HIV, antiretroviral treatment, hypertension, and stroke in Malawian adults: A case-control study

Abstract

Objective: To investigate HIV, its treatment, and hypertension as stroke risk factors in Malawian adults.

Methods: We performed a case-control study of 222 adults with acute stroke, confirmed by MRI in 86%, and 503 population controls, frequency-matched for age, sex, and place of residence, using Global Positioning System for random selection. Multivariate logistic regression models were used for case-control comparisons.

Results: HIV infection (population attributable fraction [PAF] 15%) and hypertension (PAF 46%) were strongly linked to stroke. HIV was the predominant risk factor for young stroke (≤45 years), with a prevalence of 67% and an adjusted odds ratio (aOR) (95% confidence interval) of 5.57 (2.43-12.8) (PAF 42%). There was an increased risk of a stroke in patients with untreated HIV infection (aOR 4.48 [2.44-8.24], p < 0.001), but the highest risk was in the first 6 months after starting antiretroviral therapy (ART) (aOR 15.6 [4.21-46.6], p < 0.001); this group had a lower median CD4(+) T-lymphocyte count (92 vs 375 cells/mm(3), p = 0.004). In older participants (HIV prevalence 17%), HIV was associated with stroke, but with a lower PAF than hypertension (5% vs 68%). There was no interaction between HIV and hypertension on stroke risk.

Conclusions: In a population with high HIV prevalence, where stroke incidence is increasing, we have shown that HIV is an important risk factor. Early ART use in immunosuppressed patients poses an additional and potentially treatable stroke risk. Immune reconstitution inflammatory syndrome may be contributing to the disease mechanisms.

Figure 1. Flow diagram of case and control recruitment

*Control selection: dwellings were visited and all eligible potential control participants were identified. If no one was home, dwellings were visited up to 3 times. Where multiple individuals were eligible, the oldest individual was recruited. If no eligible individual was identified, the next dwelling intersected was visited. Recruitment continued in each residential neighborhood until the prespecified numbers of individuals in each age and sex category had been met. QECH = Queen Elizabeth Central Hospital.

Figure 2. HIV treatment status, viral load, CD4⁺ count, and stroke risk

(A) Multivariate analysis of HIV treatment status, CD4⁺ count, and stroke risk, represented graphically. (B) Univariate and multivariate analysis of HIV treatment status, CD4⁺ count, and stroke risk. Explores the association of HIV treatment status and stroke risk after adjusting for immunosuppression. *Adjusted for frequency-matched variables: age, sex, and urban location. **Adjusted for hypertension, recent infection, abdominal obesity, HIV treatment status, smoking, current alcohol use, CD4⁺ T-lymphocyte count, hypercholesterolemia, cannabis use, age, sex, type of housing, and urban location. †A combined p value was calculated using a likelihood ratio test for variables with >2 categories. Data were missing for the following cases and controls: 13 HIV treatment status (including HIV viral load data), 12 CD4⁺ T-lymphocyte cell count, 2 recent infection, 2 waist-hip ratio, 1 alcohol, 8 pregnant, 1 substance use. Missing observations were included in the analysis by creating missing value categories.