 10 minutes,although in most healthy subjects it falls within therange two to eight minutes. A prolonged bleeding timein a person with a normal platelet count is mostfrequently due to either von Willebrand's disease or a
functional platelet defect, although many other causesexist (table 1). A normal bleeding time does not excludeeither of thee two diagnostic possibilities.The bleeding time in thrombocytopenia: When thewhole blood platelet count is in the range 20-100 x I09/l, the relation between it and the templatebleeding time can be expressed by the Harker-Slichterequation,2 in which an "expected bleeding time" for a
LowI
---
aSPDGrey Platelet
859
copyright. on 17 A
ugust 2018 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.42.8.858 on 1 A
given platelet count can be calculated using thefollowing:
Expected bleeding time = 30 - Platelet count (I0O/l)(in minutes)
This equation has been used as an alternative toplatelet aggregometry where the latter is precluded insevere thrombocytopenia (see below). Because of theinherent variability of both the bleeding time and theblood platelet count, the results obtained using thisequation should be interpreted with caution. If theactual bleeding time exceeds the expected bleedingtime by more than about five minutes, however, afunctional platelet defect is indicated.
Platelet aggregation
PRINCIPLEMany methods for monitoring platelet aggregationhave been described. In selecting a technique forroutine use, the important considerations are its speed,simplicity, reproducibility and, above all, its ability toidentify and classify those platelet defects which haveclinical importance. The method recommended isbased on the nephelometric principle originallydeveloped by Born3 wherein the changes in the opticaldensity of platelet rich plasma was found to reflect thedegree of platelet aggregation induced by one of anumber of agonists.
PRECAUTIONS PRIOR TO STUDYING PLATELETAGGREGATIONUnless their effect on platelet aggregation is beingspecifically investigated, all drugs known to influenceplatelet function should be excluded for several days
before testing, or for at least the time required for theirelimination from the body. In particular, at least 10days should have elapsed since the patient last ingestedaspirin, as this compound permanently acetylatescyclo-oxygenase, one of the enzymes involved inthromboxane A2 formation in platelets. Most otherdrugs have a short term and less predictable effect,only influencing platelet function while they, or theirmetabolites, remain in the circulation. Some drugswhich are known to affect platelet function are listed intable 2. This list is by no means comprehensive and, ifany doubt exists about a particular compound, thecompany manufacturing it should be consulted.Many normal dietary constituents such as alcohol,
onions, garlic, peppers and ginger may also inhibitplatelet function if ingested to excess and should beborne in mind when interpreting results of plateletaggregation studies. Because chylomicrons can inter-fere with the nephelometric evaluation of plateletfunction, studies should not be carried out shortlyafter a fatty meal. It is neither necessary nor desirablethat the patient should be fasted overnight.
COLLECTION AND PROCESSING OF THE BLOODSAMPLEPlatelets adhere to foreign surfaces, and it is importantto minimise this process by using siliconised glass orplastic ware at all stages. Because of the vital role ofcalcium ions in platelet responses, the anticoagulantused must be standardised and for routine purposes;one volume of 0 106M tri-sodium citrate dihydrate tonine volumes ofblood is recommended. Some workersprefer to add 1% dextrose to the anticoagulant, butthis may cause confusion if it necessitates having adifferent collection tube from that used for routineblood coagulation tests. For critical studies, the ratioof citrate to blood should also be adjusted to take
Table 2 Drugs which can interfere with plateletfunction
Association of Clinical Pathologists, Broadsheet 122, August 1989account of the haematocrit. This can be done by usingthe formula below, assuming a normal haematocrit of40%.Amount of citrate required for a 5 ml blood sample =
0-5 x(100 - haematocrit)
60Chilling activates platelets and so the citrated blood
should be processed at 20-25°C. Because the functionof the platelets deteriorates progressively in storedblood, it is advisable to start processing immediatelyafter blood collection. Platelet rich plasma (PRP) isprepared by centrifugation for 10-15 minutes at200 x g. The PRP is carefully removed, avoidingcontamination with erythrocytes or the buffy coat,and stored at 20-37°C until tested. Although theprecise temperature of storage is not critical, it doesinfluence the results somewhat and so should beadhered to for all such samples.
Platelet aggregation depends on pH, and so the PRPshould be maintained within the pH range 7-7-8-0.Satisfactory control ofpH can be achieved by storingthe PRP in full, tightly stoppered tubes and bycompleting the tests within two hours of bloodcollection. Buffers such as N-2-hydroxyethyl-piperazine-N'-2-ethanesulphonic acide (HEPES) canthemselves influence platelet aggregation and are notrecommended.
Immediately after centrifugation, platelets arerefractory to aggregating agents and so the PRPshould be allowed to "recover" for 30 minutes beforecommencing testing. After removal of the PRP, theremaining blood is centrifuged again for 20 minutes at2000 x g to obtain platelet poor plasma, which is thenused to blank the aggregometer and to dilute the PRPif necessary (see below).PLATELET COUNTA platelet count should be carried out on the PRP afterwhich many workers adjust the count to somearbitrary "standardised" level. The rate of aggrega-tion of the PRP within the cuvette is, to some extent,dependent on the platelet count. Over the range 200-400 x 109/l such an effect is small but it is usual,particularly when studying hypoaggregable platelets,to standardise the count to about 300 x 109/l withautologous platelet poor plasma (prepared as des-cribed above). Occasionally for research studies andparticularly when testing patients with thrombocyth-aemia, it may be more appropriate to test the patient'sPRP without dilution instead of or in addition totesting at 300 x 109/1. At PRP platelet counts below200 x 109/l there is a sigificant, progressive decreaseofin vitro platelet aggregatory responses, due partly tobiophysical factors such as the lower frequency ofplatelet-platelet collisions, and partly to the sensitivitylimits of the aggregometer. It is recommended that
under these circumstances, the aggregation patterns ofthe patient are compared with those of healthycontrols whose PRP has been adjusted to a similarcount. Attempts to concentrate platelets by cen-trifugation followed by resuspension in a smallerplasma volume, virtually always induce a functionalchange in the platelets (usually refractoriness) and it isnot recommended. Where the whole blood plateletcount falls between 10 and 100 x 109/1, the Harker-Slichter formula relating bleeding time to plateletcount (see above) may give some indication of func-tional integrity.
PLATELET AGGREGOMETERSA wide range of platelet aggregometers is availableand the choice is largely a personal one. For a routinelaboratory, a two channel aggregometer with pre-settemperature and stirring speed may be satisfactory; forresearch applications it may be desirable to purchase amachine on which these parameters can be varied.
Additional features are also available, such as theability to measure adenosine triphosphate release orchanges in ionised calcium concentrations simultan-eously with platelet aggregation. Other machineswhich allow aggregation to be studied in whole bloodrather than on PRP are also obtainable (see below).For diagnostic purposes, it is highly desirable that theaggregometer chosen should be able to distinguishclearly between the primary and secondary waves ofplatelet aggregation (see below).
Platelet responses are dependent on temperatureand routinely, 37C is selected for aggregation studies.Similarly, because stirring speed in the aggregometerinfluences the aggregation pattern, this should be set at800-1000 rpm, which produces satisfactory results.Another variable is the chart speed of the pen
recorder. Although this is not so critical, a speed of1 cm/minute is convenient because it allows a succinctrecord of the results to be copied and entered into thepatient's notes if required.The final variable is the sample size accepted by the
aggregometer. Most machines operate with a finalvolume of0 5 ml in the cuvette, but with some (Paytonaggregometers), a range of cuvettes are availableallowing sample volumes of 0 1-1 0 ml of PRP to beused. It is important to remember that the aggregationpatterns will vary somewhat with the sample volumesand so a normal range must be established for each sizeof cuvette used.
AGGREGATING AGENTSThe five aggregating agents listed below should besufficient to discriminate between the various func-tional platelet defects. For research purposes, severalother agonists can also be used including thrombin,serotonin, vasopressin, thromboxane A2 (TXA2)
861
copyright. on 17 A
ugust 2018 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.42.8.858 on 1 A
Association of Clinical Pathologists, Broadsheet 122, August 1989analogues, platelet activating factor (PAF) and thecalcium ionophore A23187. Most of these are nowcommercially available.I Adenosine 5-diphosphate (ADP): A stock solutionof 1 mmol/l of the di-sodium salt is prepared in 20%Owren's barbitol buffer in isotonic saline, pH 7 35(OBS), and stored in small aliquots at below - 20°C,whence it is stable for at least three months. Oncethawed, the solution should be used within three hoursor discarded. Further dilutions as required are made inOBS.The pattern of response to ADP depends on its final
concentration. At 2 imol/l, clearly defined primaryand secondary waves of aggregation can usually bedistinguished, the first representing the direct agonist-induced effect and the latter being due to the release ofendogenous ADP and to a lesser extent of TXA2,which itself induces platelet aggregation. Below2 pmol/l, progressively fewer normal subjects show asecondary response and the primary wave usuallyreverses as the added ADP is enzymatically degraded.Above 3 pmol/l, the primary wave may be so intensethat the distinction between it and the secondary phaseis masked. ADP induces a change in the shape ofplatelets from a bi-convex disc to a spiky sphere. Thischange initially causes a slight increase in the opticaldensity (OD) of the PRP which can usually only bedistinguished clearly if primary aggregation isimpaired.2 Adrenaline: A stock solution of 1 mmol/l of thebitartrate salt is made in OBS. It should be stored andused exactly as prescribed for ADP. With adrenaline,the concentrations used and the aggregation patternsobtained are very similar to those seen with ADP. Theprimary wave, however, is not preceded by a shapechange, does not reverse in the absence of a secondarywave, nor is it ever so intense that the secondary waveis masked.3 Collagen: Several suitable collagen suspensions arecommercially available, and for purposes of stan-dardisation, are preferable to home-made prepara-tions. The potency of different collagen preparationsvaries considerably and so it is advisable to keep to onesupplier and to establish a normal range for thematerial used. A very potent, stable suspension ofequine tendon collagen fibrils (1 mg/ml) is availablefrom Hormon-Chemie, Munich, West Germany and iswidely used. It is stored at 4°C and must be well mixedimmediately before dilution in the buffer packagedwith it. It should be used at a final concentration of0-5-2 0 ug/ml and diluted suspensions are stable at4°C for one week.With collagen, no primary wave of aggregation
occurs. The response is usually defined by the durationof the lag phase before aggregation begins and by theextent of the latter. As collagen induces release ofendogenous platelet ADP, a slight increase in OD, dueto the platelet shape change, usually precedes aggrega-tion.
4 Ristocetin: Several sources of ristocetin (such asLundbeck Ltd) are now available. The dry powdershould be stored at 4°C until dissolved in OBS (5-15 mg/ml). Solutions should be stored frozen at below- 20°C where they are stable for many months. At afinal ristocetin concentration of 1-0-1 -25 mg/ml inPRP, distinct primary and secondary waves ofaggregation are usually discernible, but above thisconcentration the direct effect is often so intense thatthe two phases merge. The primary phase reflects theinteraction between platelet membrane glycoproteinlB and von Willebrand factor. Its intensity can betaken as a measure of the amount of von Willebrandfactor in the plasma, while the second phase reflectsthe release of endogenous substances-for example,ADP from the platelets. It is becoming commonpractice also to test PRP with a lower ristocetinconcentration (0.3-0-5 mg/ml) as this is reported toidentify the type Ilb variant of von Willebrand'sdisease. At this lower concentration, platelets fromother vWD variants or normal subjects do not res-pond.5 Arachidonic acid: Sodium arachidonate (99%purity) is dissolved in OBS to a concentration of10 mmol/l. Small aliquots are placed in darkened glassphials which are then flushed with nitrogen to preventoxidation, then tightly stoppered, and stored below- 20°C. When added to PRP at final concentrations of0 5-1 5 mmol/l, a large, monophasic wave of aggrega-tion usually occurs, sometimes preceded by a short(< 10 seconds) lag phase. Higher concentrations maycause some platelet lysis and should be avoided.
Aggregation procedure
The precise details for carrying out platelet aggrega-tion studies will depend to a large extent on themachine used. Most will operate on a sample volumeof 0 5 ml, in which case the protocol given below willgive satisfactory results. It is important to use thecorrect size ofcuvettes and stir bars as indicated in themanual supplied with the machine. It is preferable touse siliconised glass or plastic cuvettes and stir bars soas to minimise platelet activation by the foreignsurfaces.
1 PRP (450 pl) is placed in a siliconised glass or clearplastic cuvette and a siliconised stir bar added. Any airbubbles present are dispersed by gentle tapping andthe cuvette is placed in the aggregometer well.
2 With two-channel aggregometers it is convenientto use half of the width of the chart paper for eachchannel. With a 10 millivolt full scale deflection, thePRP settings used are 0 5 and 5*5 millivolts forchannels one and two, respectively, and the correspon-ding blank values (which are set using platelet poorplasma) would be 4 5 and 9 5 millivolts.For critical work it is advisable to compare the
862
copyright. on 17 A
ugust 2018 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.42.8.858 on 1 A
Association of Clinical Pathologists, Broadsheet 122, August 1989
patient and control responses to a given agonist usingthe same channel, as the sensitivity settings and hencethe aggregation characteristics ofeach channel will beslightly different. For most purposes, however, this isunimportant.
3 Once the aggregometer is set up, the PRP cuvette isincubated in the well for about two minutes to warm to37°C. Some machines have pre-warming ports whichallows this stage to be bypassed.
4 Aggregating agent (50 p1) in an appropriate con-centration is then delivered to the bottom of thecuvette using a semiautomatic pipette. This ensuresadequate mixing and the transient interruption to thelight path provides a useful marker of the startingpoint. Spring-loaded syringes, which inject the agonistunder high pressure, are unsuitable as they introducemicrobubbles which affect light transmission for sometime, and may cause disruption of some platelets.
5 The change in light transmission of the PRP ismonitored continuously for three to five minutes. Thisprocess is repeated with each agonist usingappropriate concentrations. The final agonist concen-tration will, ofcourse, be one tenth that ofthe strengthadded to the PRP.
INTERPRETATION OF RESULTSWhen attempting to interpret platelet aggregationtracings it is generally more instructive to examineresponse to several concentrations of each agonistrather than to calculate a numerical value for the rateor extent of aggregation. Each agonist producescharacteristic features (see above under aggregatingagents) and the important point to determine iswhether aggregation has occurred and, if so, can bothprimary and secondary waves be elicited? Using thesecriteria, several types ofinherited platelet defect can bedelineated (table 3). The most appropriate additionalinvestigations to confirm the diagnosis may then beselected. The results with acquired platelet defects cansometimes be classified using similar criteria, but inmany cases are more difficult to interpret.
If, for convenience of reporting results, a numericalvalue for aggregation is required, then some
reproducible aspect ofthe curve can be measured, suchas the percentage fall in OD between settings for PRP
and platelet poor plasma three minutes after additionof a fixed dose of the agonist, and the patient's resultreported along with the normal range. Recommendedfinal concentrations of different agonists for thispurpose are ADP 2.pmol/l, adrenaline 2 pmol/l,collagen 1 pg/ml, ristocetin 1 25 mg/ml and ara-chidonic acid 1 mmol/l (table 3).
Common sources of error
No response on addition ofaggregating agent Stir baromitted or stuck in cuvette, stirrer motor not switchedon.Rapid large oscillations of pen before onset ofaggregation Air bubbles or other foreign body incuvette. A low platelet count will also cause this effectdue to the high sensitivity setting required. The intenseagglutination caused by ristocetin is also accompaniedby large excursions of the pen.Fall in optical density before addition of the agonistSpontaneous platelet aggregation which may be due tohyperaggregability or, more commonly, to a dirtycuvette or stir bar. A small fall in OD often occurs inlipaemic samples as some lipids dissolve on warmingto 37TC.Trace has normal biphasic shape but extent offall in ODis reduced PRP is contaminated with red cells, whitecells, or chylomicrons which contribute to the OD butnot to aggregation. It may be difficult to eliminatemicrocytic red cells by differential centrifugationwithout concomitant loss of some larger platelets.Fall in OD occurs on addition of agonist withoutoscillations ofpen Agonist has caused platelet lysis sothat the PRP becomes clear. This only usually occurswith high concentrations of arachidonic acid, TXA2analogues or A23 187.
Whole blood aggregometryAs mentioned earlier, aggregometers which measureplatelet responses in whole blood, using the principleof electrical impedance, are now available. To theextent that this permits aggregation to be monitored inthe presence ofred cells and leucocytes, these machinesmay be considered to be more physiological than thosebased on the nephelometric technique. Comparedwith the latter, however, the aggregation patterns inwhole blood are rather less consistent and less readilydistinguishable in defects of platelet release, such as
Table 3 Characteristics ofaggregation traces in different congenital defects ofplateletfunction
Bernard-Soulier syndrome Biphasic Biphasic Normal Normal AbsentThrombasthenia None None None None First phaseStorage pool deficiency First phase First phase Reduced Normal or reduced First phaseDefective release mechanism First phase First phase Reduced Reduced First phaseGray platelet syndrome Usually normal Usually normal Usually normal Normal Normal
863
copyright. on 17 A
ugust 2018 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.42.8.858 on 1 A
864 Association ofClinical Pathologists, Broadsheet 122, August 1989
storage pool disease. For this reason, measurement ofplatelet aggregability in PRP is preferred for routinepurposes, whereas whole blood aggregometry remainsmainly a research tool.
FURTHER INVESTIGATION OF PLATELET FUNCTIONThe methods described above will assist in the detec-tion of a functional platelet defect, although before adefinitive diagnosis can be made, more specialisedinvestigations may be needed. These include thedetermination of intraplatelet concentration ofplatelet adenine nucleotides (ATP and ADP) andintraplatelet or plasma concentration of plateletspecific peptides such as f thromboglobulin andplatelet factor 4, thromboxane A2 assays, calcium fluxmeasurements and phospholipid turnover studies.Measurement ofmembrane glycoproteins is necessaryfor the unequivocal diagnosis of Bernard-Souliersyndrome and Glanzmann's thrombasthenia. Thesespecialised investigations are available at specialist orreference laboratories. The glass bead adhesion test,once quite popular for the detection of von Wille-brand's disease, platelet function disorders, andhypercoagulable states, has been superseded byplatelet aggregometry and is no longer widely used.Similarly, the Prothrombin Consumption Index, for-mally used widely to detect both platelet and bloodcoagulating disorders, has fallen into disuse in mostlaboratories. It is, however, a sensitive marker of adefect in platelet procoagulant activity and issometimes used for this purpose.
If an abnormal platelet aggregation pattern isobserved, it is advisable to repeat the assessment on atleast one further occasion to check for the consistencyof the abnormality. A recent report suggested that aproportion of patients with storage pool deficiencyshowed a normal platelet aggregation pattern. Ifconfirmed, this observation suggests that the assay ofplatelet adenine nucleotides should be carried out onall patients tested for platelet function. This maynecessitate referral to a specialised centre where suchtests are routinely carried out.
Platelet sizing
The growing literature on this subject suggests thatunder certain circumstances platelet sizing may be avaluable diagnostic discriminant of underlying path-ological processes. Only time will tell if the initialenthusiasm is justified but, in any case, the blood orplasma concentration ofvarious substances containedin or released by the platelets such as adeninenucleotides, / thromboglobulin, and 5-hydroxytryp-tamine will depend to some extent on platelet volumeas well as on platelet number and metabolic capacity.The potential to measure platelet volume accurately isa feature ofmany ofthe newer blood cell counters usedin routine haematology laboratories. The principlesinvolved and the detailed methodology used are
beyond the scope ofthis Broadsheet and those interes-ted should contact the manufacturers of thesemachines directly. Cell sizing is currently being asses-sed as a means of monitoring loss of single plateletsinduced by aggregating agents. Some interestingresults are being produced, but at present, the value ofthis approach for diagnosis ofplatelet function defectsis uncertain.The technique of platelet sizing has a number oflimitations:(i) Platelet size is not constant in shed blood. Whenstored in vitro in anticoagulant, platelets swellprogressively and this must either be prevented byfixation or the tests undertaken at a set time afterblood collection.(ii) Various physiological activities such as exerciseand other adrenergic stress can affect both plateletnumbers and platelet size, and this must be taken intoaccount in evaluating results.(iii) Racial variation in platelet count and volumehave been described, the mean count being lower andthe mean volume higher in Mediterranean racescompared with Northern Europeans.
Results obtained of platelet volume in variouspathological states must therefore be interpreted in thelight of these physiological and technical influences.
Referewnes
1 Mielke CH, Haneshiro MM, Maher IA, Weiner JM, Rapaport SI.The standardised normal Ivy bleeding time and its prolongationby aspirin. Blood 1969;34:204-15.
2 Harker LA, Slichter SJ. The bleeding time as a screening test forevaluation of platelet function. N EnglJ Med 1972;287:155-9.
3 Born GVR. Aggregation of blood platelets by adenosine diphos-phate and its reversal. Nature 1962;194:927-9.
General revews
Harker LA, Zimmerman TS. Measurement of platelet function.Edinburgh: Churchill Livingstone.
Ludlam CA. Assessment of platelet function. In: Bloom AL, ThomasDP, eds. Haemostasis and thrombosis. 2nd edition. Edinburgh:Churchill Livingstone, 1987:933-52.
Yardumian DA, Mackie IJ, Machin SJ. Laboratory investigation ofplatelet function: a review of methodology. J Clin Pathol1986;39:701-12.
RA HUTTONCA LUDLAM*
Katharine Dormandy Haemophilia Centre andHaemostasis Unit,
Department ofHaematology,Royal Free Hospital and School ofMedicine, London,
and the Department ofHaematology,Royal Infirmary, Edinburgh, Scotland
This Broadsheet has been prepared by the authors at theinvitation of the Association of Clinical Pathologists whoreserve the copyright. Further copies ofthis Broadsheet may beobtained from the Publishing Manager, Journal of ClinicalPathology, BMA House, Tavistock Square, WCIH 9JR.
copyright. on 17 A
ugust 2018 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.42.8.858 on 1 A
