Outcomes of subjects who are lean, overweight or obese with nonalcoholic fatty liver disease: A cohort study in China.

The ability to determine the prognosis of lean nonalcoholic fatty liver disease (NAFLD) is essential for decision making in clinical settings. Using a large community-based Chinese cohort, we aimed to investigate NAFLD outcomes by body mass index (BMI). We used the restricted cubic splines method to investigate the dose-response relationship between BMI and outcomes in subjects with NAFLD and those without NAFLD. We included 73,907 subjects from the Kailuan cohort and grouped all subjects into four phenotypes by using NAFLD and BMI (<23 kg/m ). The probability of developing outcomes for individuals with lean NAFLD (LN), overweight/obese NAFLD (ON), overweight/obese non-NAFLD (ONN), and lean non-NAFLD (LNN) was estimated. We found a U-shaped association between BMI and death but a linear positive association concerning cardiovascular disease (CVD) after adjusting for age and other covariates. Compared with the LNN group, the adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of the LN, ON, and ONN groups were 1.30 (1.14-1.49), 0.86 (0.80-0.91), 0.84 (0.80-0.89) for all-cause death, 2.61 (1.13-6.03), 0.74 (0.44-1.26), 1.10 (0.70-1.74) for liver-related death, 2.12 (1.46-3.08), 1.23 (0.99-1.54), 1.19 (0.98-1.43) for digestive system cancers, and 2.04 (1.40-2.96), 1.30 (1.05-1.61), 1.21 (1.01-1.46) for obesity-related cancers. Subjects with LN had a significantly higher risk of colorectal cancer and esophagus cancer. However, the ON group had the highest CVD risk (HR, 1.39; 95% CI, 1.27-1.52). The LN group with hypertension had a higher risk of adverse outcomes, and those without hypertension had a similar risk compared to LNN. Conclusion: Subjects with LN may experience a higher risk of all-cause death, digestive system cancers, and obesity-related cancers than the other three groups but a lower risk of CVD than ON subjects. LN with hypertension may be a high-risk phenotype.