The association of vascular risk factors with cognitive impairment in ART‐treated HIV‐positive adults aged 50 years or older in Northern Tanzania

Abstract

AbstractBackgroundThe ageing of the global HIV‐positive population brings with it the novel challenges of geriatric medicine, such as management of comorbidities including cardiovascular disease and HIV‐associated neurocognitive disorder (HAND). Recent literature from high‐income countries suggests vascular risk factors, which are increasingly prevalent throughout sub‐Saharan Africa (SSA), are stronger predictors of HAND than HIV‐disease severity. There is a paucity of evidence surrounding the association between HAND, vascular risk factors and end‐organ damage (EOD) in older cohorts of people living with HIV (PLWH) in SSA. This study aimed to establish prevalence, and evaluate the relationship between vascular risk factors, EOD and HAND in individuals aged ≥50 receiving HIV treatment in TanzaniaMethodA systematic sample of outpatients with HIV on combined antiretroviral therapy (cART) were recruited in Kilimanjaro and assessed for vascular risk factors (hypertension, hypercholesterolaemia, diabetes, smoking etc) and markers of vascular EOD (prior myocardial infarction and LVH on ECG, stroke on neurological examination or MRI‐brain, estimated‐glomerular filtration rate and retinal arteriovenous ratio measurements). HAND were defined by AAN criteria following a detailed locally normed neurocognitive assessment battery, neurological and clinical examination and informant history, and will subsequently be confirmed by consensus panel.ResultComplete data were available for 145 individuals (mn age 56, 72.7% female). Provisional HAND prevalence was high (64.7%, 31.4% symptomatic) despite well‐managed HIV‐disease (70.5% suppressed viral load). 34.0% of the cohort were hypertensive, 10.5% obese, 33.3% had hypercholesterolaemia, 5.3% had diabetes and 4.6% smoked. Vascular EOD prevalence ranged from 1.3% (prior myocardial infarction) to 12.5% (LVH). No significant association was found between vascular factors and HAND. Conversely symptomatic HAND was significantly correlated with lower BMI (p=0.034) and lower total serum cholesterol (p=0.014).ConclusionHAND prevalence was high, however prevalence of vascular risk factors was lower than in recent Tanzanian community studies. The lack of association between vascular factors and HAND does not support the hypothesis that HAND is driven by vascular damage. However, the high prevalence of HAND observed suggests an alternative aetiology within this cohort. Further research is urgently required to explore factors contributing to the high burden of neurocognitive impairment among PLWH in SSA.