A case-control study of stroke risk factors and outcomes in African American stroke patients with and without crack-cocaine abuse

Abstract

Background: Stroke in crack-cocaine abusers is increasingly recognized. We aimed to identify significant differences in stroke risk factors, characteristics, and outcomes between hospitalized stroke patients with and without crack-cocaine abuse.

Methods: This was a retrospective study that compared stroke patients with crack-cocaine abuse (cases) to stroke patients without crack-cocaine (controls).

Results: We identified 93 crack-cocaine cases and 93 controls admitted between January 2004 and May 2006 to one teaching hospital. There were significant differences between crack-cocaine cases and controls in age (48.7 years vs. 55 years) (P = 0.0001), male gender (65.6% vs. 40.9%) (odds ratios, OR = 1.64, 95% CI 1.22-2.21), arterial hypertension (61.1% vs. 83.9%) (OR = 0.30, 95% CI 0.15-0.60), hypercholesterolemia (18.7% vs. 68.5%) (OR = 0.10, 95% CI 0.05-0.21), diabetes (20.9% vs. 41.9%) (OR = 0.36, 95% CI 0.19-0.70), cigarette smoking (70.6% vs. 29%) (OR = 5.86, 95% CI 3.07-11.20), ischemic stroke (61.3% vs. 79.6%) (OR = 0.40, 95% CI 0.21-0.78), and intracerebral hemorrhage (33.3% vs. 17.2%) (OR = 3.03, 95% CI 1.53-6.00). Also, there were significant differences in National Institutes of Health Stroke Scale scores (3.3 vs. 7) (P < 0.0001), and MRS scores (1.8 vs. 2.5) (P = 0.0022) at hospital discharge. Using univariable and multivariable logistic regression, we found that crack-cocaine abusers had 2.28 higher odds of having a favorable functional outcome (MRS score ≤ 2) at hospital discharge, after adjusting for stroke risk factors and characteristics.

Conclusions: Our study suggests that crack-cocaine abusers with stroke had fewer traditional risk factors, and more favorable functional outcome as compared to non-crack-cocaine abusers.