Central obesity and nonalcoholic fatty liver disease risk after adjusting for body mass index

AIM: To investigate whether central obesity is associated with nonalcoholic fatty liver disease (NAFLD) formation after adjusting for general obesity. METHODS: The online databases PubMed, EMBASE, and ISI Web of Science were searched for studies estimating the influence of central obesity on NAFLD occurrence published through April 2014. Studies that did not adjust for body mass index (BMI) were excluded. In addition, the independent effect of BMI was also assessed with the included studies. The pooled effect sizes and 95% confidence intervals (CIs) were calculated using random- or fixed-effects models based on the degree of heterogeneity. Furthermore, subgroup analyses, meta-regression, sensitivity analyses, and publication bias were performed. RESULTS: Twenty eligible studies were identified. The summary odds ratio (OR) values per-unit increase in waist circumference (WC) and BMI for NAFLD formation were 1.07 (95% CI: 1.03-1.10, I-2 = 73.9%, n = 11 studies) and 1.25 (95% CI: 1.13-1.38, I-2 = 88.7%, n = 11 studies), respectively. When the indices were expressed as binary variables (with the non-obesity group as reference), the pooled OR in WC, waist-to-hip ratio, and BMI were 2.34 (95% CI: 1.83-3.00, I-2 = 41.8%, n = 7 studies), 4.06 (95% CI: 1.53-10.79, I-2 = 65.7%, n = 3 studies), and 2.85 (95% CI: 1.60-5.08, I-2 = 57.8%, n = 5 studies), respectively. Using the same studies as the latter (n = 5), pooled OR in WC was 3.14 (95% CI: 2.07-4.77), which is greater than that in BMI. CONCLUSION: Central obesity may pose a greater threat to national health than general obesity, although both are independently associated with increased risk of NAFLD.