Adiposity in relation to risks of fatty liver, cirrhosis and liver cancer: a prospective study of 0.5 million Chinese adults

Abstract

Adiposity is an increasing public health problem in China. We aimed to examine the associations of adiposity with non-alcoholic fatty liver disease (NAFLD) and other chronic liver diseases in Chinese adults. The prospective China Kadoorie Biobank recruited 512,891 adults aged 30-79 years from 10 areas. During 10 years of follow-up, 7,386 incident liver disease cases were recorded among 503,991 participants without prior cancer or chronic liver disease at baseline. The mean body mass index (BMI) (SD) was 23.7 (3.3) kg/m² and mean waist circumference (WC) 80.3 (9.8) cm, with 33% having BMI ≥25 kg/m². Throughout the range examined (BMI 15-50) BMI showed a log-linear positive association with NAFLD (n = 1,298), with adjusted HR per 5 kg/m² of 2.81 (95% CI 2.63-3.01), adjusting for regression dilution. There were also positive associations of percent body fat, WC, and waist-to-hip ratio with NAFLD, with HRs per 1-SD of 2.27 (2.14-2.41), 2.60 (2.44-2.76), and 1.84 (1.76-1.92). BMI was unrelated to viral hepatitis (n = 1,477), and had a U-shaped association with cirrhosis (n = 2,082) and an inverse association with liver cancer (n = 2,568), which disappeared after excluding the first 5 years of follow-up. Among Chinese adults, adiposity was a major risk factor for NAFLD but not other chronic liver diseases.

Figure 1

Cumulative incidence of NAFLD and other liver diseases among men and women. Survival curves to show event proportions of participants by age in males (blue) and females (red). Liver diseases include (a) NAFLD, (b) viral hepatitis, (c) cirrhosis, and (d) liver cancer. Curves are truncated at age 85.

Figure 2

Adjusted HRs for hospital-reported NAFLD by usual level of (a) BMI and (b) WC, with the right panel showing the age and sex-specific HRs per 1-SD higher usual BMI and WC. In the left panel, BMI is classified as <20.0, 20.0 to <22.5, 22.5 to <25.0, 25.0 to <27.5, 27.5 to <30.0, and ≥30.0 kg/m². WC is classified by sextiles as <70.7, 70.7–75.2 (reference), 75.3–79.6, 79.7–84.0, 84.1–89.9, and ≥90.0 cm. Hazard ratios (HRs) are plotted against the mean level in each group. Log-scale is used for the y-axis. The squares represent HRs, and the vertical lines represent 95% CIs. Numerical values above the 95% CI represent the hazard ratio and values beneath the 95% CI represent the number of cases of NAFLD in each group. In the right panel, boxes represent sex-specific estimates by age-at-risk categories. Diamonds represent summary HRs for each age-at-risk category and the overall HR. Estimates and 95% CI of the summary HRs are in bold. In both panels, the sizes of the boxes are proportional to the inverse of the variance of the log hazard ratios. The analyses are stratified by sex, region, and HBsAg, and adjusted for age at baseline, education, smoking, alcohol, and total physical activity. SD was 3.3 kg/m² for BMI and 9.8 cm for WC. HR per 1-SD was corrected for regression dilution (regression dilution ratio: BMI = 0.93, WC = 0.84).

Figure 3

Adjusted HRs for NAFLD by usual level of other adiposity traits. Adiposity traits are modelled as sextiles and HRs are plotted against the mean level in each group. All adiposity traits are usual levels, except for BMI at age 25 and weight change since age 25. BMI at age 25 is missing in 81,409 participants. Log-scale is used for the y-axis. The squares represent HRs, and the vertical lines represent 95% CIs. The area of the squares is inversely proportional to the variance of the log HRs. Numerical values above the 95%CI represent the hazard ratio and values beneath the 95% CI represent the number of cases of NAFLD in each group. The analyses are stratified by sex, region, and HBsAg, and adjusted for age at baseline, education, smoking, alcohol, and physical activity. SD was 0.06 for WHR, 6.9 cm for HC, 8.4% for %BF, 10.8 kg for weight, 2.6 kg/m² for BMI25, and 9.1 kg for weight change since age 25. HR per 1-SD was corrected for regression dilution (regression dilution ratio: WHR = 0.70, HC = 0.81, %BF = 0.88, weight = 0.96).