Abstract
BACKGROUND: The relevance of measures of general and central adiposity for cardiovascular disease (CVD) risks in populations of European descent is well established. However, it is less well characterised in South Asian populations, who characteristically manifest larger waist circumferences (WC) for equivalent body mass index (BMI). This systematic review and meta-analysis provide an overview of the literature on the association of different anthropometric measures with CVD risk among South Asians. METHODOLOGY: MEDLINE and Embase were searched from 1990 to the present for studies in South Asian populations investigating associations of two or more adiposity measures with CVD. Random-effects meta-analyses were conducted on the associations of BMI, WC and waist-to-hip ratio (WHR) with blood pressure, hypertension and CVD. Quality assessment was performed using the Newcastle-Ottawa scale. RESULTS: Titles and abstracts were screened for 7327 studies, yielding 147 full-text reviews. The final sample (n=30) included 2 prospective, 5 case-control and 23 cross-sectional studies. Studies reported generally higher risks of hypertension and CVD at higher adiposity levels. The pooled mean difference in systolic blood pressure (SBP) per 5 kg/m(2) higher BMI was 3 mmHg (2.90 (95% CI 1.30 to 4.50)) and 6 mmHg (6.31 (95% CI 4.81 to 7.81) per 13 cm larger WC. The odds ratio (OR) of hypertension per 5 kg/m(2) higher BMI was 1.33 (95% CI 1.18 to 1.51), 1.45 (95% CI 1.05 to 1.98) per 13 cm larger WC and 1.22 (95% CI 1.04 to 1.41) per 0.1-unit larger WHR. Pooled risk of CVD for BMI-defined overweight versus healthy-weight was 1.65 (95% CI 1.55 to 1.75) and 1.48 (95% CI 1.21 to 1.80) and 2.51 (95% CI 0.94 to 6.69) for normal versus large WC and WHR, respectively. Study quality was average with significant heterogeneity. CONCLUSIONS: Measures of both general and central adiposity had similar, strong positive associations with the risk of CVD in South Asians. Larger prospective studies are required to clarify which measures of body composition are more informative for targeted CVD primary prevention in this population.