Adrenal sex steroids exert diverse metabolic and neurobiological actions. Their levels have been associated with cardiovascular disease, but data concerning cerebrovascular disease are lacking. The objective of our study was to investigate the role of adrenal sex steroids in a female population suffering an acute stroke. We addressed the question of whether their levels are associated with disease severity and prognosis. A 2-year cohort study was performed in 2 tertiary hospitals, where we prospectively studied 302 consecutive postmenopausal female patients hospitalized for an acute stroke. Neurological severity on admission was assessed by the National Institutes of Health Stroke Scale; and handicap 1 month after stroke, with the modified Rankin Scale. Δ4-androstenedione levels were positively and dehydroepiandrosterone sulfate was inversely associated with stroke severity (r = 0.142, P = .014 and r = −0.153, P = .008, respectively), and both parameters remained as significant determinants even after entering other confounders in the multivariate model (r = 0.118, P = .039 and r = −0.150, P = .011, respectively). Levels of Δ4-androstenedione were significantly associated with 1-month mortality in the multivariate analysis (odds ratio with 95% confidence intervals: 1.540 [1.107-2.138)], P = .010). Δ4-androstenedione and dehydroepiandrosterone sulfate levels were associated with poor outcome in the univariate analysis, that is, combined severe handicap (modified Rankin Scale ≥4) and death, 1 month poststroke, although this was not significant in the multivariate analysis. Adrenal sex steroids, and especially Δ4-androstenedione, are significantly associated with stroke severity on admission and short-term prognosis among female stroke subjects. Well-designed prospective studies will further clarify their role in cerebrovascular disease.