Cocaine-induced cerebral vasoconstriction detected in humans with magnetic resonance angiography.

CONTEXT Clinical observations and case reports suggest that there are important cerebrovascular complications of cocaine use, but no studies have documented a direct link. OBJECTIVE To determine whether low-dose cocaine administration induces cerebral vasoconstriction in healthy cocaine users. DESIGN Randomized controlled trial. SUBJECTS Twenty-four healthy and neurologically normal men (mean age, 29 years) reporting median cocaine use of 8 lifetime exposures (range, 3 to >40). INTERVENTION Double-blind intravenous administration of cocaine (0.4 or 0.2 mg/kg) or placebo, with cerebral magnetic resonance angiography performed at baseline and 20 minutes following infusion. MAIN OUTCOME MEASURE Cocaine-induced angiographic change indicative of vasoconstriction, as independently and concordantly rated by 2 reviewers blind to treatment condition. RESULTS Cocaine-induced cerebral vasoconstriction in a dose-related fashion (P=.03), with angiograms indicative of vasoconstriction found in 5 of 8 and 3 of 9 subjects receiving 0.4- and 0.2-mg/kg cocaine, respectively, compared with 1 of 7 subjects administered placebo. Outcome stratification by frequency of self-reported lifetime cocaine use (3-10 times, 11-40 times, or >40 times) revealed a statistically stronger dose-related effect (P<.001), suggesting that greater lifetime cocaine use was associated with a greater likelihood of vasoconstriction. CONCLUSIONS Cocaine administration induced dose-related cerebral vasoconstriction on magnetic resonance angiograms. These changes occurred at low cocaine doses and in the absence of other risk factors, including polydrug abuse, hypertension, or cerebrovascular disease. Outcome stratification by prior cocaine use statistically strengthened the relationship between cocaine administration and vasoconstriction, suggesting that cocaine may have a cumulative residual effect in promoting cerebrovascular dysfunction.