Studies of Human Platelet α-Granule Release In Vivo

Abstract

AbstractThe clinical usefulness of measuring platelet factor 4 (PF4) and β-thromboglobulin (βTG) by radioimmunoassay in plasma, urine, and platelets has been evaluated in 105 normal subjects and 184 selected patients with suspected alterations in platelet reactivity. Since blood collection techniques are critical in minimizing in vitro artifact when measuring plasma PF4 and βTG, we developed a method for sampling and processing with respect to anticoagulant, inhibitory agents, temperature, centrifugation, and interval before processing. In 105 normal subjects, the levels of PF4 and βTG were 1.8 ± 1.0 (± 1 SD) and 6.0 ± 3.6 ng/ml, respectively, in plasma; 1.36 ± 0.24 and 4.29 ± 1.38 ng/fl × 10−6, respectively, in platelets; and 0.18 ± 0.16 ng PF4/mg creatinine in urine. Serum values for PF4 and βTG in these normal subjects, as well as in individuals with thrombocytopenia or thrombocytosis, correlated closely with the platelet concentration in whole blood (coefficient of correlation 0.947 and 0.826, respectively; p < 10−4), but the plasma levels of these platelet-specific proteins showed no correlation with platelet counts (p > 0.2). Patients with acute events such as cardiopulmonary bypass showed parallel depletion of PF4 and βTG from platelets, striking elevation of these proteins in plasma and clearance of PF4 into urine. Untransfused, severely thrombocytopenic patients with aplastic anemia had low levels of PF4 and βTG in their plasma, i.e., 0.3 ± 0.3 (p < 0.0001) and 4.8 ± 5.0 (p = 0.48) ng/ml, respectively, as well as in their platelets, 0.19 ± 0.17 (p < 0.0001) and 0.35 ± 0.16 (p < 0.0001) ng/fl × 10−6, respectively, compared with normal values. Despite marked increases in platelet destruction (shortened platelet survival and increased platelet turnover), patients with immune thrombocytopenic purpura (ITP) had normal levels of plasma PF4 and βTG (p = 0.81 and 0.29, respectively, compared with normal results). The two proteins were slightly elevated in the plasma of 41 hyperlipidemic subjects. However, these measurements could not differentiate patients into those with and without chronic vascular disease (p = 0.93). Measurements of these proteins in platelets or PF4 in urine were no more helpful than the plasma levels in predicting vascular disease. Levels in patients with atherosclerotic coronary artery disease showed substantial overlapping with levels in normal subjects. Patients with coronary artery disease who had survived out-of-hospital ventricular fibrillation without infarction had normal values of plasma PF4 (2.3 ± 1.5 ng/ml; p = 0.13) but significant elevations of plasma βTG (17.5 ± 11.9; p < 0.0001). In patients with documented atherosclerotic cardiovascular disease and angina pectoris, there was a small but statistically significant increase in plasma PF4 and βTG (3.1 ± 1.4 and 12.1 ± 2.9 ng/ml, p = 0.0003 and p < 0.0001 for PF4 and βTG, respectively, compared with normal values). Patients with a documented remote myocardial infarction also showed small increases in mean plasma PF4 values (3.1 ± 3.9 ng/ml; p = 0.002); treatment with aspirin did not decrease the plasma PF4 level (p > 0.1). We conclude that measurements of PF4 and βTG in plasma and platelets and of PF4 in urine do not identify individuals with stable chronic atherosclerotic vascular disease and thus do not generally provide a means for identifying patients at risk or following the effects of therapy. In contrast, marked elevation in plasma levels and depletion of these proteins from circulating platelets occur with acute in vivo platelet activation and release.