Background and Aims: Diet as a modifiable factor for remission of fatty liver disease (FLD) is rarely investigated. Therefore, this study investigated the association between FLD remission and 18 food items consumption.

Methods: This is a prospective open cohort study. Participants included 11,927 health examination participants from the Beijing MJ healthcare aged ≥ 18 years with diagnosis of FL by abdominal ultrasound examination at baseline. Semi-quantitative food frequency questionnaire was used to assess 18 food items intake during the follow-up. The percentage of energy from food items were divided by quartile. Hazard ratios (HRs) and 95% CIs of FLD remission across food items consumption were estimated with Cox proportional hazard regression models, adjusted for age, sex, total energy, and major confounding factors, and further stratified by non-alcoholic fatty liver disease (NAFLD).

Results: During a median follow-up of 1.3 years (interquartile range, 1.0-2.5), 4,541 individuals occurred FLD remission. The HRs (95%CI) for FLD remission comparing the highest with the lowest quartile of food items group were 1.19(1.09-1.30) for milk, 1.15(1.06-1.26) for eggs, 0.89(0.82-0.98) for fishery products, 0.89(0.82-0.98) for dark green vegetable, 1.19(1.09-1.31) for bread, 0.83(0.76-0.90) for sugar sweeten beverage, 0.82(0.74-0.91) for ultra processed food. Egg, fishery products, dark green vegetable, bread, and ultra processed food were all displayed linear dose-response relationships with FLD remission (P for non-linearity >0.05). Similar results were found by NAFLD.

Conclusions: A high intakes from milk, egg, dark green vegetables, and bread, and a low intakes from fishery products, sugar-sweetened beverages, and ultra-processed food can independently improve fatty liver. These findings emphasize that patients can modify their dietary intake to alleviate FLD.

Disclosure

N. Yi: None. Z. Dehao: None. B. Bi: None.

Funding

Research Foundation for Advanced Talents of Hainan (820RC649)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.