OBJECTIVE Cocaine-mediated erythrocytosis is one of several effects that cocaine may have on hematologic indices; however, the precise mechanism by which cocaine induces peripheral erythrocytosis is not fully understood. The objective of this study was to examine the contribution of the bone marrow to cocaine-mediated erythrocytosis. METHODS Differences in mean hemoglobin concentration, hematocrit, and reticulocyte counts were measured in consecutive cocaine-exposed and cocaine-unexposed patients who presented to the emergency department (ED) with chest pain. Acute cocaine exposure (<3 hours) was confirmed by history and toxicologic analysis of the urine on all patients. Means were compared using independent-samples t-test and covariates were examined using multiple regression. RESULTS Seventy-nine patients met enrollment criteria. Hemoglobin and hematocrit levels were significantly elevated in the cocaine-using subjects (13.5/39.8) compared with controls (12.6/37.7; p < 0.02). However, no corresponding elevation in reticulocyte count (p = 0.2) was observed. Multivariate logistic regression revealed that male chest pain patients were significantly more likely to be exposed to cocaine (OR 5.15 [95% CI = 1.77 to 15.3]) than females (p = 0.001), and all relative increases in hemoglobin concentration in the cocaine-exposed group were attributable to gender. Cocaine exposure was not significantly associated with reticulocyte count. Multivariate linear regression revealed that of demographic, medical, and substance use covariates, only a history of diabetes mellitus was significantly associated with an elevated reticulocyte count (p = 0.009). CONCLUSIONS Acute cocaine exposure is not associated with erythrocytosis in younger ED patients with chest pain. The lack of an elevated reticulocyte count suggests that bone marrow does not contribute to any transient erythrocytosis that may occur.