Determining the prevalence and optimising the diagnosis of metabolic syndrome in people living with HIV

Background and Purpose: With the introduction of potent antiretroviral therapy (ART) leading to longevity, cardio-metabolic diseases are emerging health concerns in people living with HIV. This thesis aims to 1) to quantify the burden of metabolic syndrome (MS), which is a constellation of cardio-metabolic risk factors, in people living with HIV infection (PLWHIV) from a global perspective; and in ≥18-year-old PLWHIV receiving HIV-care in the Western Cape, South Africa to 2) determine the MS prevalence and the agreement between the popular MS diagnostic criteria, 3) assess the distribution of cardio-metabolic risk factor clustering by adiposity levels, 4) determine the optimal waist circumference (WC) thresholds, and 5) HbA1c to define abdominal obesity and dysglycaemia, respectively, for the purpose of MS screening in South African PLWHIV. Methods: A systematic review with meta-analysis was conducted to determine the MS prevalence globally. The main study comprised a representative cross-sectional study of PLWHIV receiving HIV-care at 17 public healthcare clinics across the Western Cape, South Africa. Results: The global prevalence of MS was 16.7%-31.3% by different diagnostic criteria with substantial heterogeneity not explained by major study characteristics. The prevalence was higher in women than in men (International Diabetes Federation [IDF]-2005, 23.2% vs. 13.4%, p=0.030), in antiretroviral therapy (ART) versus non-ART users (Adult Treatment Panel III [ATPIII]-2001, 18.4% vs. 11.8%, p=0.001), and varied significantly by participant's age, duration of HIV diagnosis, CD4 count level, ART regimens. In the study conducted in the Western Cape (N=748, median age 38 years), MS prevalence among PLWHIV was 28.2% (JIS-2009), 26.5% (IDF-2005), and 24.1% (ATPIII-2005), which was higher in women, participants with longer duration of diagnosed HIV infection, ART users not receiving 1st line regimen (all p≤0.039). There was a good agreement between sets of the criteria that was not affected by HIV-related factors (all kappa ≥0.81). Cardio-metabolic risk factors clustered across all categories of adiposity levels: 11.7% of normal-weight, and 15.1% of obese PLWHIV had two or more factors, and this distribution was not affected by HIV-specific features. The optimal WC thresholds for abdominal obesity were 92 cm (sensitivity 64%, specificity 64%) in women and 87 cm (sensitivity 48%, specificity 85%) in men, which differed from the internationally recommended 80 cm (women) and 94 cm (men). The optimal HbA1c thresholds to define oral glucose tolerance test diagnosed dysglycaemia was 5.75% (39.3 mmol/mol) (sensitivity 52%, specificity 85%), similar to the threshold of 5.7% (39 mmol/mol) recommended by the American Diabetes Association. The MS prevalence by the JIS was 28.2% and 29.7% using glucose-defined and HbA1c-defined dysglycaemia, with a good agreement between the original and modified criteria (kappa=0.81). Conclusions: The high prevalence of cardio-metabolic risk factor clustering in PLWHIV highlights the need for effective management strategies. Optimal performance of MS criteria in South African PLWHIV requires the use of the African-population-specific WC thresholds, while using HbA1c to diagnose dysglycaemia could be both effective and more practical than blood glucose defined dysglycaemia. However, these findings need to be confirmed and the effects of their introduction in routine care on healthcare behaviour and patient outcomes assessed through impact and implementation studies