Platelet Aggregation

Platelet Aggregation

CHAPTER 26 Platelet Aggregation Lisa K. Jennings and Melanie McCabe White Vascular Biology Center of Excellence and Department of Medicine, Universit...

708KB Sizes 43 Downloads 188 Views


Platelet Aggregation Lisa K. Jennings and Melanie McCabe White Vascular Biology Center of Excellence and Department of Medicine, University of Tennessee Health Science Center, Memphis, Tennessee

I. Introduction

(aspirin) can totally inhibit the aggregation of platelets in response to low doses of collagen, but at higher doses aggregation will still occur. Specialized aggregometers called lumiaggregometers assess the release of adenine nucleotides from platelet storage granules concomitant with the extent of aggregation response. This assay is adapted from a method originally described by Holmsen et al.3 Lumiaggregation is a quantitative bioluminescent determination. It is based on the conversion of ADP released from the platelet dense granules to ATP which then reacts with firefly lantern extracts (luciferin and luciferase) generating adenyl-luciferon. Light is emitted when oxidation of adenyl-luciferin occurs. The light emitted is proportional to the nanomoles of ATP present in the aggregometer cuvette.4,5 Studies of platelet secretion take the investigation of platelet function one step further than simple aggregation testing and allow the assessment not only of platelet aggregation but also release (Fig. 26-3). This is useful in the diagnosis of bleeding disorders such as storage pool disease and release defects especially in cases where the patients have clinical bleeding in the absence of the abnormal aggregation tracings usually associated with these disorders.6,7 Secretion studies that provide a quantitative determination of second wave aggregation can also be useful in the investigation of platelet activation or inhibition. A more detailed methodology for assessing platelet aggregation and secretion can be found in published research and clinical laboratory procedures.8 In addition to the classic PRP system, platelet aggregation testing may be carried out in a whole blood system. The most common whole blood method is electrical impedance. This method can be used with either PRP or whole blood and measures an increase in impedance across electrodes placed in the anticoagulated blood as activated platelets accumulate on them.9 Aggregometry recordings obtained by the electrical method do not discriminate two waves of platelet aggregation or correlate with platelet secretion as

Historically, studies of platelet function were performed in an effort to elucidate the basis of bleeding in patients with a history of epistaxes, bruising, gingival bleeding, etc. The original aggregometer described in 1962 by Born1 consisted of an absorbptiometer and experiments evaluating platelet function were performed at room temperature. In the same year, O’Brien reported his results of aggregation studies using a photoelectric colorimeter run at three different temperatures.2 This methodology has progressed to an instrument designed specifically for measuring platelet aggregation that is basically a spectrophotometer attached to a chart recorder or computer. The current instrument is standardized for each subject using platelet-rich plasma (PRP) as the most opaque setting possible (0% aggregation) and autologous platelet-poor plasma (PPP) as the maximally transparent situation (100% aggregation). As platelets aggregate in response to the addition of an exogenous platelet agonist, the sample becomes more “clear” and an increase in light transmission through the test sample is recorded (Fig. 26-1a). Platelet aggregation response is calculated by dividing the distance from baseline to the maximal aggregation achieved by the distance from baseline to the theoretical 100% aggregation (Fig. 26-1b). The platelet aggregation pattern is classically thought of in terms of a primary response to the addition of an exogenous agonist, such as adenosine diphosphate (ADP), followed by a secondary response to the release of adenine nucleotides that are stored within the dense granules of platelets. These responses are often referred to as the first and second “wave” of aggregation (Fig. 26-2a). This biphasic response can be masked if high concentrations of agonists are added. With the agonist collagen, the aggregation pattern reflects the adhesion of platelets to the collagen fibrils and then the aggregation in response to the activation caused by that event (Fig. 26-2b). Acetylsalicylic acid 495 PLATELETS

Ch026-P369367.indd 495

Copyright © 2007, Elsevier Science (USA) All rights of reproduction in any form reserved.

9/23/2006 11:53:47 AM

Lisa K. Jennings and Melanie McCabe White



B Figure 26-1. A. A representation of platelet response measured in an aggregometer cuvette. Stirring platelets in platelet-rich plasma (PRP) inhibits the transmission of light through the specimen. After an agonist is added, aggregation commences and the instrument measures an increase in light transmission. The maximal amount of light transmission possible for any sample is that seen with autologous platelet-poor plasma (PPP). Adapted from White and Jennings,8 Fig 2-10, page 43. B. Calculation of percent platelet aggregation. Measure the distance between 0% aggregation and maximal aggregation (A). This value, divided by the distance between 0% aggregation and 100% aggregation (B) × 100 equals the % platelet aggregation. Abbreviations: PPP, platelet-poor plasma; PRP, platelet-rich plasma.

well as recordings obtained by the traditional optical method. In a study by Podczasy et al.,10 when inhibitors of platelet function were added to PRP, both the rate and extent of aggregation were inhibited, but the main consequence on impedance was a decrease in its rate and not in its extent. Increases in impedance and secretion of ATP were also measured in whole blood after preincubation of an antibody specific for platelet surface glycoproteins. This study10 showed that increases in impedance lagged several minutes behind the formation of platelet aggregates and the secretion

Ch026-P369367.indd 496

of platelet ATP. Therefore, this study and others suggest that there are advantages and disadvantages to both methods of measurement of platelet aggregation and that the parameters being measured must be clearly understood to properly interpret the results. In recent years, several point-of-care technologies have been developed to assess platelet aggregation. These systems include the VerifyNow Rapid Platelet Function Analyzer (see Chapter 27) and the Plateletworks/ICHOR system (see Chapter 23).11–13 VerifyNow is a unique platelet function

9/23/2006 11:53:47 AM

26. Platelet Aggregation



Figure 26-3. Platelet aggregation measured simultaneously with the release of ADP from platelet dense granules, as determined by lumiaggregometry.

B Figure 26-2. A. A classic platelet aggregation pattern showing a first wave of aggregation in response to addition of an exogenous agonist and a second wave in response to release of adenine nucleotides from dense granules. B. Collagen-induced platelet aggregation showing the delayed aggregation in response to release of stored ADP.

Ch026-P369367.indd 497

assay system that relies upon the initial agglutination of platelets to fibrinogen-coated beads and platelet response to agonist exposure. This system measures the rate of platelet response but not the extent of aggregation response. Thus data obtained from this system are more similar to slope determinations in classical light transmission aggregometry. Recent data suggest that this system has certain limitations in assessing the level of platelet inhibition to the glycoprotein (GP) IIb-IIIa antagonists ex vivo.13 New VerifyNow test cartridges for the assessment of aspirin and thienopyridine mediated inhibition of platelet function are currently under investigation. The Plateletworks/ICHOR system is another point-of-care platelet function system that measures the extent of platelet aggregation similar to that obtained with light transmission aggregometry. Initially the single platelet count in a whole blood sample is determined and, upon agonist exposure, the number of single platelets remaining in the sample is determined. These values are used to determine the percent of platelet aggregation. The Plateletworks/ ICHOR, while still a point-of-care system, has similar results to that obtained with light transmission aggregometry and provides more flexibility in terms of agonist and anticoagulant choices. The basic theory and assessment of platelet aggregation in PRP has changed very little from its origins. What has

9/23/2006 11:53:47 AM


Lisa K. Jennings and Melanie McCabe White

changed over time is the use of platelet aggregation to assess more than bleeding problems. The aggregometer has become a very useful tool in the assessment of inhibition of platelet aggregation by new anti-platelet therapies. Initially used to monitor inhibition by aspirin, light transmission aggregometry is now used to determine the pharmacodynamics of thienopyridines, antithrombins, and GPIIb-IIIa antagonists.14,15

II. Variables of Platelet Aggregation Testing In platelet aggregation testing, it is essential to develop normal ranges for the agonists being tested and to carefully control variables that might affect the test results.

A. Venipucture Blood from adult donors should be obtained using a 19 to 21-gauge needle and plastic syringe. In the case of pediatric patients, a smaller gauge needle such as a 23 to 25-gauge needle may be used. A single syringe, as opposed to the two syringe technique, may be used as long as the venipuncture is clean with no necessity for probing to find a vein.16 VacutainerTM (Becton Dickinson) collection of blood is not considered suitable for platelet aggregation measurements as increased responsiveness to low-dose ADP is typically observed from Vacutainer versus syringe-derived PRP (White and Jennings, unpublished observations). Until a Vacutainer is developed that does not increase the platelet activation or alter pharmacodynamic measurements, we recommend the use of a syringe.

B. Anticoagulant 1. Citrate. Sodium citrate (0.102 M, 0.129 M citrate, buffered and nonbuffered) at a ratio of nine parts blood to one part anticoagulant is the anticoagulant typically chosen for platelet aggregation testing. Laboratories involved in aggregation testing do not use Vacutainers to obtain blood due to the concern that the platelets may become activated by the shear force of the vacuum. Instead, a plastic syringe and a butterfly needle are used to obtain the specimen. This practice is also helpful in assuring anticoagulant consistency because all of the above listed varieties of citrate anticoagulant come in blue stoppered Vacutainers. Some laboratories correct for a subject’s hematocrit, especially if the hematocrit is very high or low, as the final plasma concentration of the anticoagulant can affect test results. Hardisty et al. showed that in individuals with a high hematocrit, more aggregating agent is necessary to produce an effect due to the decreased amount of free calcium available in the

Ch026-P369367.indd 498

plasma.17 One can correct for hematocrit either by changing the amount of citrate added to a fixed amount of blood16 or by changing the amount of whole blood added to a set amount of anticoagulant using the formula (5/(1 − 0. Hct) = amount of whole blood to add to 1.0 mL anticoagulant). We have found the latter approach to be more convenient. Agonists such as ADP, collagen, and epinephrine, at moderate concentrations, show little or no difference in aggregation results between blood drawn into 0.102 M versus 0.129 M citrate. However, when testing in the lowest concentration ranges of agonist (e.g., ADP, where higher concentrations of citrate can result in lower aggregation responses), there may be a difference in the results obtained with the different citrate concentrations. If the pharmacodynamics of the GPIIb-IIIa antagonists are to be evaluated, it is necessary to use a noncitrate-based anticoagulant. Studies have shown the binding of these antagonists to GPIIb-IIIa is calcium concentration dependent.14,18 A citrate-based anticoagulant can enhance the amount of inhibition observed in ex vivo or in vitro testing. For general aggregation testing in which citrate anticoagulant is used, buffered citrates are preferable to the nonbuffered varieties because they help maintain the pH of the PRP, thus negating the possible effects of pH change. Figure 26-4 demonstrates the platelet aggregation response to ADP in three citrate anticoagulants: 0.102 M buffered citrate, 0.129 M buffered citrate, and ACD (acid-citrate-dextrose). The latter is only used for planned preparation of washed platelets and not for plasma-based aggregation assays. 2. Heparin. Heparin, which inhibits the generation and activity of thrombin via its complex with antithrombin III, can be used for platelet testing, but in many donors the PRP platelet count will be significantly lower when collected into heparin as compared to citrate. “Spontaneous” aggregation in the presence of heparin may be seen in a small percentage of the population. Heparin is therefore not an anticoagulant of choice for platelet aggregation testing. 3. EDTA. Because platelet aggregation is dependent on the presence of free calcium in the plasma, EDTA is not suitable for use in aggregation testing. 4. PPACK. d-phenylalanine-proline-arginine chloromethyl ketone (PPACK), an antithrombin, has become a familiar anticoagulant for platelet aggregation in systems that are being used to assess platelet inhibition by the GPIIbIIIa antagonists. Because the Food and Drug Administration (FDA) approved antagonists (eptifibatide, abciximab, and tirofiban) have a calcium-dependent inhibition response, a nonchelating anticoagulant has to be used to avoid overestimation of inhibition during ex vivo testing.14 The problems associated with heparin have made it a poor choice as an anticoagulant (see above). PPACK, which has the bene-

9/23/2006 11:53:48 AM

26. Platelet Aggregation




6. ACD-A. This formulation of ACD keeps the pH of the PRP at 7.2 and may be acceptable for aggregation testing.19 If simultaneous measurements of aggregation and release are being made, it should be noted that anticoagulants such as PPACK and heparin cannot be used. The release reaction seen in citrated blood, which is used in the diagnosis of SPD and release defects, is not seen with the agonist ADP in blood anticoagulated with PPACK or heparin (Fig. 26-5). In our laboratory, the results of platelet aggregation testing performed in both PPACK and citrated blood were very similar at high concentrations of agonist. At lower concentrations of collagen, platelets from PPACK anticoagulated blood sometimes gave lower responses than platelets drawn into citrate. With very low-dose ADP (1 µM), lower responses are obtained in citrate than PPACK but at higher concentrations (5 µM) the response is greater in citrate. We find no response to 5 µM epinephrine in PPACK anticoagulant versus approximately 75% response when citrate anticoagulant is used.

C. Glass versus Plastic Processing Tubes


Figure 26-4. The response of platelets to 1 µM ADP in (A) 3.2% (0.102 M) buffered citrate, (B) 3.8% (0.129 M) buffered citrate, and (C) ACD anticoagulants. (Adapted from White and Jennings,8 Fig 2-1, page 29.)

fits of not chelating calcium and therefore not exerting an effect on platelet function based on available plasma calcium, has filled the void. Unfortunately, PPACK is very expensive relative to the other anticoagulants listed and its anticoagulant effect can be short-lived depending on the concentration used. A final concentration of 1.6 mg per 10 mL whole blood (0.3 mM final concentration) prevents the specimen from clotting for several hours. 5. ACD. This anticoagulant brings the pH of the PRP to 6.5, and is, therefore, unsuitable for use in aggregation experiments (see Section II.F below). For washing or gelfiltering platelets, it is an excellent choice and avoids platelet aggregate formation in the centrifugation process.

Ch026-P369367.indd 499

The preparation of platelets for aggregation studies should always be carried out in either plastic or siliconized glass tubes. Uncoated glass can cause platelet activation and will, therefore, artifactually affect results. In our experience, polypropylene tubes are superior to either polystyrene or polycarbonate when used for platelet preparation and/or storage.

D. Platelet Count Correction There are a variety of opinions on whether or not it is necessary or beneficial to standardize the platelet count of the PRP used in platelet aggregation assays.8,16,20 Because it has been reported that aggregation responses can vary in relation to platelet count, comparison of aggregation responses from donor to donor or in the case of multicenter studies necessitates standardization of the platelet count. This practice is recommended because established normal ranges of response to each agonist are typically carried out with adjusted PRP platelet counts.

E. Red Blood Cell Contamination and Lipemia Because platelet aggregation in PRP is based on optical transmission, the presence of any contaminating particles, such as red blood cells, or the presence of lipids can affect the ability of the aggregometer to measure platelet aggregates and can lead to a decrease in the percent aggregation.

9/23/2006 11:53:48 AM

Lisa K. Jennings and Melanie McCabe White





Figure 26-5. Platelet aggregation (lower tracing) and release (upper tracing) measured in (A) 0.102 M buffered citrate, (B) PPACK, and (C) heparin. Note the lack of release in platelets drawn into PPACK and the very diminished response in platelets drawn into heparin.

Red cell lysis can result in the release of ADP from the red cells, which in turn may cause the platelets to become refractory to the addition of exogenous ADP.

F. pH Platelet aggregation is pH sensitive and, therefore, when preparing a specimen for aggregation studies, pH must be maintained between 7.2 and 8.0. If the pH of the plasma drops below 6.4, no aggregation will occur and at a pH above 8.0, spontaneous aggregation can occur. If the pH approaches 10, inhibition of aggregation once again is evident. The change in pH of the plasma is mediated by the diffusion of CO2 out of the plasma. As the CO2 diffuses out of the plasma, the pH rises. To avoid this situation, PRP should be kept in a tube that minimizes the surface area exposed to the atmosphere (small diameter tubes), the tube should be kept capped, and the tube should be mixed as little as necessary.16 Finally, although various buffers may not affect the pH, they may affect the platelet aggregation response. Usually, isotonic saline is the diluent of choice for agonists. Phosphate buffers in particular have the effect of lowering aggregation responses.

G. Temperature While platelet aggregation should be performed at 37°C to mimic the in vivo situation, the method used to store the platelets prior to and during aggregation studies is a matter of debate. Although it has been reported that platelets stored at room temperature are more sensitive than platelets stored

Ch026-P369367.indd 500

at 37ºC, the aggregation response of platelets stored at 37ºC is less stable. Studies have suggested that storage of platelet preparations at 4ºC can prolong responsiveness, but exposure of platelets to cold temperatures can lead to a spontaneous aggregation response upon rewarming and stirring of the platelet suspension.16 Furthermore, when the prechilled platelets have been incubated at warmer temperatures for longer periods, the agonist-induced aggregation response is higher compared to a control platelet sample. Thus, the choice of the “correct” method of storage for platelets to be used in aggregation testing is unclear; however, the most consistent results are those obtained with storage at room temperature in a capped tube.16

H. Aggregometer Stir Speed In order to aggregate, platelets must come in contact with each other. If an agonist is added to nonstirred platelets, they will become activated but will not aggregate. This situation produces a very typical aggregation tracing (Fig. 26-6). The optimal stir speed for any instrument is based on the height of the PRP column, the diameter of the aggregation cuvette, and the size of the stir bar used. Most aggregometer manufacturers recommend the optimal stir speed for their system. We have performed simultaneous studies on four different manufacturers’ instruments (Helena Laboratories, Beaumont, TX; Chronolog Corporation, Havertown, PA; Bio/ Data Corporation, Horsham, PA; and Payton Scientific Inc., Buffalo, NY) based on their recommended stir speed and found that the interinstrument reproducibility was excellent.

9/23/2006 11:53:48 AM

26. Platelet Aggregation


a clearly biphasic curve. At lower concentrations of ADP, fibrinogen binding is usually reversible and platelets disaggregate. Higher ADP concentrations (typically 10 or 20 µM) can mask the biphasic response elicited by the release of endogenous ADP. This is still considered a biphasic response because ADP release has occurred, but it is not evident in the aggregation tracings. Aspirin will inhibit the ADP aggregation response observed with lower concentrations of agonist, due to the inhibition of the cyclooxygenase pathway and the release of granular constituents.

B. Epinephrine Figure 26-6. Platelets in PRP exposed to 10 µM ADP with (upper tracing) and without (lower tracing) a stir bar. Without platelet–platelet contact, shape change will occur but not aggregation.

I. Time Frame of Platelet Aggregation In assessing the effect of time after venipuncture on the responsiveness of platelets, aggregation studies were carried out with three concentrations of ADP (2, 5, and 10 µM) over a 5-hour period. When the platelets were tested immediately after preparation of PRP, the responses to all three concentrations were diminished. For the higher concentrations of ADP, the response reached a stable level when tested 30 minutes after preparation of PRP (White and Jennings, unpublished observations). It took 1 hour of “resting” for stable responses to be seen with all three concentrations of agonist. This stability remained for up to 3 hours after platelet preparation and then began to diminish at the lower concentration of ADP. While the responses to higher concentrations of agonist remained relatively stable for over 4 hours (White and Jennings, unpublished observations), it is our recommendation to complete the testing of platelet aggregation within 3 hours of the time the PRP is prepared.

Epinephrine 5–10 µM is typically used in platelet aggregation testing. Epinephrine is the most erratic and unreliable of the agonists for platelet aggregation. Classically, a small first wave of response is seen, sometimes followed by a larger, full scale secondary response (Fig. 26-7). This second wave of aggregation, when present, is inhibited by aspirin (Fig. 26-8), nonsteroidal anti-inflammatory drugs (NSAIDs), antihistamines, some antibiotics, and many other prescription and over-the-counter compounds.

C. Collagen Collagen, either from bovine or equine tendon, is typically used at concentrations ranging from 1 to 5 µg/mL. Collagen is the strongest of the typical agonists used in the clinical laboratory. Collagen-induced platelet aggregation usually reflects a lag phase of approximately 1 minute, during which the platelets adhere to the collagen fibrils and undergo shape change and then release (Fig. 26-2b). The aggregation response measured is in fact the “second wave” of aggregation subsequent to platelet activation and release. At low concentrations of collagen, aspirin and other anti-platelet drugs may totally inhibit the aggregation response (Fig. 26-8).

D. Arachidonic Acid

III. Platelet Agonists A. ADP Concentrations of 1 to 10 µM ADP are typically used in the assessment of platelet aggregation. However, studies assessing the GPIIb-IIIa antagonists have primarily used 20 µM ADP. Lower ADP concentrations (1–3 µM) produce either a single (monophasic) aggregation response curve or

Ch026-P369367.indd 501

Arachidonic acid in a reaction with cyclooxygenase, is converted to thromboxane A2, a potent platelet agonist. Aspirin inhibits the cyclooxygenase pathway and the aggregation response to arachidonic acid (see Chapter 60). Subjects who have taken aspirin or other anti-platelet drugs, or who have an intrinsic release defect or Glanzmann thrombasthenia, will have abnormal arachidonic acid-induced platelet aggregation. Patients with SPD should exhibit normal arachidonic acid-induced platelet aggregation.

9/23/2006 11:53:48 AM


Lisa K. Jennings and Melanie McCabe White

Figure 26-7. Platelet aggregation and release tracings from a normal individual in response to ADP, collagen, epinephrine, and anti-CD9 mAb7 which activates via FCγRII/CD9 crosslinking. (Adapted from White and Jennings,8 Fig 3-1, page 74.)

Figure 26-8. Platelet aggregation and release tracings from an individual who had ingested aspirin the day prior to testing. Agonists used: ADP, collagen, epinephrine, and anti-CD9 mAb7 which activates via FCγRII/CD9 crosslinking. These tracings could also represent a patient with an intrinsic release defect. (Adapted from White and Jenning,8 Fig 3-2, page 75.)

E. Ristocetin In the presence of normal platelets and a normal complement of von Willebrand factor (VWF) antigen, the antibiotic ristocetin, at a concentration of 1.5 mg/mL, causes a GPIb/VWF-dependent platelet agglutination. If abnormal

Ch026-P369367.indd 502

agglutination in response to ristocetin is observed, von Willebrand disease or Bernard–Soulier syndrome (inherited lack of the GPIb-IX-V complex — the VWF receptor [see Chapters 7 and 57]) should be considered. Abnormal ristocetin-induced agglutination has also been reported in SPD.21

9/23/2006 11:53:49 AM

26. Platelet Aggregation F. Thrombin Thrombin is a very potent platelet agonist (see Chapter 9). The fact that thrombin cleaves fibrinogen and leads to the formation of a clot makes it a very difficult agonist to use for platelet aggregation testing. The synthetic peptide GlyPro-Arg-Pro (GPRP) inhibits thrombin-induced fibrin polymerization (and therefore clot formation) while allowing thrombin-induced platelet aggregation.21a Alternatively, γ-thrombin can be used for aggregation in place of αthrombin in a plasma system, but it is not readily available commercially. α-Thrombin at a concentration of 0.1 to 0.5 U/mL may be used to activate platelets in washed or gel-filtered platelet preparations.

G. TRAP Thrombin receptor activating peptide (TRAP) is a synthetic peptide that corresponds to the new N-terminal amino acid sequence of the “tethered ligand” generated after thrombin hydrolysis of the thrombin protease-activated receptor (PAR1)22 (see Chapter 9). The addition of TRAP (10 µM) to platelets elicits the very strong activation response to thrombin without the complications of fibrinogen cleavage and clot formation. Most platelet defects reflect a normal aggregation response to TRAP except for Glanzmann thrombasthenia (see Chapter 57). TRAP is now being used in platelet aggregation testing to monitor the pharmacodynamic effects of new anti-platelet drugs which block fibrinogen binding to the platelet or target the platelet PAR receptors.

IV. Trouble Shooting Because there are no quality control (QC) kits currently available for aggregometers, it is important to be aware of problems in both sample preparation and instrument calibration that can cause inaccuracy in platelet aggregation results. One common problem encountered in aggregation testing is obtaining a result of 100% or greater. It is virtually impossible for platelets to aggregate to such an extent that 100% light transmission is achieved. If the sample clots, 100% light transmission may occur, but this is not a true measure of platelet aggregation. More often the problem is in the preparation of the PPP sample. If there are residual platelets in the PPP sample, then 100% or greater aggregation may be reported. If this situation occurs, a platelet count should be performed on the PPP. If the platelet count is less than 5 × 109/L in the PPP, then there may be a problem with the calibration of the instrument. To check this, set the PRP (baseline, 0%) and PPP (100%) limits and then put an extra PPP tube into the PRP well. The instrument should reflect 100% light transmission; if it does not there is a problem in the calibration

Ch026-P369367.indd 503


of the instrument. This test can also be useful in troubleshooting two channels that are set to run duplicates but do not give answers that agree. Occasionally, instruments that are set to read against a PPP standard will not read PPP as “100%.” Unless the aggregometer manufacturer provides a calibration technique, this instrument will have to be serviced. Often interlaboratory variability in platelet aggregation testing is a result of the manner in which the results are calculated. Whether the laboratory is manually calculating the results or the instrument performs the calculations automatically, error can be introduced by methodology. If an instrument automatically calculates the answers to an aggregation assay, it is important to be sure that it is reading the maximal aggregation at actual maximum rather than at a predetermined time. If time is used to determine maximum aggregation and disaggregation has occurred, the instrument will report out a falsely low aggregation value. Slopes that are automatically calculated should be reported as change per minute and the instrument must be checked to be certain that the line it has drawn for the calculation of slope is tangential to the actual aggregation and not to shape change. For a more complete description, see the detailed protocols by White and Jennings.8

V. Medications That May Affect Platelet Aggregation Many prescription and over-the-counter medications can affect platelet function (see Chapter 58). If platelet aggregation results are not clearly indicative of any classic defect but resemble partial defects, there is a significant likelihood that the patient has ingested aspirin, thienopyridines, or NSAIDs within the past week to 10 days and has simply forgotten or neglected to relate this fact. Because it is often difficult to get a precise drug history from patients coming to the laboratory, it is very important to reconfirm a defect before making the diagnosis of an intrinsic platelet function disorder.

A. Antibiotics Antibiotics that have a β-lactam ring structure, such as the penicillins and cephalosporins, may affect platelet function. The mechanism of action is postulated to be a membrane change that blocks receptor–agonist interactions or affects Ca2+ influx.23

B. Dipyridamole Dipyridamole is a pyrimidopyrimidine that inhibits adenosine uptake into platelets, endothelial cells, and erythrocytes

9/23/2006 11:53:49 AM


Lisa K. Jennings and Melanie McCabe White

(see Chapter 63 for details). This inhibition causes an increase in the local concentration of adenosine that stimulates platelet adenylate cyclase and increases levels of cyclic 3′, 5′-adenosine monophosphate (cAMP). Elevation of cyclic AMP lowers the ability of platelets to be aggregated by platelet agonists such as platelet activating factor (PAF), collagen, and ADP.24 The overall benefit of dipyridamole has been more evident on prosthetic surfaces. This drug as an extended release formulation is used in combination with low-dose aspirin as AggrenoxTM (Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT).

is typically at least an order of magnitude greater than that considered to have potentially lethal effects in vivo. There have been reports that patients undergoing general anesthesia not only had prolonged bleeding times but also had reduced aggregation response ex vivo to weak agonists. Although there are conflicting reports, data suggest that halothane, sevoflane, and propofol inhibit platelet function signaling pathways at clinically relevant concentrations.29,30

F. Thrombin Inhibitors C. Fibrinolytics Fibrinolysis and the formation of fibrin degradation products (FDPs) have been associated with a decrease in platelet aggregation. FDPs compete with fibrinogen for binding to the platelet membrane and interfere with platelet aggregation. One recent study25 has shown that subjects treated with tenecteplase and alteplase, two newer fibrinolytic compounds, exhibited significant inhibition of platelet aggregation when tested in whole blood, traditional PRP aggregation systems, or point-of-care analyzers. In another study26 comparing reteplase, alteplase, and streptokinase, inhibition of platelet aggregation was observed with all three treatments. The decrease in aggregation was most pronounced with streptokinase, followed by reteplase and then alteplase. Decreased levels of plasma fibrinogen and impaired binding of fibrinogen to GPIIb-IIIa correlated with the severity of the platelet aggregation defect with these agents.

D. Dextran Intravenous infusion of dextran can result in reduced platelet function. In peripheral arterial disease patients, it was recently shown that Dextran 40 reduced spontaneous and agonist-mediated platelet aggregation and the expression of activation markers such as P-selectin on the platelet surface.27

E. Anesthetics Anesthetics have been demonstrated to have an effect on the aggregation response of platelets and have been implicated in an increased risk of hemorrhagic complications.29,30 Anesthetics such as lidocaine, dibucaine, cocaine, etc. have a direct effect on the platelet membrane. Cocaine added to platelets in vitro causes a reduction in fibrinogen binding to the activated GPIIb-IIIa receptor.28 The concentration at which these anesthetics mediate a platelet inhibitory effect

Ch026-P369367.indd 504

Thrombin is a key regulator in the pathophysiology of acute coronary syndromes. It mediates the conversion of fibrinogen to fibrin, activates factor XIII that aids in clot stabilization, and is a potent agonist of platelets. Heparin is the most used clinical anti-thrombin, but it has many limitations with regard to unpredictable bioavailability and safety. The recent generation of direct thrombin inhibitors that act independently of antithrombin III can inhibit clotbound thrombin as well as thrombin-induced platelet activation.31

G. Thienopyridines ADP is a key platelet agonist that functions by binding to G protein coupled receptors, P2Y1 and P2Y12 (see Chapter 10). The P2Y12 receptor is the primary ADP receptor that mediates fibrinogen binding and sustained aggregation response.32 The thienopyridines, ticlopidine and clopidogrel, irreversibly and covalently bind to this receptor and inhibit platelet aggregation by ADP on the order of 40 to 60% (see Chapter 61).33 Platelet aggregation to other agonists is not inhibited but may be attenuated because released ADP contributes to the platelet aggregation response, particularly in conjunction with weak agonists.

H. GPIIb-IIIa Antagonists GPIIb-IIIa antagonists bind to the GPIIb-IIIa (integrin αIIbβ3) receptor and prevent the binding of fibrinogen or VWF to the activated platelet (see Chapter 62).14,34 These agents, eptifibatide, abciximab, and tirofiban, are the most potent of the anti-platelet agents because when bound to GPIIb-IIIa, platelet aggregation to all agonists (e.g., ADP, collagen, TRAP) is inhibited significantly. Typically, the level of platelet inhibition by these agents is tested ex vivo using the PPACK anticoagulant and 20 µM ADP.35 Many other drugs have been identified that alter normal platelet function (see Chapter 58 for a comprehensive review).

9/23/2006 11:53:49 AM

26. Platelet Aggregation

VI. Inherited Platelet Function Defects


and thrombin (Fig. 26-10), and an abnormal clot retraction.

A. Storage Pool Disease In contrast to normal platelet aggregation (Fig. 26-7), SPD and release defects are usually distinguished by an abnormal second wave of aggregation secondary to the absent or decreased release of adenine nucleotides from platelet dense granules (Figs. 26-8 and 26-9). However, SPD may have normal platelet aggregation.6,7 The diagnosis of these patients is made using studies of adenine nucleotide release. SPD and release defects are discussed in detail in Chapters 15 and 57.

C. von Willebrand Disease Patients with von Willebrand disease exhibit bleeding similar to those with intrinsic platelet defects. The aggregation responses in these individuals are normal to all agonists with the exception of ristocetin. In the rare type IIb and platelet-type von Willebrand disease, there is increased ristocetin-induced platelet aggregation.38

D. Bernard–Soulier Syndrome B. Glanzmann Thrombasthenia 36

Patients with Glanzmann thrombasthenia have a mutation in the GPIIb-IIIa receptor resulting in either the absence or deficiency of functional GPIIb-IIIa (see Chapter 57 for details). While the absence of GPIIb-IIIa yields a definitive diagnosis of Glanzmann thrombasthenia, variants have been described in which platelet GPIIb-IIIa is present, but a mutation renders it nonfunctional.37 The diagnostic features are: absent platelet aggregation to ADP, collagen, epinephrine,

Bernard–Soulier syndrome39 patients have a deficiency of GPIb-IX-V which renders their platelets incapable of interacting with VWF (see Chapter 57 for details). This leads to a problem with platelet adhesion to injury-induced exposure of subendothelium. The platelet aggregation responses seen in Bernard–Soulier syndrome are normal to all agonists but ristocetin. If measuring only platelet aggregation in the diagnosis of a patient with a bleeding disorder, Bernard– Soulier could be mistaken for von Willebrand disease.

Figure 26-9. Platelet aggregation and release tracings from an individual with storage pool disease, in response to ADP, collagen, epinephrine, and anti-CD9 mAb7 which activates via FCγRII/CD9 crosslinking. Note that, compared to the patient with release defect (aspirin, Fig. 26-8), this patient exhibits a markedly decreased release in response to mAb7. (Adapted from White and Jennings,8 Fig 3-3, page 76.)

Ch026-P369367.indd 505

9/23/2006 11:53:49 AM


Lisa K. Jennings and Melanie McCabe White

Figure 26-10. Platelet aggregation and release tracings from a patient with Glanzmann thrombasthenia. Agonists used: ADP, collagen, epinephrine, and anti-CD9 mAb7 which activates vial FCγRII/CD9 crosslinking. Due to the lack of GPIIb-IIIa expression, platelets are unable to bind fibrinogen and aggregate. Note that there is release in response to strong agonists, showing that, although the platelets are incapable of aggregating, activation can occur. (Adapted from White and Jennings,8 Fig 3-4, page 77.)

However, flow cytometric analysis of GPIb-IX-V surface density is able to distinguish between these disorders (see Chapter 30).

VII. Acquired Platelet Function Defects Acquired platelet function defects occur in uremia, preleukemia and acute leukemias, myeloproliferative disorders, dysproteinemias, liver disease, and anti-platelet antibodies. These disorders are discussed in detail in Chapter 58.

References 1. Born, G. V. R. (1962). Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature, 194, 927–929. 2. O’Brien, J. R. (1962). Platelet aggregation: Part II. Some results of a new method. J Clin Path, 15, 452–455. 3. Holmsen, H., Holmsen, I., & Bernhardsen, A. (1966). Microdetermination of adenosine diphosphate and adenosine triphosphate in plasma with the firefly luciferase system. Anal Biochem, 17, 456–473. 4. Feinmann, R. D., Kubowsky, J., Charo, I., et al. (1977). The lumiaggregometer: A new instrument for simultaneous measurement of secretion and aggregation by platelets. J Lab Clin Med, 90, 125–129.

Ch026-P369367.indd 506

5. White, M. M., Foust, J. T., Mauer, A. M., et al. (1992). Assessment of lumiaggregometry for research and clinical laboratories. Thromb Haemost, 67, 572–577. 6. Nieuwenhuis, H. K., Akkerman, J.-W. N., & Sixma, J. J. (1987). Patients with prolonged bleeding time and normal aggregation tests may have storage pool deficiency: Studies on 106 patients. Blood, 70, 620–623. 7. Israels, S. J., McNicol, A., Robertson, C., et al. (1990). Platelet storage pool deficiency: Diagnosis in patients with prolonged bleeding times and normal platelet aggregation. Br J Haematol, 75, 118–121. 8. White, M. M., & Jennings, L. K. (1999). Platelet protocols: Research and clinical laboratory procedures. San Diego: Academic Press. 9. Ingeman-Wojenski, C., Smith J. B., & Silver, M. J. (1983). Evaluation of electrical aggregometry: Comparison with optical aggregometry, secretion of ATP, and accumulation of radiolabeled platelets. J Lab Clin Med, 101, 44–52. 10. Podczasy, J. J., Lee, J., & Vucenik, I. (1997). Evaluation of whole-blood lumiaggregation. Clin Appl Thromb Hem, 3, 190–195. 11. Steinhubl, S. R., Talley, J. D., Braden, G. A., et al. (2001). Point of care measured platelet inhibition correlates with a reduced risk of an adverse cardiac event after percutaneous coronary intervention. Circulation, 103, 2572– 2578. 12. Kereiakes, D. J., Broderick, T. M., Roth, E. M., et al. (1999). Time course, magnitude and consistency of platelet inhibition by abciximab, tirofiban, or eptifibatide in patients with unsta-

9/23/2006 11:53:50 AM

26. Platelet Aggregation ble angina pectoris undergoing percutaneous coronary intervention. Am J Cardiol, 84, 391–395. 13. White, M. M., Krishnan, R., Kueter, T. J., et al. (2005). The use of the point of care Helena ICHOR/Plateletworks and the Accumetrics Ultegra RPFA for assessment of platelet function with GPIIb-IIIa antagonists. J Thromb Thrombolysis, 18, 163– 169. 14. Jennings, L. K., Jacoski, M. V., & White, M. M. (2002). The pharmacodynamics of parenteral glycoprotein IIb-IIIa Inhibitors. J Intervent Cardiol, 15, 45–60. 15. Tardiff, B. E., Jennings, L. K., Harrington, R. A., et al. (2001). Pharmacodynamics and pharmacokinetics of eptifibatide in patients with acute coronary syndromes: Prospective analysis from the Pursuit Trial. (Tardiff and Jennings are co-first authors.) Circulation, 104, 399–405. 16. Triplett, D. A., Harms, C. S., Newhouse, P., et al. (1978). Platelet function: Laboratory evaluation and clinical application. Chicago: American Society of Clinical Pathologists. 17. Hardisty, R. M., Hutton, R. A., Montgomery, D., et al. (1970). Secondary platelet aggregation: A quantitative study. Br J Haematol, 19, 307–319. 18. Phillips, D. R., Teng, W. S., Arfsten, A., et al. (1997). Effect of Ca ++ on Integrilin GPIIb-IIIa interactions: Enhanced GPIIb-IIIa binding and inhibition of platelet aggregation by reductions in the concentration of ionized calcium in plasma anticoagulated with citrate. Circulation, 96, 1488– 1494. 19. Pignatelli, P., Pulcinelli, F. M., Ciatti, F., et al. (1995). Acid citrate dextrose (ACD) formula A as a new anticoagulant in the measurement of in vitro platelet aggregation. J Lab Clin Anal, 9, 138–140. 20. Luxembourg, M. H., Klaffling, C., Erbe, M., et al. (2005). Use of native or platelet count adjusted platelet rich plasma for platelet aggregation measurements. J Clin Pathol, 58, 747–750. 21. Pujol-Moix, N., Hernandez, A., Escolar, G., et al. (2000). Platelet ultrastructural morphometry for diagnosis of partial δ-storage pool disease in patients with mild platelet dysfunction and/or thrombocytopenia of unknown origin. A study of 24 cases. Haematologica, 85, 619–626. 21a. Michelson, A. D., Ellis, P., Barnard, M. R., et al. (1991). Downregulation of the platelet surface glycoprotein Ib-IX complex in whole blood stimulated by thrombin, ADP or an in vivo wound. Blood, 77, 770–779. 22.Coughlin, S. R. (2000). Thrombin signaling and protease activated receptors. Nature, 407, 258–264. 23. Burroughs, S. F., & Johnson, G. J. (1990). Beta-lactam antibiotic-induced platelet dysfunction: Evidence for irreversible inhibition of platelet activation in vitro and in vivo after prolonged exposure to penicillin. Blood, 75, 1473– 1480.

Ch026-P369367.indd 507


24. Philp, R. B., & Lemieux, J. P. V. (1975). Interactions of dipyridamole and adenosine on platelet aggregation. Nature, 221, 1162–1164. 25. Serebruany, V. L., Malinin, A. I., Callahan, K. P., et al. (2003). Effect of tenecteplase versus alteplase on platelets during the first 3 hours of treatment for acute myocardial infarction: The Assessment of the Safety and Efficacy of a New Thrombolytic Agent (ASSENT-2) platelet substudy. Am Heart J, 145, 636–642. 26. Moser, M., Nordt, T., Peter, K., et al. (1999). Platelet function during and after thrombolytic therapy for acute MI with reteplase, alteplase or streptokinase. Circulation, 100, 1858– 1864. 27. Robless, P., Okonko, D., Milhailidis, D. P., et al. (2004). Dextran 40 also reduces in vitro platelet aggregation in peripheral arterial disease. Platelets, 15, 215–222. 28. Jennings, L. K., White, M. M., Sauer, C. M., et al. (1993). Cocaine-induced platelet defects. Stroke, 24, 1352–1359. 29. Kozek-Langenecker, S. A. (2002). The effects of drugs used in anaesthesia on platelet membrane receptors and on platelet function. Curr Drug Targets, 3, 247–258. 30. Dordoni, P. L., Frassanito, L., Bruno, M. F., et al. (2004). In vivo and in vitro effects of different anaesthetics on platelet function. Br J Haematol, 125, 79–82. 31. Chen, M. S., & Lincoff, A. M. (2005). Direct thrombin inhibitors. Curr Cardiol Rep, 7, 355–359. 32. Conley, P. B., & Delaney, S. M. (2003). Scientific and therapeutic insights into the role of the platelet P2Y12 receptor in thrombosis. Curr Opin Hematol, 10, 333–338. 33. Gurbel, P. A., & Bliden, K. P. (2003). Durability of platelet inhibition by clopidogrel. Am J Cardiol, 91, 1123–1125. 34. Jennings, L. K. (2001). Clinical trials of the parenteral glycoprotein IIb-IIIa inhibitors: The influence of dose selection on results. A CME Booklet, Health Science Communications. 35. Saucedo, J. F., Wolford, D. C., Cook, S. L., et al. (2004). Comparative pharmacodynamic evaluation of eptifibatide and tirofiban HCl in patients undergoing coronary intervention— The TAM1 Study. Am J Cardiol, 93, 1270–1282. 36. Glanzmann, E. (1918). Heredititare hamorrhagische thrombasthenic: Ein Beitray zur pathologic der blutplatchen. Jahrbuch fur Kinderheilkunde, 88, 113. 37. Jackson, D. E., White, M. M., Jennings, L. K., et al. (1988). A Ser162-Leu mutation within glycoprotein (GP) IIIa (integrin β3) results in an unstable αIIbβ3 complex that retains partial function in a novel form of Type II Glanzmann’s thrombasthenia. Thromb Haemost, 80, 42–48. 38. Schneppenheim, R. (2005). Blood Coagul Fibrinolysis, Suppl 1, S3–S10. 39. Bernard, J., & Soulier, J. P.(1948). Sur une nouvelle variete de dystrophie thrombocytaire hemorragipare congenitale. Sem Hop Paris, 24, 3217.

9/23/2006 11:53:50 AM