The role of obstructive sleep apnoea on chronic inflammation and cardiometabolic risk in people living with HIV

Abstract

Obstructive sleep apnoea (OSA) is a proinflammatory respiratory disorder characterised by interruptions to breathing during sleep. OSA pathophysiology potentiates cardiometabolic risk (CMR) factors. Cohort follow-up studies support that with early OSA detection and intervention, CMR variables like hypertension status, circulating cholesterol levels, and certain peripheral markers of systemic inflammation can be reduced. Despite often showing more favourable cardiometabolic risk profiles using traditional risk factors of CMR, more people living with HIV (PLWH) develop cardiovascular disease and at a younger age than people without HIV. OSA prevalence in the general population can be up to 30%, yet there are no data available on OSA prevalence for PLWH in South Africa, nor whether OSA prevalence is associated with increased traditional (hypertension, insulin resistance, dyslipidaemia) and non-traditional (systemic and endothelial inflammation) CMR factors. Aim: This project intended to characterise the prevalence of high risk of OSA using the Berlin Questionnaire in South African people living with HIV, and to investigate the association between high risk OSA and CMR using traditional CMR factors and inflammatory markers. Methods: This is a cross-sectional, observational study of a well-characterised cohort of antiretroviral treatment-experienced people living with HIV residing in Johannesburg. Between September 2021 and November 2022, 210 participants previously enrolled in ADVANCE were recruited into the CHARACTERISE trial. Participants were excluded if pregnant or unable to complete the Berlin Questionnaire (BQ), a screening tool to detect OSA validated in “Global North” populations. All participants had been transitioned onto dolutegravir-based ART at least two years before recruitment. We measured blood pressure, body mass index, waist circumference and assayed blood for fasting lipid profiles and glucose concentration. A CMR score was calculated as a composite of calculated z-scores of waist circumference (WC), fasting glucose, high-density lipoprotein cholesterol (HDL- C), triglycerides, and mean arterial blood pressure. OSA risk was evaluated using the BQ. In 95 participants (24 with high risk OSA), we then assayed stored plasma for markers of systemic inflammation (IL-1 beta, IL-6, IL-8, TNF alpha) and endothelial activation (ESM, MCP1). v Results: In a relatively young (mean age (SD) = 38.4 years (± 7.5), 62% female) cohort of PLWH, 43 (20.5%) scored high risk of OSA. All participants self-identified with Black African ancestry, and 68% were employed at the time of enrolment. Three-quarters of the participants had hypertension or obesity (75%). Median CMR score (p = 0.0024; 95% CI: 0.090 - 0.437), body mass index (p < 0.0001; 95% CI: 2.318 - 7.708), WC (p < 0.001; 95% CI: 3.446 - 12.062), and triglyceride concentration (p = 0.0385; 95% CI: -0.059 - 0.272) were greater in patients at high risk of OSA than those at low risk for OSA. Five of the six markers of inflammation and endothelial activation were detectable in all participants for whom we had plasma (TNF alpha only in 35 participants, including 11 with high risk OSA). We found no significant difference between inflammation/endothelial activation in the high risk and low risk OSA groups. Conclusion: Individuals with high risk OSA had significantly higher cardiometabolic risk scores using traditional markers of CMR. No difference in systemic inflammation nor endothelial activation was found. However, the cohort-wide obesity and hypertension prevalence, and overall detectable rates of inflammation are concerning for the growing contribution of cardiometabolic diseases to mortality of PLWH in South Africa. A case can be made for the necessity of point-of-care OSA screening in HIV primary care to manage progression of cardiometabolic diseases. These findings further impress upon the potentials of monitoring impaired sleep quality from ART side-effects as a public health imperative.