HIV Infection and Incidence of Ischemic Stroke

Julia L. Marcus; Wendy A. Leyden; Chun R. Chao; Felicia C. Chow; Michael A. Horberg; Leo B. Hurley; Daniel B. Klein; Charles P. Quesenberry Jr; William J. Towner; Michael J. Silverberg


AIDS. 2014;28(13):1911-1919. 

In This Article

Abstract and Introduction


Objective: To determine the association of HIV infection and immunodeficiency with incidence of ischemic stroke.

Design: Cohort study of HIV-positive and matched HIV-negative adult Kaiser Permanente Northern and Southern California (KPNC and KPSC, respectively) members during 1996–2011 (KPNC) or 2000–2011 (KPSC).

Methods: We used Poisson models to obtain rate ratios for incident ischemic stroke associated with HIV infection, both overall and stratified by CD4+ cell counts (cells/μl) and HIV RNA copies (copies/ml), with HIV-negative individuals as the reference group. We also obtained rate ratios for risk factors in the HIV-positive subset.

Results: Among 24 768 HIV-positive and 257 600 HIV-negative individuals, the ischemic stroke rate per 100 000 person-years was 125 (n = 151 events) for HIV-positive and 74 (n = 1128 events) for HIV-negative individuals, with an adjusted rate ratio of 1.4 [95% confidence interval (CI) 1.2–1.7). Compared with HIV-negative individuals, HIV-positive individuals with recent CD4+ cell counts of 500 cells/μl at least (rate ratio 1.0, 95% CI 0.8–1.4) or recent HIV RNA less than 500 copies/ml (rate ratio 1.1, 95% CI 0.9–1.4) had no excess risk of ischemic stroke, with similar results for HIV-positive individuals with nadir CD4+ cell counts of 500 cells/μl at least (rate ratio 1.4, 95% CI 0.8–2.2) or 200–499 cells/μl (rate ratio 1.2, 95% CI 0.9–1.5). Among HIV-positive individuals only, recent CD4+ cell count less than 200 cells/μl (rate ratio 2.5, 95% CI 1.3–4.6) was associated with an increased risk of ischemic stroke after adjustment for recent HIV RNA and nadir CD4+ cell count, whereas recent HIV RNA and nadir CD4+ were not independent risk factors.

Conclusion: Ischemic stroke incidence in HIV-positive individuals with high CD4+ cell count or low HIV RNA is similar to that of HIV-negative individuals.


Combination antiretroviral therapy (ART) has dramatically improved survival among HIV-positive individuals. By 2015, over half of the HIV-positive population in the United States will be 50 years of age or older.[1] As this population ages, aging-associated conditions such as cardiovascular disease increasingly contribute to morbidity and mortality,[2] with a doubling of ischemic stroke hospitalizations among HIV-positive individuals during 1997–2006.[3] However, age-adjusted rates of ischemic stroke in this population may have stabilized or decreased in recent years with improvements in risk factor management or the wider and earlier use of ART regimens associated with improved lipid profiles.[4,5] Furthermore, it is unclear whether HIV-positive individuals with well controlled infection are at an increased risk of ischemic stroke, independent of age, compared with HIV-negative individuals.

Although numerous studies have explored the risk of myocardial infarction associated with HIV infection,[6–11] few large-scale epidemiologic studies have directly compared ischemic stroke incidence and risk factors by HIV status. Although ischemic stroke rates are reportedly higher in HIV-positive individuals compared with the general population,[12,13] few studies have directly compared these patients with an internal control group of HIV-negative individuals. In one of the only studies examining ischemic stroke using an internal HIV-negative comparison group and with adjustment for stroke risk factors, Chow et al.[14] found an elevated risk of ischemic stroke among approximately 4000 HIV-positive individuals. Although that study identified high HIV RNA but not low CD4+ cell count as a risk factor, others have found that low CD4+ cell count is associated with increased risk.[12,13,15] Findings on the relationship of ART use and ischemic stroke have also been mixed, with some studies suggesting that longer duration of any ART use is associated with a decreased risk[14,16] and others finding no association.[12,13]

In this study, we determined the association of HIV infection and immunodeficiency with ischemic stroke in a large cohort of nearly 25 000 HIV-positive individuals, with a demographically similar comparison group of HIV-negative individuals from the same healthcare system. Among HIV-positive individuals, we assessed the role of traditional and HIV-specific risk factors for ischemic stroke, including CD4+ cell counts, HIV RNA levels, and class-specific duration of ART use.