Background: Cocaine use has been associated with arterial occlusion resulting from platelet-rich thrombi and with an accelerated, often atypical atherosclerotic lesion that could be ascribed to platelet activation and platelet alpha-granule release.
Methods and results: Using a flow cytometric method to quantitate the percent of circulating activated platelets in whole blood (those that express the alpha-granule membrane protein P-selectin), we found that 5 of 25 samples from 12 long-term cocaine users had a baseline level of circulating activated platelets > 3 SD (range, 19% to 60%) above the mean (4.4 +/- 3.7%, mean +/- 1 SD) for 85 nonusers (sample n = 130). This subset resulted in a significantly higher mean baseline level of circulating activated platelets (11.8 +/- 14.4%) for all cocaine users (P = .01). By contrast, cocaine and its metabolites, at concentrations documented as obtainable during in vivo cocaine use (10(-7) to 10(-5) mol/L), had no effect on in vitro platelet activation or aggregation, either directly or in concert with platelet agonists. However, in experiments in which cocaine users received blinded infusions of placebo or cocaine, the mean percent of circulating activated platelets rose significantly (P < .05) after infusion of either placebo (peak 77 +/- 31%) or cocaine (peak 65 +/- 28%), the latter at doses resulting in peak plasma cocaine levels averaging < 10(-6) mol/L.
Conclusions: Long-term cocaine use in some subjects is intermittently associated with high basal levels of circulating platelets that have undergone alpha-granule release. The inability of cocaine and its metabolites at concentrations of 10(-7) to 10(-5) mol/L to cause platelet P-selectin expression in vitro in this study, coupled with the acute increase in circulating activated platelets observed in vivo after either cocaine or placebo infusion, suggests that in vivo platelet alpha-granule release associated with cocaine use may occur through indirect rather than direct effects of the drug.