Diet and Risk of Non-Alcoholic Fatty Liver Disease, Cirrhosis, and Liver Cancer: A Large Prospective Cohort Study in UK Biobank

Abstract

Background and Aims: Epidemiological evidence has shown the association between nutritional habits and liver disease. However, results remain conflicting. This study investigated the influence of dietary factors on the risk of incident non-alcoholic fatty liver disease (NAFLD), cirrhosis, and liver cancer. Methods: Data from the UK Biobank database were analyzed (n = 372,492). According to baseline data from the food frequency questionnaire, two main dietary patterns (Western and prudent) were identified using principal component analysis. We used cox proportional hazards models to explore the associations of individual food groups and dietary patterns with NAFLD, cirrhosis, and liver cancer. Results: During a median follow-up of 12 years, 3527 hospitalized NAFLD, 1643 cirrhosis, and 669 liver cancer cases were recorded among 372,492 participants without prior history of cancer or chronic liver diseases at baseline. In multivariable adjusted analysis, participants in the high tertile of Western dietary pattern score had an 18% (95%CI = 1.09−1.29), 21% (95%CI = 1.07−1.37), and 24% (95%CI = 1.02−1.50) higher risk of incident NAFLD, liver cirrhosis, and liver cancer, respectively, compared with the low tertile. Participants in the high tertile of prudent scores had a 15% (95%CI = 0.75−0.96) lower risk of cirrhosis, as compared with those in the low tertile. In addition, the higher consumption of red meat and the lower consumption of fruit, cereal, tea, and dietary fiber were significantly associated with a higher risk of NAFLD, cirrhosis, and liver cancer (ptrend < 0.05). Conclusions: This large prospective cohort study showed that an increased intake of food from the Western dietary pattern could be correlated with an increased risk of chronic liver diseases, while the prudent pattern was only correlated with a reduced liver cirrhosis risk. These data may provide new insights into lifestyle interventions for the prevention of chronical liver diseases.

Keywords: cirrhosis; cohort study; dietary pattern; liver cancer; non-alcoholic fatty liver disease.

Figure 1

Hazard ratios for the risks of different liver diseases by dietary patterns (tertiles) in UK Biobank. HRs and 95% CIs were adjusted for age, sex, race, education level, Townsend deprivation index (quartiles), drinking status, smoking status, exercise, BMI, and diabetes. Definition of abbreviations: NAFLD, non-alcoholic fatty liver disease; HR, hazard ratio; 95%CI, 95% confidence interval; SD, standard deviation.