A longitudinal investigation of the cardiovascular health of a black South African cohort infected with the human immunodeficiency virus

Abstract

Motivation - The human immunodeficiency virus (HIV) continues to be a global health concern, with an estimation of 38 million people living with HIV. Out of these 38 million people living with HIV, 70% are residing in Sub-Saharan Africa (SSA) countries. Alongside the high burden of HIV infection, cardiovascular diseases (CVD) are also an important health concern in this region. The introduction of antiretroviral therapy (ART) has increased the life expectancy of HIV-infected individuals with consequential exposure to comorbidities such as hypertension, dyslipidaemia and diabetes mellitus as well as liver and renal disease – all known to be associated with an increased risk for the development of CVD. With 90% of South Africans living with HIV expected to be on ART by 2030, the prevalence of CVD risk factors is also expected to increase. Hence, it is vital to fully understand the impact of HIV, ART and conventional cardiovascular risk factors in the development of CVD in HIV-infected individuals. This understanding is currently limited by available evidence from longitudinal studies, especially in the SSA region. To better understand the long-term risk of developing CVD in people living with HIV, the cardiovascular health transition of HIV-infected Africans who are at risk of developing CVD was investigated for this study. Aim - The central aim of this study was to investigate the longitudinal cardiovascular health of a black South African cohort infected with HIV and a literature review based on studies conducted in the SSA region. Methods - The narrative review included a synthesis of studies done in SSA reporting on blood pressure (BP) and hypertension in HIV. PubMed and EBSCO Discovery Service were searched for manuscripts published up to May 2020. Ultimately, 19 studies were included that reported on estimates of BP and/or hypertension

prevalence in treated HIV-infected individuals versus untreated HIV-infected individuals and uninfected controls, residing in SSA countries. The original papers of this study are based on the South Africa leg of the Prospective Urban and Rural Epidemiology (PURE) study, North West Province. Manuscript two and three, which form part of the PURE study, included 117 HIV-infected individuals and 131 uninfected controls and 126 HIV-infected and 126 uninfected controls, respectively. The study participants comprised of African men and women older than 35 years. Participation in the PURE study was voluntary and written informed consent was obtained from each participant. Standard protocols and methods were used to gather all data and consisted of questionnaire data (demographic, lifestyle and medical history). Other data included HIV-related factors (ART use and duration, CD4 cell count), body composition and cardiovascular measurements (blood pressure, carotid femora pulse wave velocity, carotid intima-media thickness), as well as biochemical analyses of all relevant biomarkers used in the study. Regarding statistical analyses, variables that were not normally distributed were log-transformed. Means and proportions were compared using the independent t-tests, dependent t-tests, Chi-square test, McNemar test or analysis of variance or covariance and repeated measure analysis. With manuscript three, partial and multiple regression analyses were used to investigate associations between the main dependent (cfPWV and cIMT) and independent (endothelial activation ad cardiovascular risk markers) variables of interest. In all cases, p≤0.05 was used to indicate statistical significance. Results of each manuscript - Manuscript one - A narrative review synthesis was conducted to determine the estimates of raised BP and hypertension in HIV-infected patients compared to untreated and HIV-uninfected controls from SSA. A total of 19 eligible studies that compared treated HIV-infected with untreated and/or HIV-uninfected controls were included. In studies comparing treated HIV-infected patients to uninfected controls, a large proportion of studies, reported lower BP 6,882 (56.30%) and hypertension prevalence 21,819 (79.2%) respectively, in the treated HIV-infected individuals than in HIV-uninfected controls. On the other hand, studies including 753 (6.16%) and

3,553 (12.9%) participants showed a higher BP and hypertension prevalence. Lastly, 4,588 (37.54%) and 2,180 (7.91%) participants showed no difference in BP and the prevalence of hypertension between the treated HIV-infected individuals and HIV-uninfected controls. Estimates of BP and hypertension prevalence were further compared for of treated versus untreated HIV-infected individuals. Studies including 5,757 (44.2%) and 4,547 (74.5%) participants reported lower BP and hypertension, respectively in treated compared to untreated HIV-infected individuals. On the other hand, studies including 200 (1.53%) and 598 (9.80%) participants showed higher BP and hypertension prevalence. Similar BP and hypertension prevalence was reported 7,073 (54.28%) and 959 (15.7%) participants respectively in the treated and untreated HIV-infected individuals. Manuscript two - This manuscript profiled the demographic and cardiometabolic factors, indicators of inflammation and oxidative stress as well as liver and renal function of HIV-infected Africans and uninfected controls over 10 years. The HIV-infected individuals living with HIV for over 10 years presented with longitudinal increases in C-reactive protein (p=0.002), alanine transaminase (p=0.006) and gamma-glutamyl transferase (p=0.046) levels as well as estimated glomerular filtration rate (eGFR) (p<0.001). The HIV-infected individuals also showed increased high-density lipoprotein cholesterol (HDL-c) (p<0.001) and total cholesterol (p<0.001) levels and decreased triglycerides: HDL-c (p=0.011) levels. Low-density lipoprotein cholesterol (LDL-c) decreased in both groups (p < 0.001), although the decrease was relatively smaller in the HIV-infected individuals. Manuscript three - In this manuscript, it was firstly determined whether individuals living with HIV for over five years have increased measures of arterial injury (carotid intima-media thickness (cIMT)) and large artery stiffness (carotid-femoral pulse wave velocity (cfPWV)) when compared to their HIV-uninfected counterparts. As a result, similar cIMT (p=0.90) and cfPWV (p=0.35) were reported in both groups. Secondly, it was determined whether baseline markers of endothelial activation and cardiovascular risk are associated with cIMT and cfPWV after five years. Although, levels of soluble intracellular adhesion molecule-1 (ICAM-1)

and soluble vascular cell adhesion molecule-1 (VCAM-1) were elevated in the HIV-infected group (all p<0.001) at baseline, these markers did not associate with cIMT and cfPWV after five years. In multivariable-adjusted regression analysis cIMT associated positively with age (β=0.31, p=0.002) and triglycerides: high-density lipoprotein-cholesterol (β=0.23; p=0.016) in the HIV-infected group. Also, cfPWV was positively associated with mean arterial pressure (MAP) (β=0.28; p=0.010) in the HIV-infected group. In the HIV-uninfected group, cIMT was associated with sex (β=0.31; p=0.004) and glycated hemoglobin (HbA1c) (β=0.24; p=0.026) while cfPWV was associated with age (β=0.17; p=0.049), sex (β=0.29; p=0.003), MAP (β=0.31; p=0.001) and HbA1c (β=0.21; p=0.041). General conclusion - A large proportion of studies conducted in SSA countries suggest that lower BP and hypertension prevalence is eminent in treated HIV-infected individuals compared to their uninfected and untreated HIV-infected controls. Despite this, deranged lipid profiles were reported accompanied by increased trajectories of inflammatory status, oxidative stress, liver enzymes and eGFR in HIV-infected individuals living with HIV for 10 years, proposing increased long-term risk for CVD. Traditional CV risk factors and, not elevated endothelial activation and HIV status, associated with measures of arterial injury and stiffness over time in HIV-infected people. This finding emphasises the need for HIV care in SSA settings where CVD is increasing due to epidemiological transition. Therefore, it is critical to screen for these CVD risk factors in HIV-infected individuals to reduce the possible burden of CVD burden in the future.