Diet Quality and its Association With Nonalcoholic Fatty Liver Disease and All-cause and Cause-specific Mortality

Eric R. Yoo; Donghee Kim; Luis M. Vazquez-Montesino; Jessica A. Escober; Andrew A. Li; Sean P. Tighe; Christopher T. Fernandes; George Cholankeril; Aijaz Ahmed

Disclosures

Liver International. 2020;40(4):815-824. 

In This Article

Results

A total of 10 858 subjects were eligible to be included in the analysis. The mean age was 42.9 years and 47.1% of the subjects were men. The prevalence of NAFLD was 34.4%, including 20.4% with moderate to severe NAFLD. As shown in Table 1, subjects with poor diet quality (quartile 1) were more likely to be younger, non-Hispanic blacks, current smokers, more caffeine consumption, sedentary and less educated compared to those with higher diet quality (quartile 4). Subjects with poor diet quality also had a higher body mass index (BMI), waist circumference and higher plasma concentrations of C-reactive protein compared to those with higher diet quality. There was no significant difference in the percentage of diabetes, fasting glucose levels, high-density lipoprotein cholesterol, HOMA-IR, haemoglobin A1c, alanine aminotransferase and alcohol consumption among the four quartiles.

We analysed cross-sectional association between NAFLD and quartiles of the HEI score and individual components (Table 2). Higher diet quality was associated with progressively lower odds of having NAFLD in the age, sex and Race/ethnicity-adjusted model (odds ratio [OR] 0.84 95% confidence intervals 0.71–0.99 for quartile 4 [highest diet quality] compared to quartile 1 [lowest diet quality], OR 0.96, 95% CI 0.90–0.99 for 1-SD of HEI score). The multivariable model was adjusted for age, sex, Race/ethnicity, education level, economic status, BMI, smoking status, diabetes, hypertension, caffeine consumption, alanine aminotransferase, alcohol consumption, cholesterol, high-density lipoprotein cholesterol, C-reactive protein, transferrin saturation, sedentary lifestyle and total calorie intake. In this multivariable model, higher diet quality was associated with significantly lower odds of presence of NAFLD, and the adjusted OR and 95% CI for the higher diet quality compared with the lowest diet quality was 0.70 (0.50–0.98, P for trend = .028). Within the individual components of the HEI score, higher consumption of fruits and vegetables and reduction of daily sodium were significantly associated with lower odds of NAFLD (OR 0.87, 95% CI 0.77–0.98 for fruits; OR 0.87, 95% CI 0.76–0.99 for vegetables; OR 0.86, 95% CI 0.76–0.98 for sodium per 1-SD increase).

During a median follow-up period of 23 years, there were 3260 deaths among 10 858 subjects. As shown in Table 3, higher diet quality was associated with a reduction in the risk of all-cause mortality in the age, sex, Race/ethnicity-adjusted model (hazard ratio [HR] 0.70, 95% CI 0.57–0.85 for quartile 3; HR 0.60, 95% CI 0.52–0.68 for quartile 4, P for trend < .001). In the multivariable model adjusted for known demographic variables and traditional risk factors, the HRs did not change materially compared to the unadjusted estimates. Subjects with high diet quality had a 19% decrease in all-cause mortality (HR 0.81, 95% CI 0.71–0.92 for quartile 4, P = .002). When we performed sensitivity analysis using the HEI score as a continuous variable, the results remained identical (HR 0.94, 95% CI 0.91–0.98 per 1-SD increase, P = .005). Among individual component of HEI, higher consumption of fruit was significantly associated with a lower risk of all-cause mortality (HR 0.93, 95% CI 0.89–0.98 for fruit per 1-SD increase). When we classified groups between subjects with NAFLD and without NAFLD (Table 4), higher diet quality was associated with a lower risk for all-cause mortality in subjects without NAFLD (HR 0.73, 95% CI 0.62–0.86 for quartile 4 [P for trend < .001]; HR 0.91, 95% CI 0.86–0.96 for 1-SD increase); however, this protective association was not seen in those with NAFLD, regardless of their diet quality. Within the individual components of the HEI score, higher consumption of fruits and lower consumption of cholesterol were significantly associated with lower risk for all-cause mortality in subjects without NAFLD (HR 0.92, 95% CI 0.87–0.98 for fruits; HR 0.90, 95% CI 0.84–0.97 for cholesterol per 1-SD increase).

When the analysis was restricted to specific causes (Table 5), there was no significant difference in cardiovascular disease-related mortality when comparing diet quality among the total population, those without NAFLD and those with NAFLD. There was an association with lower risk for cancer-related mortality in those with high diet quality for the total population (HR 0.64, 95% CI 0.48–0.85 [quartile 4], P for trend = .001), and for subjects without NAFLD (HR 0.52, 95% CI 0.35–0.80 [quartile 4], P for trend = .005), but this association was not seen in those with NAFLD (P for trend = .460). Subjects with high diet quality had a 36% reduced risk in cancer-related mortality compared to those with poor diet quality.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....