Etiology and risk factors of stroke in HIV-infected patients in Siriraj Hospital: a case-control study.

Abstract

Stroke and HIV infection are major health problems in Thailand. There is limited data regarding the etiology and risk factors of stroke in HIV-infected Thai patients. To study the risk factors, types, and mechanisms of stroke in HIV-infected patients. The authors reviewed records of consecutive HIV-infected patients with acute first stroke in a large urban medical center from August 1, 2009 through December 31, 2010. Age-matched controls of HIV-infected patients without stroke were consecutively recruited at a 2:1 ratio. Data collection included demographics, stroke subtypes, risk factors of stroke, and HIV disease parameters. Multiple logistic regression analysis (p < 0.05) identified factors associated with stroke in HIV-infected patients. There were 37 subjects and 74 controls. In HIV-positive stroke patients, 81.1% were males and mean age was 50.5 years. There were 33 and 4 cases of ischemic and hemorrhagic strokes respectively. HIV infection was previously diagnosed in 70%, mean CD4 count was 287 cells/uL and 33% had CD4 counts < 200 cells/uL. Prior antiretroviral medications were used in 49%. TOAST classification of stroke was as follows: large artery atherosclerosis 2 (6.1%), small vessel occlusion 9 (27.3%), cardioembolism 2 (6.1%), other determined etiology 9 (27.3%) (vertebral artery dissection 1, anti-thrombin III deficiency 1, thrombotic thrombocytopenic purpura 1, tuberculous meningitis 4, cryptococcal meningitis 1, intravenous heroin 1) and undetermined 11 (33.2%) (incomplete evaluation 10, negative evaluation 1). Multivariate analysis demonstrated the following to be significant risk factors of stroke: smoking p = 0.001, adjusted OR 6.9 (95% CI 12.3, 21.1) and tuberculous meningitis p = 0.034, adjusted OR 11.9 (95% CI 1.2, 117.2). Stroke etiology in HIV-infected patients is more heterogeneous than in non-immunocompromised hosts. Smoking and concurrent tuberculous meningitis were significantly associated with stroke in HIV-infected Thai patients. Further prospective cohort studies should be performed in a larger population of more severely immunocompromised patients in Thailand.