Background and aimsAdiposity is associated with an increased risk of primary liver cancer (PLC). As the most commonly used indicator of adiposity, the body mass index (BMI) has been questioned for its limitations in reflecting visceral fat. This study aimed to investigate the role of different anthropometric indicators in identifying the risk of PLC by accounting for potential non-linear associations. MethodsSystematic searches were conducted in the PubMed, Embase, Cochrane Library, Sinomed, Web of Science, and CNKI databases. Hazard ratios (HRs) and the corresponding 95% confidence intervals (CIs) were used to assess the pooled risk. The dose-response relationship was assessed using a restricted cubic spline model. ResultsSixty-nine studies involving more than 30 million participants were included in the final analysis. Regardless of the indicator used, adiposity was strongly associated with an increased risk of PLC. When comparing the HRs per 1-standard deviation increment across indicators of adiposity, the association was strongest for waist-to-height ratio (WHtR) (HR = 1.39), followed by waist-to-hip ratio (WHR) (HR = 1.22), BMI (HR = 1.13), waist circumference (WC) (HR = 1.12), and hip circumference (HC) (HR = 1.12). A strong non-linear association was observed between each anthropometric parameter and the risk of PLC, regardless of whether the original or decentralised value was used. The positive association between WC and PLC risk remained substantial after adjusting for BMI. The incidence of PLC was higher with central adiposity (52.89 per 100,000 person-years, 95% CI = 50.33–55.44) than general adiposity (39.01 per 100,000 person-years, 95% CI = 37.26–40.75). ConclusionCentral adiposity seems to contribute more to the development of PLC than general adiposity. A larger WC, independent of BMI, was strongly associated with the risk of PLC and might be a more promising predictive indicator than BMI.
Read full abstract