Altered Platelet Function Detected by Flow Cytometry
Effects of Coronary Artery Disease and Age
Abstract Platelet activation state and responsiveness to physiological agonists were measured in 65 patients with documented coronary artery disease (54 male and 11 female; mean age, 58 years). Twelve patients (mean age, 52 years), selected at random from the male cohort, were compared with 12 age-matched male control subjects (mean age, 52 years) and with 10 normal, young male subjects (mean age, 25 years). Whole-blood flow cytometry was used to measure platelet activation status ex vivo and platelet responsiveness to physiological agonists in vitro. Peripheral blood samples were analyzed for bound fibrinogen and expression of P-selectin, GPIb, and GPIIb-IIIa at rest and in response to ADP (0.1 to 10 μmol/L) and thrombin (0.02 to 0.32 μ/mL). No significant differences were seen in the basal levels of fibrinogen binding between any of the groups, but P-selectin expression was significantly lower in patients compared with age-matched control subjects (P=.0005). When stimulated with agonists, patients’ platelets had significantly decreased fibrinogen binding (P<.03) but no difference in P-selectin expression compared with the age-matched group. Both agonist-induced fibrinogen binding and P-selectin expression were, however, higher in the young subjects compared with either the older control group or the patients (P<.05). GPIb and GPIIb-IIIa expression were lowest in the patients with angina and highest in the young control subjects, with levels in the age-matched control subjects falling between these values. Data from the total patient cohort (n=65) were identical to those in the smaller cohort (n=12). In conclusion, atherosclerosis impairs platelet aggregatory responses (fibrinogen binding) over and above the decreased response seen with age. Platelet degranulation (P-selectin expression) is also impaired in patients with coronary artery disease, but only in comparison with younger subjects, not age-matched controls.
- Received June 26, 1996.
- Accepted November 28, 1996.
Platelets are thought to play a key role in atherogenesis and the development of the thrombotic complications of atheroma,1 2 and antiplatelet therapy is effective in limiting complications in patients with vascular disease3 and angina.4 Despite this, it remains unclear whether platelet behavior is altered in patients with stable atherosclerotic disease. A variety of platelet abnormalities have been described in patients with stable angina. These include an increased tendency to aggregation,5 6 7 8 enhanced prostaglandin generation and abnormal sensitivity to prostacyclin,9 increased TxB2 production,10 decreased platelet survival,11 12 increased levels of βTG, FPA,13 14 15 and PF4,16 but these have not been consistent findings.17 18 19 20 21 22 23 24 Furthermore, in two large epidemiological studies,25 26 angina was not associated with increased platelet aggregation.
To a large extent these conflicting results reflect the methodological problems of measuring platelet activation in vivo and the inadequacy of isolated in vitro tests of platelet function. Earlier studies relied on indirect measurement of the consequences of platelet activation in vivo, using aggregometry or the measurement of platelet release products. Direct assessment of the activation status of individual platelets can be measured by the sensitive technique of fluorescence-activated flow cytometry, which uses fluorescently labeled antibodies to detect the expression of activation antigens on the platelet surface.27 Activation causes a conformational change in the platelet membrane GPIIb-IIIa complex, exposing the receptor site for fibrinogen and leading to binding of plasma fibrinogen, which can be recognized with specific antibodies.28 29 30 Platelet degranulation is accompanied by translocation of granule membrane to the platelet surface, bringing with them specific membrane glycoproteins such as alpha granule P-selectin,31 which appear as neoantigens. Activation of platelets in vitro can also lead to increased numbers of GPIIb-IIIa complexes and reduced numbers of GPIb-IX complexes on the cell surface.32 Whole-blood flow cytometry of unfixed blood also allows the responsiveness of the platelets to ex vivo stimulation with platelet agonists such as thrombin and ADP to be studied in the presence of other blood components. Thus, in addition to detecting changes in the basal activation state of platelets, the technique can also detect more subtle alterations in platelet behavior.
We used flow cytometry to investigate platelet function in patients with documented coronary disease and stable angina pectoris, comparing these with both healthy, young subjects and age-matched control subjects.
Subjects and Patients
Platelet function was studied in three groups: patients with stable angina awaiting angioplasty and two groups of healthy male control subjects, one group of age-matched subjects and another group of younger subjects.
Sixty-five patients (54 male and 11 female) with stable angina were studied. All had limiting angina and were admitted for elective angioplasty. Patients were between 39 and 79 years of age (mean, 58 years). All subjects received aspirin (150 mg) and a β-blocker (atenolol 50 mg or bisoprolol 10 mg OD) as their only antianginal medication. Thirty of the patients (45%) had single-vessel, 27 (42%) had double-vessel, and 8 (13%) had triple-vessel disease. Additional clinical and demographic details (clinical presentation, previous myocardial infarction and revascularization, diabetes, smoking, impaired left ventricular function, and lipid profile) are given in Table 1⇓. Patients were studied after fasting, on the morning of angioplasty, before taking their morning medication.
Twenty-two healthy male volunteers participated in the study. Twelve male subjects were age-matched to 12 of the patients with angina, who were selected at random from the male cohort. The age range of these subjects and the paired angina patients was 40 to 60 years (mean, 52 years). Ten younger subjects between 21 and 29 years of age (mean, 25 years) were also studied. All control subjects underwent a pretrial examination, including electrocardiography, to verify health status. All had normal resting blood pressure and normal results on resting electrocardiograms with no history of, or risk factors for, coronary artery disease. No subject had taken any platelet-active drug during the 2 weeks preceding the study period, and none of the healthy subjects smoked.
All subjects and patients gave written consent to participate in this study, which was approved by the Ethical Practices Committees of the Royal Brompton and Royal Free Hospitals.
All patients and control subjects rested supine for 30 minutes before collection of blood. All blood samples were obtained by clean venipuncture, using minimal stasis, via a 21-gauge butterfly needle into Monovette tubes; conditions designed to prevent artifactual activation of platelets during phlebotomy. On each occasion, the first 2.7 mL of blood was drawn into an EDTA tube and used to obtain blood counts. The next 5 mL was aspirated into a 1/10-vol of 0.106 mol/L trisodium citrate and used, within 10 minutes, for the flow cytometric analysis. A subsequent 5 mL was drawn into an ice-cold citrate tube and immediately centrifuged at 4°C for 30 minutes at 3000×g, and plasma was taken from the center of the tube and frozen in aliquots at −70°C until analyzed for plasma fibrinogen and soluble P-selectin. A final 10 mL of blood was collected into a serum/gel tube for analysis of cholesterol and lipoproteins.
Measurement of Platelet Volume
EDTA samples were stored at room temperature for 2 hours before routine measurement of red and white blood cell count, hemoglobin, and MPV in a Coulter Stacks S blood analyzer. Measurement of MPV has been shown to be stable in EDTA samples under these conditions.33
Plasma and Serum Assays
Cholesterol was measured by a standard enzymatic method (cholesterol oxidase). HDL cholesterol was measured after first precipitating the apolipoprotein B-containing fractions with phosphotungstic acid/MgCl2 reagent, and LDL-cholesterol was calculated using the Friedewald formula.34 Lp(a) was measured by ELISA.
Plasma fibrinogen was determined by the Clauss method on a KC10 coagulometer. sP-selectin was measured by ELISA.
Flow Cytometric Analysis of Platelet Activation
Platelets were identified with a MAb to GPIbα (RFGP37), and GPIIb-IIIa was identified with a CD41 MAb (RFGP56), both raised in our laboratory.35 The MAbs were purified from ascites on protein G-Sepharose and coupled to FITC using standard techniques.36 Platelet-bound fibrinogen was detected with an FITC-conjugated polyclonal antibody to human fibrinogen (Rαfgn-FITC), as described by Warkentin et al29 and Janes et al.30 P-selectin was identified with a FITC-conjugated immunoglobulin G1 mouse MAb. All antibodies were used at optimum concentration for maximum fluorescence with minimum nonspecific binding, determined by titration.
ADP and human α-thrombin were purchased from Sigma Chemical Company Ltd. Samples incubated with thrombin also contained 0.125 mmol/L glycyl-prolyl-l-arginyl-l-proline peptide to inhibit fibrin polymerization and consequent clot formation.37 The dilution buffer used was HEPES-buffered saline (145 mmol/L NaCl, 5 mmol/L KCl, 1 mmol/L MgSO4, and 10 mmol/L HEPES, pH 7.4) that had been passed through a 0.22-μm filter to remove dust particles.
Blood samples were prepared for flow cytometric analysis using the whole-blood method essentially as described by Janes et al.30 Five microliters of citrated blood was added to LP3 tubes containing 50 μL of HEPES-buffered saline and 5 μL of appropriate concentrations of antibodies and agonists. After gentle mixing, the samples were incubated for 20 minutes, then diluted with 0.5 mL of 0.2% (vol/vol) formyl saline to inhibit further activation. Incubations were carried out at room temperature (20 to 22°C), and all samples were run in duplicate. Fibrinogen binding was measured in response to ADP (0.1 to 10 μmol/L) and thrombin (0.02 to 0.32 μ/mL). P-selectin expression was measured in response to maximal ADP stimulation (10 μmol/L) and to thrombin (0.02 to 0.32 μ/mL).
Samples were analyzed, within 2 hours of collection, in a Coulter EPICS Profile II flow cytometer. The instrument was aligned daily with Immuno Check and Standard Brite beads to calibrate the light scatter and fluorescence parameters, respectively. The platelet population was identified from its light scatter characteristics, and its identity was confirmed using the anti-GPIbα MAb. An electronic bit map was set around the platelet population and adjusted for each sample to ensure that >98% of the particles analyzed were positive for GPIb. The negative cutoff levels for fluorescence in unstimulated samples were set at 2%.30 Five thousand platelets were analyzed, and the results represent the means of duplicate samples. The percentage of platelets positive for the marker and the mean fluorescence intensity for each sample were used to calculate the binding index for the marker from the following equation: binding index=(percent positive×mean fluorescence intensity)×100−1.
All data are shown as the mean±SD. However, to allow for the possibility that data were not normally distributed, the relationship between independent variables was analyzed using nonparametric (Wilcoxon) tests. Correlation between variables was plotted using Excel software and found, in all cases, to be either linear or random. Correlation coefficients were calculated using Pearson rank sum correlation.
Hemoglobin and blood cell counts were normal in all patients and subjects studied. Platelet counts were normal in all patients and subjects and were not significantly different between any of the groups. Table 2⇓ shows the platelet counts, MPV, and plasma fibrinogen for the control groups and the patients with angina, with the latter presented in two groups, the subgroup of 12 angina patients randomly selected for age-matching to the healthy control subjects and the total patient group (n=65). MPV was similar in the patient groups and the age-matched control subjects but was significantly higher (P<.001) in the younger subjects compared with all the older groups. Plasma fibrinogen was within the normal range (2 to 5 g/L) in all subjects (with the exception of one angina patient who had a level of 5.1 g/L) but was higher in the patients compared with both the age-matched and the younger control subjects (P<.05).
Flow Cytometric Analysis
Baseline Platelet Activation Status
Table 3⇓ shows the level of fibrinogen binding and expression of P-selectin, GPIbα, and GPIIb-IIIa in the different groups. There was no significant difference between the main group of patients and the subgroup in expression of any antigen in the unstimulated samples. There was no significant difference in fibrinogen binding in unstimulated samples between any of the groups. However, the patients with angina had significantly lower expression of P-selectin compared with the age-matched control subjects (P=.0005). There was no significant difference in P-selectin expression between angina patients and young control subjects or between the control groups.
GPIb and GPIIb-IIIa expression were lowest in the angina patients and highest in the young control subjects, with levels in the age-matched control subjects falling between these values. These differences were significant for both antigens between the angina patients and the young control subjects (P=.005), but only GPIbα expression was significantly lower between the angina patients and the age-matched control subjects (P=.001), and only GPIIb-IIIa expression was significantly different between the two control groups.
The differences in GPIb and GPIIb-IIIa expression between the young cohort of subjects and the other three groups could relate to differences in platelet volume. However, although expression of GPIb per platelet was correlated with MPV in all subjects (r=.433; P=.0001), this was not so for GPIIb-IIIa (P>.3). In addition, differences in platelet volume cannot account for the differences seen between the patients and age-matched control subjects because the platelet volume was no different between these groups.
Platelet Responsiveness to Agonist Stimulation
Fibrinogen binding was studied in response to stimulation with ADP (0.1 to 10 μmol/L) and thrombin (0.02 to 0.32 μ/mL).
Fig 1⇓ shows dose-response curves for fibrinogen binding in response to thrombin stimulation. Fig 1⇓, a, shows data from the subgroup of 12 angina patients and the age-matched control subjects together with data from the young control group. Fibrinogen binding was highest in the young control subjects and lowest in the angina patients, with the curve for the age-matched control subjects lying in between. The difference between the angina patients and the control group attained statistical significance (P<.03), but the differences between the two control groups did not. Fibrinogen binding in the subgroup of 12 angina patients selected for age-matching was identical to that seen in the total patient population (Fig 1⇓, b).
A very similar pattern was seen for fibrinogen binding in response to ADP (Fig 2⇓). Again, there was a statistically significant difference between the angina patients and the age-matched control group (P=.02) but not between the two control groups (Fig 2⇓, a). The subgroup of 12 angina patients did not differ from the main patient group (Fig 2⇓, b).
Within the patient group, fibrinogen binding in response to ADP and thrombin stimulation showed an inverse relationship with age, but this was significant only with the higher concentrations of thrombin (eg, 0.32 μ/mL thrombin; r=−.442, P=.0004).
P-selectin expression was studied in response to maximal ADP stimulation (10 μmol/L) and to thrombin (0.02 to 0.32 μ/mL) stimulation.
A different pattern was observed from that of fibrinogen binding. P-selectin expression in response to thrombin stimulation was no different between the patients and the age-matched control group (Fig 3⇓, a). However, both older groups showed significantly reduced P-selectin expression compared with the young control subjects (P=.0006). As before, there was no significant difference between the subgroup of 12 angina patients and the main cohort (Fig 3⇓, b).
A similar trend was observed for P-selectin expression in response to maximal ADP stimulation (Fig 4⇓), with higher levels in the young subjects compared with all other groups and similar levels of expression between the older control subjects and the patients. However, these differences did not reach statistical significance (P>.07). Within the patient group, P-selectin expression did not correlate with age.
Measurement of sP-Selectin
Levels of sP-selectin were no different in the plasma from the patients (56.7±3.9; range, 20.8 to 110.6 mg/L) compared with that from the age-matched control subjects (53.0±8.0; range, 37.8 to 74.4 mg/L).
Effect of Variables
Platelet Count and Plasma Fibrinogen
It has been shown 30 that platelet fibrinogen binding is not influenced by platelet count or the level of plasma fibrinogen (both factors that could influence the assay) within the normal range for these parameters. There was little variation in platelet count within any of the groups or between the groups (Table 2⇑), and there was no consistent correlation between platelet count or plasma fibrinogen and fibrinogen binding. For illustration, in the main patient cohort, platelet count showed an inverse correlation with fibrinogen binding in response to ADP (10 μmol/L; r=−.300, P=.008) but not in response to thrombin. Similarly, although in this group there was a degree of inverse correlation between plasma fibrinogen levels and fibrinogen binding in response to thrombin (0.32 μ/mL; r=−.404, P=.003), no correlations were seen with other stimuli. In addition, similar correlation was observed between thrombin-induced P-selectin expression and plasma fibrinogen (0.32 μ/mL thrombin; r=−.285, P=.05).
Plasma cholesterol was significantly higher in the patients compared with the age-matched control subjects (P<.05), but no correlations between plasma cholesterol (expressed as either total cholesterol or LDL cholesterol) and platelet activation were seen within the large patient cohort. Similarly, there were no statistically significant differences in platelet activation between patients with total cholesterol <6.5 mmol/L and those with total cholesterol >6.5 mmol/L (Table 4⇓) .
Lp(a) levels of >0.3 g/L were seen in 24 (38%) of the patients, but again, there was no relationship between platelet activation markers and Lp(a) (Table 4⇑).
There was a tendency for platelet activation and platelet agonist responsiveness to be lower in patients who smoked than in nonsmokers or ex-smokers (data not shown). However, these differences were not consistent and were not statistically significant (P>.05).
Extent of Disease
Patients with two or three diseased vessels had lower levels of fibrinogen binding than those with single-vessel disease, and these differences were significant in the resting and ADP-stimulated samples (P<.05) but not in those stimulated with thrombin (Table 5⇓). No differences were seen, however, in P-selectin, GPIbα or GPIIb-IIIa expression between these groups (Table 5⇓). Patients with a history of myocardial infarction (n=20) showed no overall difference in platelet activation or responsiveness from the overall patient cohort (data not shown).
We showed that patients with coronary artery disease have abnormalities of platelet activation and function that are measurable by whole-blood flow cytometry. In contrast to some earlier studies using different methodologies, we found that fibrinogen binding (a precursor to aggregation) in response to stimulation with physiological agonists is impaired in patients with coronary artery disease compared with age-matched, healthy control subjects. These differences were over and above the reduced binding seen with age in healthy subjects. In contrast, there was no difference in agonist-induced P-selectin expression (a marker for platelet degranulation) in the patients compared with their age-matched, healthy counterparts, but a marked decrease was seen with age.
Formation of intracoronary thrombus, consequent upon plaque rupture, is thought to be the initiating event in the transition from stable to unstable coronary artery disease. Platelets might be expected to be activated in this situation, and decreased platelet prostacyclin binding,38 increased TxA2-prostaglandin H2 platelet receptor numbers,39 increased TxA2 biosynthesis,23 and increased βTG and PF4 levels40 have all been reported in patients with unstable angina. The platelet abnormalities described in the acute coronary syndromes may be transient consequences of plaque rupture, but it is also possible that atherosclerosis induces chronic alterations in platelet reactivity, which may precede, and contribute to, the triggering of acute coronary syndromes. No clear consensus on platelet behavior in patients with stable, as opposed to unstable, angina has been reached from existing studies.
Enhanced platelet aggregation to ADP has been reported using a variety of techniques in studies of small numbers of patients,5 6 7 and more recently, a study of 78 patients with coronary artery disease8 showed a marked increase in in vitro aggregation to ADP and collagen when the patients were compared with 113 healthy subjects. Abnormal platelet aggregation was not related in this study to either smoking, increased plasma cholesterol, or hypertension.
In contrast, other studies have found no difference in ADP-induced platelet aggregation in patients with stable angina compared with age- and sex-matched control subjects.17 18 19 Similarly, circulating platelet aggregates, although increased in patients with myocardial infarction, were not increased in patients with stable angina,20 21 and no differences were found in circulating platelet microaggregates from the aorta and pulmonary artery in 41 patients with coronary artery disease undergoing cardiac catheterization compared with 17 subjects with normal coronary arteries.22
Measurement of platelet-release products in patients with stable angina show similar inconsistencies. Small but significant increases in levels of βTG13 14 and PF416 have been reported in patients with stable angina and were related to the extent of vessel disease,14 but in general the levels were within14 or overlapped13 the normal range.
There is good evidence41 42 that patients with myocardial infarction have larger, younger platelets than normal, probably as a result of increased platelet consumption at the atherosclerotic plaque, and such platelets are thought to have heightened reactivity. In contrast to the findings in myocardial infarction, a recent large study43 in 426 patients with stable angina awaiting coronary artery bypass grafting found no difference in MPV compared with non-age-matched control subjects. This disparity between findings in patients with myocardial infarction and those with stable angina is reflected in large epidemiological studies. In the Caerphilly Collaborative Heart Disease Study,25 investigators studied 1811 men and found platelet aggregation to be associated with electrocardiographic evidence of ischemia and history of myocardial infarction but not angina. Thaulow et al,26 despite finding a link between mortality and platelet count (487 men) and platelet aggregation (150 men) over 13.5 years of follow-up, found no relation between platelet count and the development of angina or positive results on exercise testing. A further study44 in 958 individuals showed a nonsignificant increase in aggregability in patients with a history of ischemic heart disease and abnormal results on electrocardiography.
We found a marked decrease in platelet agonist responsiveness, in particular a decreased degranulation response, in the older subjects. This contrasts with previous reports that suggest increased agonist-induced aggregation in platelet-rich plasma45 46 and increased levels of βTG and PF447 in the plasma of older individuals.46 However, Reilly and FitzGerald48 found no increase in βTG, circulating platelet aggregates, or response to arachidonic acid with age despite an increase in Tx excretion. The subjects in the above studies were considerably older than the patients and age-matched control subjects in our study, so the findings may not be directly comparable. It is clear, however, that currently there is little consensus on the effects of advancing age on platelet function. Although we saw no direct statistical correlation between patient age and any of the platelet parameters measured, this may reflect the relatively small age range within this middle-aged population. In addition, the wide prevalence of atherosclerosis in middle-aged men makes it difficult to separate the effects of age from those of disease. The age-matched control subjects were selected because they had a low risk of coronary artery disease, but the majority of such middle-aged men could still be expected to have some clinically insignificant atheroma. It is therefore possible that the differences in vascular disease between the age-matched control subjects and the patients may be relative rather than absolute.
Thus, although studies using a number of techniques have indicated platelet abnormalities in patients with myocardial infarction and acute coronary syndromes, a definitive picture of platelet behavior in stable coronary disease and in older individuals has not emerged. In part this is related to age differences in the patient and control groups and to the different phlebotomy procedures used for patients and control subjects in some studies. Phlebotomy was carefully controlled in the present study; all subjects were sampled under the same conditions, and phlebotomy was performed by the same individual throughout the study.
The lack of consensus reached by investigators also reflects the methodological problems of assessing platelet function in vivo. Most studies use indirect methods to detect platelet activation, many of which are highly susceptible to artifacts induced by phlebotomy and processing. Furthermore, testing of platelet responsiveness in vitro, when platelets are isolated from other blood components, may have limited relevance to the situation in vivo. By using the sensitive technique of fluorescence-activated flow cytometry in whole, unfixed blood, the activation status of individual platelets can be assessed directly. In addition, the technique allows measurement of platelet responsiveness to agonists such as thrombin and ADP to be studied in the presence of other blood components, which may detect more subtle alterations in platelet behavior. These techniques have previously been found to detect activated and hyperresponsive platelets in other clinical conditions.49
Levels of GPIb and GPIIb-IIIa per platelet were significantly lower in patients with angina and in older subjects. Total expression of these antigens has been shown, in certain experimental situations, to alter on activation, with GPIIb-IIIa increasing by translocation from intracellular stores and GPIb decreasing, probably by internalization induced to the platelet cytoskeleton following activation.50 51 The changes seen in the present study do not resemble those seen in stimulated platelets in vitro and probably do not reflect activation in vivo.
Flow cytometry detects the fluorescence intensity of each platelet, which reflects both the surface receptor density and the size of the platelets. Platelet volume was lower in all the older subjects, and this was reflected in lower expression of GPIb and, although not attaining statistical significance, lower expression of GPIIb-IIIa per platelet. Platelet volume differences could not, however, account for the differences in GPIb, GPIIb-IIIa, and fibrinogen binding between the patients and the age-matched control subjects because MPV did not differ between these groups. Measurement of MPV was carefully controlled, and all samples were analyzed 2 hours after collection to account for the swelling seen in EDTA anticoagulant. The decrease in fibrinogen binding did not appear to be related to lower numbers of GPIIb-IIIa complexes per platelet because there was no direct statistical correlation between these parameters under any condition.
Fibrinogen binding was detected with a polyclonal antibody that recognizes both plasma and platelet-bound fibrinogen. Variation in plasma fibrinogen could therefore lead to altered platelet fibrinogen binding by direct competition for the antibody. In a similar way, increased platelet count could lead to reduced antibody binding and hence to lower levels of fluorescence per platelet. We previously showed30 that fibrinogen binding to resting and ADP-stimulated platelets is unaffected by levels of fibrinogen in the plasma, or platelet counts, within the normal range, and this was borne out in the present study. Plasma fibrinogen levels were, with one exception, in the normal range, as were platelet counts, and there was no overall correlation with fibrinogen binding and either of these parameters.
A number of possible mechanisms may underlie the observations of reduced platelet responsiveness with age and disease seen in the present study. First, there is in vitro evidence that platelets exposed to substimulatory levels of agonists such as thrombin are desensitized to subsequent rechallenge with the same agonist,52 a phenomenon that can be detected using the present assay technique.53 Platelets from patients with atherosclerotic arteries could be exhibiting a similar phenomenon in vivo, through constant contact with a low level of thrombin induced by the diseased endothelial surface. Thrombin generation has been reported in patients with stable angina, as evidenced by elevated levels of FPA.14 15
Levels of blood lipids may also affect platelet function. In vitro studies suggest that LDLs that have not undergone any oxidation can inhibit platelet aggregation and fibrinogen binding in a plasma milieu.54 In contrast, low levels of LDL oxidation can enhance aggregation and fibrinogen binding.54 55 Activated and/or hyperresponsive platelets have been reported in hyperlipidemic patients, but this may be associated with alterations in cell membrane lipid composition.56 57 The patients in this study did not show a statistical correlation between platelet responsiveness and lipid parameters. However, platelets from patients with stable coronary artery disease may circulate in a state of hyporesponsiveness, yet switch to abnormal, heightened aggregation in response to alterations in lipid oxidation status. Such a biphasic response, which may underlie the triggering of acute coronary syndromes and myocardial infarction, deserves further study.
An alternative explanation is that platelets that become activated in the circulation are rapidly removed by binding to endothelial cells, leukocytes, or become assimilated into thrombi, leading to a remaining circulating pool of hyporesponsive platelets. In a previous flow cytometric study comparing 16 patients with severe peripheral arterial disease with healthy age-matched control subjects, Galt et al58 found no difference in baseline P-selectin expression and no difference in plasma βTG between the groups. Interestingly, however, the control subjects formed significantly more platelet aggregates in response to in vitro stimulation with ADP than patients, in keeping with our findings in patients with coronary artery disease. Further studies, analyzing platelet leukocyte aggregates by flow cytometric methods (currently under development) are needed. In the present study levels of sP-selectin were not raised. It is unclear whether sP-selectin in the circulation is derived from platelets or endothelial cells, but our findings do not suggest an increase in sP-selectin from either source.
The patients in our study, but not the control subjects, were taking aspirin and β-blockers. We previously showed that aspirin, although a powerful antiplatelet agent via inhibition of cyclo-oxygenase, does not affect the flow cytometric detection of platelet activation in response to agonist stimulation. The flow cytometric assay detects aggregation-independent platelet activation events that are therefore independent of the cyclo-oxygenase pathway and are unaffected by aspirin.59 60 These latter studies were carried out in normal subjects, and it is possible that aspirin in patients with hyperresponsive platelets might serve to suppress platelet function. However, this would not explain the lower levels of platelet response seen in the patients compared with the control subjects. Furthermore, we have also shown61 that in healthy volunteers, the administration of atenolol has no effect on fibrinogen binding or P-selectin expression in unstimulated samples and tends to increase the expression of activation markers to agonist stimulation, the opposite effect of that seen in the patients in this study. Thus, the changes in the patients are likely to be disease related rather than pharmacologically induced.
In conclusion, in contrast to previous reports, we have shown that platelets from patients with stable angina are more resistant to agonist-provoked aggregation than those from age-matched control subjects. In addition, there appears to be a decrease in the ability of platelets to degranulate with advancing age.
Selected Abbreviations and Acronyms
|MPV||=||mean platelet volume|
|PF4||=||platelet factor 4|
C.J.K. is a British Heart Foundation Junior Research Fellow, D.J.P. has been supported as a fellow by the Augustus Newman Foundation, and M.P. is funded by a grant from the Royal Brompton Hospital Clinical Research Committee. This project was funded in part by a grant from Pfizer U.K., Ltd. We thank the Haemophilia Centre and the Department of Chemical Pathology (Royal Free Hospital) for the measurement of plasma fibrinogen and serum lipids, respectively.
Antiplatelet Trialists’Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. Br Med J. 1988;296:320-331.
Hampton JR, Gorlin R. Platelet studies in patients with coronary artery disease and in their relatives. Br Heart J. 1972;34:465-471.
Frishman WH, Weksler B, Christodoulou JP, Smithen C, Killip T. Reversal of abnormal platelet aggregability and change in exercise tolerance in patients with angina pectoris following oral propranolol. Circulation. 1974;50:887-896.
Tada M, Kuzuya T, Inoue M, et al. Elevation of thromboxane B2 levels in patients with classic and variant angina pectoris. Circulation. 1981;64:1107-1115.
Fuster V, Cheesbro JH, Frye RL, Elveback LR. Platelet survival and the development of coronary artery disease in the young adult: effects of cigarette smoking, strong family history and medical therapy. Circulation. 1981;63:546-551.
Files JC, Malpass TW, Yee EK, Ritchie JL, Harker LA. Studies of human platelet α-granule release in vivo. Blood. 1981;58:607-618.
Levine SP, Lindenfeld J, Ellis JB, Raymond NM, Krentz LS. Increased plasma concentrations of platelet factor 4 in coronary artery disease—a measure of in vivo platelet activation and secretion. Circulation. 1981;64:626-632.
Steele PP, Weily HS, Davies H, Genton E. Platelet function studies in coronary artery disease. Circulation. 1973;48:1194-1200.
Wu KK, Hoak JC. A new method for the quantitative detection of platelet aggregates in patients with arterial insufficiency. Lancet. 1974;ii:924-926.
Elwood PC, Renaud S, Sharp DS, Beswick AD, O’Brien JR, Yarnell JWG. Ischemic heart disease and platelet aggregation—the Caerphilly Collaborative Heart Disease Study. Circulation. 1991;83:38-44.
Thaulow E, Erikssen J, Sandvik L, Stormorken H, Cohn PF. Blood platelet count and function are related to total and cardiovascular death in apparently healthy men. Circulation. 1991;84:613-617.
Shattil SJ, Cunningham M, Hoxie JA. Detection of activated platelets in whole blood using activation-dependent monoclonal antibodies and flow cytometry. Blood. 1987;70:307-315.
Michelson AD, Ellis PA, Barnard MR, Matic GB, Viles AF, Kestin AS. Downregulation of the platelet surface glycoprotein Ib-IX complex in whole blood stimulated by thrombin, adenosine diphosphate or an in vivo wound. Blood. 1991;77:770-779.
Freidewald WT, Levy RJ, Fredrickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma without the use of a preparative ultracentrifuge. Clin Chem. 1979;18:499-502.
Goodall AH. Platelet activation during preparation and storage of concentrates: detection by flow cytometry. Blood Coagul Fibrinolysis. 1991;21:377-389.
Golding JW. Monoclonal antibodies: principles and practice. London, UK: Academic Press; 1986.
Kestin AS, Ellis PA, Barnard MR, Errichetti A, Rosner BA, Michelson AD. Effect of strenuous exercise on platelet activation state and reactivity. Circulation. 1993;88:1502-1511.
Neri Serneri GG, Fortini A, Lombardi A, Modesti PA, Abbate R, Gensini GF. Reduction in prostacyclin platelet receptors in active spontaneous angina. Lancet. 1984;ii:838-841.
Modesti PA, Colella A, Cecioni I, et al. Increased number of thromboxane A2-prostaglandin H2 platelet receptors in active unstable angina and causative role of enhanced thrombin formation. Am Heart J. 1995;129;873-879.
Sobel M, Salzman EW, Davies GC, et al. Circulating platelet products in unstable angina pectoris. Circulation. 1981;63:300-306.
Martin JF, Plumb J, Kilbey RS, Kishk YT. Changes in volume and density of platelets in myocardial infarction. Br Med J. 1983;287:456-459.
Meade TW, Vickers MV, Thompson SG, Stirling Y, Haines AP, Miller GJ. Epidemiological characteristics of platelet aggregability. Br Med J. 1985;290:428-432.
Michelson AD, Barnard MR. Thrombin induced changes in platelet membrane glycoproteins Ib, IX and IIb-IIIa complex. Blood. 1987;70:1673-1678.
Hourdille P, Heilmann E, Combrie R, Winkler J, Clemetson KJ, Nurden AT. Thrombin induces a rapid redistribution of glycoprotein Ib-IX complexes within the membrane systems of activated human platelets. Blood. 1990;76:1503-1513.
Wilson DJ, Domingos MO, Goodall AH. Regulation of thrombin receptor mediated activation of human platelets. Thromb Haemost. 1995;73:931. Abstract.
Naseem KM, Goodall AH, Bruckdorfer KR. Differential effects of native and oxidatively modified low-density lipoproteins on platelet function. Platelets.1997;8:163-175.
Weidtmann A, Scheithe R, Hrboticky N, Pietsch A, Lorenz R, Sies W. Mildly oxidized LDL induces platelet aggregation through activation of phospholipase A2. Arterioscler Thromb Vasc Biol. 1995;15:1131-1138.
Rinder CS, Student LA, Bonan JL, Rinder HM, Smith BR. Aspirin does not inhibit adenosine diphosphate-induced platelet α-granule release. Blood. 1993;82:505-512.
Knight CJ, Panesar M, Wilson DJ, et al. Different effects of calcium antagonists, nitrates, and beta blockers on platelet function: possible importance for the treatment of unstable angina. Circulation. In press.