In vitro assessment of platelet function

In Vitro A s s e s s m e n t of P l a t e l e t F u n c t i o n Marcus E. Cart, Jr

p

LATELETS ARE critical to normal hemostasis. If the platelet concentration is decreased, the risk of hemorrhage is increased. If platelet function is abnormal, bleeding may be excessive, even at normal platelet concentrations. Although determination of adequate platelet concentration is relatively simple, rapid, and accurate; assessment of platelet function has remained problematic. Platelets play multiple roles in hemostasis. When vessel wall endothelium is damaged, platelets immediately adhere to the area (Fig 1). This adhesion is mediated by glycoprotein receptors on the platelet membrane. Glycoproteins Ib and IIb/IIIa anchor platelets to subendothelial yon Willebrand Factor. Adherent platelets undergo a complex series of activation events resulting in exposure of additional glycoprotein IIb/IIIa receptors on the platelet surface. The platelet cyclooxygenase pathway rapidly produces thromboxane A2, a potent platelet activating agent. Platelet granules are moved to the center of the platelet, fuse with the platelet outer membrane, and secrete their contents into the platelet cannicular system. The released materials result in rapid recruitment of additional platelets to the injury site. Recruited platelets aggregate on the adherent platelet surface via bridging of glycoprotein IIb/IIIa receptors by fibrinogen. Clotting factors released from platelet granules and absorbed from the surrounding blood are concentrated into reaction complexes on platelet surface membranes. The tenase and prothrombinase complexes thus produced result in rapid production of thrombin on the platelet surface. Thrombin leads to additional platelet activation and aggregation, and cleaves fibrinogen to fibrin monomer. Polymerizing fibrin forms a network within and on top of the platelet plug. The fibrin network is anchored to the platelets via glycoprotein IIb/IIIa. As network formation proceeds, platelets extend cytoplasmic projections

From the Departments of Internal Medicine and Pathology, Medical College of Virginia/VCU and MeGuire V.A. Medical Center, Richmond, VA. Address reprint requests" to M.E. Cart, Jr, Box 980230, Medical College of Virginia/VCU, Richmond, VA 23298-0230. Copyright 9 1997 by W.B. Saunders Company 0887-7963/97/1102-000353.00/0

106

out along fibrin strands. Retraction of these fibrinadherent pseudopodia reduces clot volume, concentrating the fibrin platelet matrix directly over the injury site. Various components of this complex series of events have served as focal points for laboratory testing of platelet function. THE BLEEDING TIME

In 1910, Duke introduced a global test of hemostasis that involved puncturing the earlobe and measuring the time required for bleeding to stop. In 1935, Ivy et al2,3 moved the puncture site from the earlobe to the forearm, prescribed a standard puncture size of 1.5-mm wide and 2.5-mm deep, and introduced the use of a blood pressure cuff inflated, throughout the measurement, to 40 mm Hg. In 1958, Borchgrevink and Waaler4 suggested blade incisions instead of punctures; and in 1969, Mielke et al 5 proposed adoption of a template to obtain uniform incisions. Since 1969, the bleeding time has been modified only by the introduction of a series of triggered blade devices that attempt to produce a uniform bleeding time incision. The introduction of multiple-triggered bleeding time devices prompted a series of studies that compare various bleeding time techniques and devices. The Simplate II | triggered blade (General Diagnostics, Morris Plains, NJ) makes two 10-ram long incisions of 1-mm depth. When compared with the Ivy bleeding time; bleeding times performed with the Simplate II are more sensitive to aspirin effects and to patient abnormalities. 6 The same study found considerable overlap between bleeding times performed in normal subjects taking aspirin, and patients with abnormal platelets. Part of the increased sensitivity obtained with the triggered blade bleeding time technique is because of the longer bleeding times achieved with these devices. Although the normal Ivy bleeding time ranges from 1.2 to 4.0 minutes (95% confidence limit), the Simplate II produces normal bleeding times ranging from 3.4 to 13.0 minutes (95% CI), and the Thrombolette (Labora Mannheim, Mannheim 31) measures normal bleeding times ranging from 1.7 to 7.9 minutes (95% CI). 7 The Surgicutt device (Intern/ttional Technidyne Corporation, Edison, NJ) makes its incision with a

Transfusion Medicine Reviews, Vol 11, No 2 (April), 1997: pp 106-115

PLATELET TESTING

107 0

Fig 1. Testable parameters of platelet participation in hemostasis: (1) Platelet shape change, (2) platelet adhesion, (3) platelet secretion, (4) platelet aggregation, (5) platelet procoagulant function with fibrin formation, and (6) platelet-mediated clot retraction.

o

0

0

\1 o

2

blade that moves through an arc instead of directly pressing down on the skin. s When compared with the Simplate-I | (Organon Teknika Corporation, Jessup, NJ) device, bleeding times obtained with this device are similar. 9 Pain experienced with both devices is similar and minimal. Scar formation with both devices varied with incision orientation. When the incisions were made parallel to the long axis of the forearm, the Simplate-I device produced no scaring at 6 months, whereas the Surgicutt resulted in scaring in 3% of subjects tested. If the incisions were made perpendicular to the long axis of the forearm, the Simplate-I device caused scarring in 3% of patients, whereas the Surgicutt did not result in scarring. In addition to scarring and pain, a major disadvantage of the bleeding time is the requirement that the patient be available for the entire 20-minute testing time. If the bleeding time could be reproduced in an in vitro form, patient blood samples could be sent to the laboratory for testing. Figure 2 is a diagram of such an approach, t0 The patient sample is drawn into a syringe for analysis. The test "vessel" is a polyethylene tube that is expanded at one end to allow attachment of a test membrane. A rectangular window is removed from the tube and replaced with a 1.0- • 0.8-cm piece of membrane. The membrane is composed of types I and III collagen, factor VIII, von-Willebrand factor, fibronectin, and calcium and is supported by a nylon mesh. A 0.3 cm to 0.5 cm horizontal slit is made in the membrane. The test is performed by pushing the blood through the test vessel with a force produced by the effect of gravity on 200g to 500g weights placed on the vertical syringe plunger. Initially, some blood oozes through the slit while the remainder passes directly out the end of the tube. The time required to stop oozing from the slit is recorded as the in vitro bleeding time. The in vitro bleeding time for normal unanticoagulated blood ranged from 20 to 60 seconds. If the blood contained citrate, the

3

4

5

6

bleeding time is 30 to 90 seconds. If the anticoagulant is ethylenediaminetetraacetic acid (EDTA), the bleeding time increases to 3 to 8 minutes. The test is very sensitive to aspirin. Further refinements of this test have not been reported. The major problem with the currently available bleeding time procedures done directly on the patient is their lack of correlation with bleeding risk. n Multiple studies are now available that show that bleeding times done preoperatively do not predict bleeding risk in patients taking aspirin or other NSAIDs, in patients with liver disease, in patients undergoing coronary bypass, or in general surgery patients. 12 For example, during thoracic surgery there was no difference shown in the perioperative fall in hemoglobin or in the amount

SLI~~~T

MEMBRANE

Fig 2. Schematic diagram showing the in vitro bleeding time test. Left panel shows the plastic tube with the w i n d o w removed and replaced by the test membrane. When a bloodfilled syringe is attached to the tube and pressure is applied (middle panel), the slit in the membrane allows passage of blood out the side of the tube. Blood also flows down the internal channel. AS platelets plug the slit, "bleeding" from the slit stops and all the blood flows out the end of the tube, or internal channel. The time required for "bleeding" to stop is recorded.

108

MARCUS E. CARR

of chest tube drainage in patients with normal (BT 4.8 --- 0.9 minutes) versus abnormal (8.0 _+ 2.7 minutes) bleeding times. 13 These unfortunate findings have led to denouncements of the direct bleeding time in the surgical, medical, and pathology literatures. 1~,14,~5 PLATELET AGGREGATION

The routine study of platelet aggregation dates back to 1962 with the introduction by Born of the turbidity technique. 16 In this procedure, various agonists are added to anticoagulated samples of platelet-rich plasma. The agonists bind to the platelet causing platelet activation and exposure of glycoprotein (GP) IIb/IIIa in the appropriate conformation. Fibrinogen binds to GP IIb/IIIa receptors on adjacent platelets and serves as a molecular adhesive allowing platelets to stick together. This process is followed optically as an increase in light transmission caused by the clearing of the solution with the formation of macroscopic aggregates. The basic technique has remained unmodified since its introduction. Through the use of multiple concentrations of various agonists (ADR collagen, ristocetin, epinephrine, thrombin, etc), information regarding the diagnoses of von Willebrand disease, BernardSoulier syndrome, storage pool disease, and abnormalities of thromboxane A2 metabolism may be obtained. Significant technical constraints include the need to schedule the test (aggregation should be measured within 4 hours of drawing blood), the need to process the sample to platelet-rich plasma (time consuming and may cause platelet activation), the requirement of significant technical expertise, the sensitivity of the test to low doses of aspirin, and the requirement for a simultaneous control run on a norma ! platelet-rich plasma. Whole blood aggregation studies can be performed on unprocessed whole blood and thus avoid at least one of the major difficulties of turbidimetric aggregation studies. 1~,18In this procedure, platelets form a mon01ayer off electrodes, and subsequent platelet aggregation results in a change in electrical impedance, which can be measured. In addition to not requiring sample processing, the whole blood test seems to function well with samples obtained in standard 3.8% sodium citrate vacutainer tube used for other coagulation tests. 19 Whole blood aggregation responses are similar for samples held at room temperature for 3 hours. The extent of collagen-induced platelet impedance aggregation is

dependent on platelet count over the range of 25,000 to 300,000 platelets per gL? ~Above 300,000, the measured impedance reachs a plateau. When ADP is the agonist, the extend of impedance change continues to increase at platelet concentrations between 300,000 to 900,000 per gL, but aggregation with ADP is undetectable below 100,000 platelets per gL. Whole blood aggregation offers several additional advantages. Using a luciferin-luciferase system that produces luminescence, the whole blood lumiaggregometer (Chronolog Corporation, Havertown, PA) allows simultaneous measurement of platelet aggregation and ATP secretion. Another potential advantage of the impedance technique is the ability to detect hyperaggregation in some disease states. In a study of patients with newly diagnosed diabetes mellitus, hyperaggregation was noted with both ADP and arachidonic acid. 2~Hyperaggregation seemed to partially correct with good diabetic control. A recent innovation in platelet aggregation is the introduction of stagnation point flow aggregometry.22 This technique flows platelet-rich plasma towards a stationary coverslip. A dark field microscope is used to monitor adhesion of the platelets to the glass and the aggregation of platelets onto adherent platelets, The deposition of platelets is visible because the increase in scattered light. The kinetics of platelet deposition is recorded as a voltage signal that increases as the scattered light intensity increases. Analysis of the voltage kinetic curves allows separation of the adhesive versus aggregation components of platelet deposition. The major advantage of this technique is that it simultaneously measures adhesion and aggregation. With this technique, increased platelet adhesion and increased spontaneous platelet aggregation would be detected in patients with peripheral artery disease. 22 Disadvantages of the technique include the requirement for sample processing (platelet-rich plasma), the necessity for anticoagulation, and the need for significant technical expertise. PLATELET ADHESION

Attempts to assess platelet adhesion have varied with respect .to the substance on which platelet adhesion is measured. In its most complex form, platelet adhesion to umbilical vein ground substance is measured. 23 Such a technique, although relevant from several physiologic aspects, is techni-

PLATELET TESTING

109

cally demanding and not easily adaptable to widespread clinical use. The propensity of platelets to stick to glass led to the development of commercially produced glass bead columns for quantitating platelet retention. Blood is pumped through the column and the change in the sample platelet count before entering and after exiting the column is measured. 24,25 Such measurements allow calculation of the percent adhesion. Some techniques use a two-stage adhesion measurement. The first stage supposedly measures primarily platelet surface adhesion, whereas the second stage assesses plateletplatelet interaction. 25 Unfortunately, variable platelet adhesion in "normal" samples leads to wide 95% confidence limits, and adhesion also varies with the column used. 26 Such test characteristics have limited the use of this test to the point that it is infrequently performed in large laboratories and unavailable in smaller ones. While the popularity of platelet glass bead adhesion measurement has declined, alternative techniques to assess platelet adhesion have been introduced. These new assays monitor platelet adhesion to glass fiber filters, Dacron filters, and collagen membranes. The newer techniques share several characteristics. They all use a regulated pressure gradient to produce flow through a filter and an anticoagulated platelet sample (platelet-rich plasma or whole blood). Typically, rate of flow and/or time to filter occlusion are measured. An apparatus used to measure flow of whole blood or platelet-rich plasma through a glass filter was described by O'Brien and Salmon 27 in 1987 and is

VALVEI FILTER. ~ LIGHT ~ SOURCE

~

CATCHRESERVOIR Fig 3. Schematic diagram showing the platelet filtration and plugging of glass fiber filters. The syringe is loaded with blood by creatinga vacuum with a syringe attached to port C of valve 2, A pressure head is generated by the pump, measured by the transducer, maintained in the air reservoir, and applied to the sample via valve 2. When pressure is applied, blood is forced through the filter. The blood exiting is counted as/drops per second and collected in EDTA tubes for cell count analysis. Measured parameters include flow rate versus time and percent platelet adhesion. The pressure head is variable, allowing measurement of the effect of shear stress on platelet adhesion.

13.5mm

3 psi SAMPLE \ ~ CLAMp__ll~gl~

O

Filterdisk ~

2gR2I CATCH

Gaskets

4~ chamber (bofforn)

RESERVOIR

Fig 4. Schematic diagram showing the apparatus for the filter bleeding time. Blood is forced to flow through a 4-mm diameter Dacron filter under a pressure head of 3 psi. Flow is monitored as drops per minute.

depicted schematically in Figure 3. The glass filter is composed of glass fibers 0.1 to 3.4 ~m in diameter. By varying the pressure gradient used to induce flow, this device can be used to study the effect of high shear on platetet aggregation. When 5 mm Hg and 40 mm Hg pressures are used to force identical samples through the filter, the filter only becomes blocked under the high shear condition. Examination of the filter shows occlusion by platelet aggregates. The effects of ptatelet glycoproteins (GP Ib, GP IIb/llIa); adhesive proteins (vWF, fibrinogen); membrane active drugs; and anticoagulants, whose aggregation is dependent on shear rate, can be assessed using this technique. By collecting and performing cell counts on the filtered sample, the effects of shear on retention of platelets and white cells can also be measured. Electron micrographs of "filtered" platelets often show evidence of activation with movement of granules to the platelet center. The filter bleeding time (FBT) (Fig 4) was introduced by Uchiyama, et a128 in 1983. In this system, anticoagulated blood is forced through a Dacron filter under a constant pressure of 3 psi. Flow is monitored as drops per minute, and the FBT is defined as the interval between initiation of flow and a slowing in flow to less than one drop in 30 seconds. "Bleeding" volume, initial "bleeding rate," and percentage platelet adhesion are also measured. The FBT in 23 normal, human volunteers was 4.33 _+ 1.79 minutes, and the FBT correlated (r = - . 9 1 ) with platelet count in thrombocytopenic dogs. FBT more than doubled in normal humans after one dose of 650 mg of oral aspirin. 29 FBT in von Willebrand pigs was greater than 4 times the value for normal pigs. 29Additional development or application, of this technique has not been reported.

110

MARCUS E. CARR

CONTROLLED VACUUM

I

I PR~ss~

1

l TPANSDUCER [

[

I

I,I ~ I!l-

~

FILTER CAPILLARY BLOOD

SAMPLE

Fig 5. Schematic diagram showing the Thrombostat 4000 and PFA 100 instrument. A controlled vacuum is used to draw anticoagulant blood through a capillary tube and into the measurement chamber. Within the chamber, the blood is passed through a collagen-coated cenulose-acetate aperture. The collagen membrane can be coated with 10 pg of epinephrine or 50 pg of adenosine 5'-diphosphate (ADP). In the PFA 100 instrument the aperture is contained within a disposable test cartridge. Initial flow (pL/min), total volume flow, and time to stop flow are measured,

In 1985, Kratzer and Born introduced the in vitro bleeding test with the Thrombostat 4000 (yon der Goltz, Seeon, Germany).3~ In this instrument (Fig 5), anticoagulated (3.2 or 3.8% sodium citrate) whole blood is drawn through a capillary and is made to pass through a collagen/cellulose aperture under the influence of a controlled vacuum. The collagen contains calcium chloride and ADR or epinephrine. In its original design, three parameters were measured: IF, initial flow rate (~tL/min); T, time to aperture occlusion; and V, the volume that passes through the aperture before closure. The measurement has been shown to be exquisitely sensitive to aspirin. 31 After a single 1000-rag oral dose of aspirin, the total volume flow was significantly prolonged for 5 days. Twenty-four hours after a 50-mg dose of aspirin, the volume flow also was significantly increased. The test seems to be abnormal in patients with vWD, as well as in patients with an uremic platelet disorder. 32 The original device has undergone significant modification to allow its introduction recently as a clinical instrument. 33,34 The new product being developed

by Dade International, Miami, FL, and marketed as the Platelet Function Analyzer (PFA 100) uses disposable ADP and epinephrine cartridges. The measured parameter is the mean PFA closure time. The coefficient of variation ranges from 7% to 8% for the ADP and approximately 12% for the epinephrine cartridge. Analysis of the Receiver/Operator characteristics curve (sensitivity versus 1-specificity) for normals and abnormals (total number 151), showed a better performance by the epinephrine cartridge. Both cartridges seemed to out-p~erform the Ivy bleeding time. 34 GLOBAL ASSESSMENT

OF H E M O S T A S I S

-

Several instruments and techniques have been developed that attempt to simultaneously assess the adequacy of both fluid phase and platelet hemostasis. The Thromboelastograph Coagulation Analyzer (TEG) (Haemoscope, Morotn Grove, IL), th6 Sonoclot Analyzer (Sienco, Morrison, CO), the Hemodyne Hemostasis Analyzer (Hemodyne, Richmond, VA), the Platelet-Induced Thrombin Generation Time (PITT), and the Haemostatometer-CSA (Xylum Corporation, Scarsdale, NY): Clot Signature Analyzer which attempt to do this, will be briefly discussed below. The thromboelastograph technique and instrument were originally introduced by Hartert in 1948. 35 In the thrombelastograph (Fig 6), whole blood is added to a sample cup into which an inner

~ T 0 r s i 0 n Wire

./." s0m

I

,, \,, / '~'4.45''~\, \

~

iln

Fig 6. Schematic diagram of the Thromboelastogram instrument and definition of measured parameters. The test sample is placed in the sample cup, which is rotated through an angle of 4.45 ~ As the sample clots, connections are established between the, inner piston and the sample cup wall. When this occurs, the inner piston begins to oscillate. The stronger the clot, the stronger the coupling, and the greater the deflection of the inner piston. (R, reaction time; ~, clot formation rate; and MA, maximum amplitude)

PLATELET TESTING

piston is suspended. The outer cup is rotated back and forth through an angle of 4.45 ~ A clot is allowed to form in the space between the outer cup and the inner piston. The forming clot couples the inner piston to the rotating outer cup. When the coupling is strong enough, the inner piston begins to move. The extent of piston movement increases with the strength of the coupling. As the mechanical strength of the clot increases, the coupling increases and the piston movement increases. The movement of the piston is recorded as a function of time. Thromboelastograph data analysis (Fig 6) involves assessment of several measured parameters: R, the reaction time; c~, the angle between the baseline and the resulting slope, and MA, the maximum amplitude traced. 36 The normal reaction time is between 6 and 8 minutes, the normal oL(clot formation rate) is more than 50 ~, and the normal MA is 50 mm to 70 mm. The current instrument is available with a software package that calculates an index of coagulation (Haemoscope, Morton Grove, IL). Despite attempts at quantitation, thromboelastograph analysis remains primarily a form of pattern recognition (Fig 7). Platelet function is believed to be reflected primarily by the MA band. Unfortunately, the MA band is a complex parameter that is also very sensitive to fibrinogen concentration. Although the MA band correlates with clot strength, cl0t modulus is not measured. Recently,

lj!

jJ !

Fig 7 Examples of Thromboelastogram patterns in various pathologic states,

111

37 ~ C HEATING BLOCK & MAGNETIC STIRRER Fig 8. Schematic diagram of the viscometer sample chamber of the Sonoclot Coagulation Analyzer. An axially vibrating probe (frequency = 250 Hz) is immersed to a defined depth into a liquid. A feedback circuit adjusts the power to keep the motion of the probe constant. The amount of power required to maintain probe oscillation increases with increasing sample viscosity. If blood, PRP, or plasma is allowed to clot within the sample chamber, various signal patterns of "clot impedance" are generated.

thromboelastography has enjoyed a resurgence in use as an index for patient monitoring during liver transplantation and cardiopulmonary bypass surgery. 37-40 The Sonoclot (Sienco, Morrison, CO) instrument and technique were first described by von Kaulla et al 4j in 1975. The technique was first proposed as a measure of platelet function by Saleem et a142 in 1983. The instrument is a dynamic viscometer (Fig 8). The test sample is placed in a glass cuvette into which a plastic probe is immersed. T h e probe oscillates through an amplitude of less than 10 ~tm at frequency of 250 Hz and is attached to a transducer. The sample is stirred with a magnetic bar and is maintained at 37~ A feedback circuit maintains constant probe movement by adjusting the power applied to the probe. The mV output of the circuit is linear with log of the sample viscosity over a viscosity range of 0.69 cP to 23 CP. This allows the instrument to be used as a plasma or whole blood viscometer. 43 If the sample is allowed to clot within the sample cup, the instrument output no longer reflects viscosity, and the signal is said to reflect mechanical impedance. This parameter is not expressed in physical units; rather the anaiogue signal is recorded versus time and expressed as centimeters of signal deflection (Fig 9). The shape of the output varies with the type of sample tested, and various portions of the curve have been interpreted by some investigators to reflect clotting events. The lack of straightforward correlation of curve characteristics with physiologic eVents, and

112

MARCUS E. CARR

15 E

0

Z 0

z

10

D

5

J c

t

o !,A,,,, 0

5

10

15

TIME ( r n i n ) Fig 9. Sonoclot signal Patterns for clots formed from platelet-rich plasma (PRP) and platelet-poor plasma (PPPL (A, lag period; B, primary wave; C, shoulder; D, secondary wave; E, peak; and F, downward Wave)

the lack of defined measurement parameters, have limited the application of this technology in the study of platelet function.44 The Hemodyne instrument and technique were initially designed to measure the forces produced by platelets during clot formation and clot retraction. 4s Clots of platelet-rich plasma, or whole blood, are formed between the surfaces of a shallow, thermostated sample cup and an overlying probe (Fig I0). The probe is attached to a transducer that generates a voltage signal if the probe is LOWERCUP

TOTRANSDUCER

AIR'CLOT INTERFACE X COVEREDWITH ~ \

moved. When platelets within the clot attempt to shrink the clot, a downward force is produced on the probe. The voltage is converted to force with a calibration constant derived by measuring the voltage generated when a force of known magnitude is imposed on the system. The force constant, which is repeatedly measured during a sample run, can be used to calculate the clot modulus. 46 Instrument output can be displayed in an unprocessed form on an x-y plotter or can be displayed as a kinetic curve of clot modulus and force development versus time on a computer screen. 47 Although clot modulus is dependent on both fibrin structure and platelet function, force development is primarily a platelet function. Force, which is a linear function of platelet concentration, is virtually absent in Glanzmann's thrombasthenia (Fig 11) and is decreased by glycoprotein IIb/IIIa blockade. 48 Force development is independent of clot structure and, over the range of 100 mg/dL to 400 mg/dL, fibrin concentration. The simultaneous measurement of clot modulus and force may allow separation of clot altering and platelet inhibitor e f f e c t s . 49 Although clinical data are limited, the inhibition of platelet force during cardiopulmonary bypass seems to correlate with peri- and postoperative blood loss, s~ and platelet force has been shown to be diminished in acute uremia, sl The platelet-induced thrombin generation time, which measures platelet aggregation and clot formation in a moving cuvette, was first described by Basic-Micic in 1992. 52 Figure. 12 depicts the principles of operation of the device. Light is transmit8000 m

M

4000

~/COUPLING AR I I . . . . . . . LATE

s,L,co.o,L \ \

6000

II /% SP CE

0

2000

"0 0

10

20

30 mm

Fig 10. Schematic diagram of the Hemodyne instrument, which is used to measure force developed by platelets during clot formation. As the clot retracts, a downward force is exerted on the plate attached to the transducer. The resultant downward deflection of the plate generates a voltage that is proportional to the downward force.

400

800 - - - 1 2 0 0

1600

TIME (sec) Fig 11. Force development during clotting of normal blood and blood obtained from a patient with Glanzmann's thrombasthenia in the Hemodyne instrument. Thrombin (1 NIH U/mL) and calcium chloride (10 mmol/L) were added at time zero.

PLATELET TESTING

LAMp

BLEND

PLASMAROTATESIN DISK-SHAPEDCUVEI"rE

113

PHOTOCELL AMPLIFIER

RECORDER

Fig 12. Schematic diagram of the device used to measure platelet-induced thrombin generation time (PITT). A 0.6 mL sample of PRP is placed in a disk shaped polyvinylchloride cuvette which is rotated at 20 revolutions per minute. Rotation is started exactly 30 minutes after blood collection. Light is passed through the cuvette and measured by a photometer. When thrombin is generated, the amount of light transmitted through the cuvette increases initially because of platelet aggregation. When clotting occurs, however, the sample becomes increasingly turbid and light transmission falls, The time interval from onset of rotation to achieve a decreased optical density is defined as the aggregation time Ta. The time interval from onset of rotation to an increase in optical density is defined as the coagulation time Tc.

ted through a rotating (20 rpm) cuvette that contains 0.6 mL of platelet-rich plasma. As platelets aggregate and clotting occurs, the amount of transmitted light changes. Test samples are anticoagulated with small amounts of hirudin or low molecular weight heparin. As the cuvette revolves, platelets become activated, and a small amount of thrombin is produced. Thrombin-induced platelet aggregation causes a precipitous decline in optical density. Platelet aggregation is rapidly followed by clotting of the sample, which produces an increase in sample turbidity. The time interval between onset of cuvette rotation and onset of aggregation is defined as the aggregation time T~, and the time from onset of rotation to onset of clotting is defined as the clotting time Tc. Normal Ta and Tc are about 6 and 7.5 minutes, respectively. Aspirin, iloprost, coumadin, and heparin all extend Ta a n d To. Patients with peripheral artery disease have shortened Ta and Tc. Testing of whole blood samples are not possible with the original design of this instrument. The Hemostatometer was initially described by G6r6g in 1986. 53This instrument attempts to model a bleeding vessel (Fig 13). Blood flows through an in vitro "vessel" (polyethylene tubing) under a constant pressure head (60 mm Hg). The sample of whole blood is maintained at 37~ and the pressure within the vessel is continuously measured. The vessel is punctured by a sharp, 0.15-mm diameter puncturing device. Blood leaks out the puncture "wounds," and the pressure in the vessel falls precipitously. As the holes are plugged by platelets, the pressure recovers. The time for initial closure of the puncture is taken to be a measure of

platelet function. Over time, a clot forms on top of the platelet plug, ultimately causing vessel occlusion. This event is monitored as a second, more gradual, fall in vessel pressure. The time required for secession of flow in the vessel is taken to be a measure of clotting function. This measurement has been shown to be abnormal in factor IX deficiency, factor VIII deficiency, and von Willebrand disease 54 and has been shown to be sensitive to changes caused by factor replacement or DDAVP therapy. When used as a preoperative screen of bleeding risk for patients undergoing coronary bypass grafting, the test correctly identified 80.8% of the nonbleeders but identified only 63.6% of the bleeders. 55 SUMMARY

With the realization that the skin bleeding time is often an unreliable measure of platelet function, efforts have been made to identify ways to assess qualitative platelet dysfunction. Currently available techniques measure platelet adhesion, platelet aggregation, the ability of platelets to retard or stop flow through filters, and the contribution of platelets to in vitro clot formation. Glass bead adhesion, which continues to be performed in some laboratories, is gradually being replaced by measures of platelet adhesion to filters composed of glass fibers, Dacron fibers, or collagen. In each instance, anticoagulated platelet-rich plasma or whole blood flows through the filter under a regulated pressure gradient. The 37 ~

HEATING BLOCK

I PKE URE

Punching Device

PARAFFIN OIL IN

BLOOD 0.1 ml/min, 60 mmHg

) CATCH RESERVOIR

Fig 13. Schematic diagram showing the CSA | Clot Signature Analyzer (formerly Haemostatometer). The test sample syringe is held in a heating block. The blood sample is forced out of the syringe by displacement with paraffin oil. The sample flows through the test "vessel" under a constant pressure of 60 mm Hg. The vessel is then punctured w i t h a mechanical punching device and blood leaks out of the vessel into surrounding saline. The device measures pressure within the vessel. When punctured, the pressure suddenly drops. As the puncture is plugged with platelets, the pressure returns to pre-puncture levels. As a clot forms over the plug, fl0w slows within the vessel and the pressure once again falls.

114

MARCUS E. CARR

amount of blood flowing through the filter versus time and/or the time to filter occlusion are measured. Recent developments in platelet aggregation have focused on whole blood and stagnation point flow aggregation techniques. Whole blood aggregation does not require blood sample processing and accommodates blood obtained from citrated vacutainer tubes. Stagnation point flow measures both platelet adhesion and aggregation and may be able to detect pathologically-enhanced platelet function.

Global measures of hemostasis attempt to simultaneously evaluate the adequacy of fluid phase coagulation and platelet function. Currently available techniques include Thromboelastography, SonoClot Analyzer, Hemodyne Hemostasis Analyzer, PITI', and Hemostatometry. Although each of these technologies have been shown to provide interesting data in the research setting, the ability of any of these techniques to detect abnormal or clinical inadequate platelet function remains to be established.

REFERENCES 1. 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 cases of hemorrhagic disease relieved by transfusion. JAMA 55:1185-1192, 1910 2. Ivy AC, Shapiro PF, Melnick P: The bleeding tendency in jaundice. Surg Gynecol Obstet 60:781-784, 1935 3. Ivy AC, Nelson D, Bucher G: The standardization of certain factors in the cutaneous "venostasis" bleeding time technique. J Lab Clin M e t 26:1812-1822, 1941 4. Borchgrevink CF, Waaler BA: The secondary bleeding time: A new method for the differentiation of hemorrhage diseases. Acta Med Scand 162:361-374, 1958 5. Mielke CH Jr, Kaneshiro MM, Maher IA, et al: The standardized normal Ivy bleeding time and its prolongation by aspirin. Blood 34:204-215, 1969 6. Koster T, Caekebeke-Peerlinck KMJ, Briet E: A randomized and blinded comparison of the sensitivity and the reproducibility of the Ivy and Simplate II bleeding time techniques. Am J Clin Patho192:315-320, 1989 7. Bain B, Forster T, Baker A: An assessment of the sensitivity of 3 bleeding time techniques. Scand J Haematol 30:311-316, 1983 8. Smith C. Surgicutt~: A device for modified template bleeding times. J Med Technol 3:29-31, 1986 9. Buchanan GR, Holtkamp CA: A comparative study of variables affecting the bleeding time using two disposable devices. Am J Clin Pathol 91:45-51, 1989 10. Brubaker DB: An in vitro bleeding time test. Am J Clin Patho191:422-429, 1989 11. Rodgers RPC, Levin J: A critical reappraisal of the bleeding time. Semin Thromb Hemost 16:1-20, 1990 12. Burns ER, Lawrence C: Bleeding time-- a guide to its diagnostic and clinical utility. Arch Pathol Lab Med 113:12191224, 1989 13. Burns ER, Billet HH, Frater RWm, et al: The preoperative bleeding time as a predictor of postoperative hemorrhage after cardiopulmonary bypass. J Thorac Cardiovasc Surg 92:310312, 1986 14. O'Kelly SW, Lawes EG, Luntley JD: Bleeding time: Is it a useful clinical tool? Br J Anaesth 68:313-315, 1992 15. Gewirtz AS, Miller ML, Keys TF: The clinical usefulness of the preoperative bleeding time. Arch Pathol Lab Med 120:353-356, 1996 16. Born GVR: Quantitative investigations into the aggregation of blood platelets. J Physiol 162:67-71, 1962 17. Mackie IJ, Jones R, Machin SJ: Platelet impedance

aggregation in whole blood and its inhibition by antiplatelet drugs. J Clin Pathol 37:874-878, 1984 18. Zwierzina WD, Kunz F: A method of testing platelet aggregation in native whole blood. Thromb Res 38:91-100, 1985 19. Sweeney JD, Hoernig LA, Michnik A, et al: Whole blood aggregometry--Influence of sample collection and delay in study performance on test results. Am J Clin Fatho192:676-679~ 1989 20. Sweeney JD, Labuzetta JW, Fitzpatrick JE: The effect of the platelet count on the aggregation response and adenosine triphosphate release in an impedance lumi-aggregometer. Am J Clin Pathol 89:655-659, 1988 21. Mandel S, Sarode R, Dash S, et al: Hyperaggregation of platelets detected by whole blood platelet aggregometry in newly diagnosed noninsulin-dependent diabetes mellitus. Am J Clin Pathol 100:103-107, 1993 22. Reininger CB, Reininger AJ, H/Srmann A, et al: Quantitative analysis of platelet function using stagnation point flow aggregometry. Int Angiol 11:247-255, 1992 23. Baumgartner HR: The role of blood flow in platelet adhesion, fibrin deposition, and formation of mural thrombi. Microvasc Res 5:167-179, 1973 24. Rossi EC, Green D: A study of platelet retention by glass bead columns ("platelet adhesiveness" in normal subjects). Br J Haematol 23:47-57, 1972 25. McPherson J, Zucker MB: Platelet retention in glass bead columns: Adhesion to glass and subsequent platelet-platelet interactions, Blood 47:55-67, 1976 26. Roper-Drewinko PR, Drewinko B, Corrigan G, et al: Standardization of platelet function tests. Am J Hematol 11:183203, 1981 27. O'Brien JR, Salmon GP: Shear stress activation of platelet glycoprotein IIb/IIIa plus vQn Willebrand factor causes aggregation: Filter blockage and the long bleeding time in yon Willebrand's disease. Blood 70:1354-1361, 1987 28. Uchiyama S, Stropp JQ, Claypool DA, et al: Filter bleeding time: A new in vitro test of hemostasis. I. Evaluation in normal and thrombocytopenic subjects. Thromb Res-3~1299- 116, 1983 29. Uchiyama S, Stropp JQ, Claypool DA, et al: Filter bleeding time: Anew in vitro test of hemostasis. II. Application to the study of von Willebrand disease and platelet function inhibitors. Thromb Res 31:H7-125, 1983. 30. KratzerMAA, BornGVR: Simulation ofprimary haemostasis in vitro. Haemostasis 15:357-362, 1985 31. Kretschmer V, Schikor B, S6hngen D, et al: In vitro

PLATE LET TESTI NG

bleeding test-- a simple method for the detection of aspirin effects on platelet function. Thromb Res 56:593-602, 1989 32. Alshameeri RS, Mammen EF: Clinical experience with the Thrombostat 4000. Semin Thromb Hemost 21:1-10, 1995 (Suppl 2) 33. Kundu S, Sit R, Mitu A, et al: Evaluation of platelet function by PFA-100 ~ . Clin Chem 40:1827-1828, 1994 34. Mammen EF, Alshameeri RS, Comp PC: Preliminary data from a field trial of the PFA-100 ~ system. Semin Thromb Hemost 21:113-121, 1995 (Suppl 2) 35. Hartert H: Blutgerinnung studien mit der thrombelastographie, einen neuen untersuchingsverfahren. Klin Wochenschr 26:577-583, 1948 36. Hartert H, Schaeder JA: The physical and biological constants of thromboelastography. Biorheology 1:31-39, 1962 37. O w e n CA, Rettke SR, Bowie EJ, et al: Hemostatic evaluation of patients undergoing liver transplantation. Mayo Clin Proc 62:761-772, 1987 38. Spiess BD, Tuman KJ, McCarthy RJ, et al: Thromboelastography as an indicator of post-cardiopulmonary bypass coagulopathies. J Clin Monit 3:25-30, 1987 39. Essell JH, Martin TJ, Salinas J, et ah Comparison of thrmboelastography to bleeding time and standard coagulation tests in patients after cardiopulmonary bypass. J Cardiothorac Vasc Anesth 7:410-5, 1993 40. Tuman KJ, McCarthy RJ, Djuric M, et al: Evaluation of coagulation during cardiopulmonary bypass with a heparinasemodified thromboelastographic assay. J Cardiothorac Vasc Anesth 8:144-9, 1994 41. von Kaulla KN, Ostendorf P, von Kaulla E: The impedence machine: A new bedside coagulation recording device. J Med 6:73-87, 1975 42. Saleem A, Blifeld C, Saleh SA, et al: Viscoelastic measurement of clot formation: A new test of platelet function. Ann Clin Lab Sci 13:115-124, 1983 43. Chandler WL, Schmer G: Evaluation of a new dynamic viscometer for measuring the viscosity of whole blood and plasma. Clin Chem 32:505-507, 1986 44. LaForce WR, Brudno DS, Kanto WP, et al: Evaluation of

115

the SonoClot Analyzere~ for the measurement of platelet function in whole blood. Ann Clin Lab Sci 22:30-33, 1992 45. Carr ME, Zekert SL: Measurement of Platelet MediatedForce Development during Plasma Clot Formation. A m J Med Sci 302:13-18, 1991 46. Cart ME, Carr SL: Fibrin structure and concentration alter clot elastic modulus but do not alter platelet mediated force development. Blood Coagul Fibrinolysis 6:79-86, 1995 47. Carr ME: Measurement of platelet force: The Hemodyne | Hemostatic Analyzer. Clin Lab Manage Rev 9:312-320, 1995 48. Carr ME, Carr SL, Hantgan RR, et al: Glycoprotein IIb/IIIa Blockade Inhibits Platelet Mediated Force Development and Reduces Gel Elastic Modulus. Thromb Haemost 73:4995O5, 1995 49. Carr ME, Zekert SL: Abnormal cl0t retraction, altered fibrin structure, and normal platelet function in multiple myeloma. Am J Physio1266:H 1195 -H 1201, 1994 50. Greilich PE, Carl" ME, Carr SL, et al: Reductions in Platelet Force Development by Cardiopulmonary Bypass Are Associated with Hemorrhage. Anesth Analg 80:459-465, 1995 51. Carr ME, Zekert SL: Force monitoring of clot retraction during DDAVP therapy for the qualitative platelet disorder of uraemia: Report of a case. Blood Coagul Fibriuolysis 2:303308, 1991 52. Basic-Micic M, Roman C, Herpel U, et al: Plateletinduced thrombin generation time: a new sensitive global assay for platelet function and coagulation. Haemostasis 22:309-321, 1992 53. G0r6g P: A new, ideal technique to monitor thrombolytic therapy. Angiology 37:99-105, 1986 54. Kovacs IB, Hutton RA, Kernoff PBA: Hemostatic evaluation in bleeding disorders from native blood. Clinical experience with the Hemostatometer. Am J Clin Pathol 91:271-279, 1989 55. Ratnatunga CR Rees GM, Kovacs IB: Preoperative hemostatic activity and excessive bleeding after cardiopulmonary bypass. Ann Thorac Surg 52:250-257, 1991