A modified Dietary Approaches to Stop Hypertension pattern and its association with non-alcoholic hepatic steatosis and fibrosis
OA Version
Citation
Abstract
BACKGROUND: The benefits of a Dietary Approaches to Stop Hypertension (DASH) dietary pattern in association with non-alcoholic fatty liver disease (NAFLD), specifically steatosis and fibrosis, and its association with odds of NAFLD have been studied by few.
OBJECTIVES: This study examined the prospective association between adherence to a newly modified DASH dietary pattern (mDASH) and NAFLD odds, specifically the relative odds of hepatic steatosis and fibrosis, assessed via vibration-controlled transient elastography (VCTE) measures of controlled attenuation parameter (CAP) and liver stiffness measure (LSM), among non-heavy drinking men and women in the Framingham Generation 3 (Gen III), Omni 2, and New Offspring Spouse (NOS) cohorts. In secondary analyses, we explored the modification of the association between mDASH adherence and steatosis by factors such as body mass index (BMI), waist circumference, waist-to-height ratio (WsHtR), physical activity, smoking, alcohol drinking, and cardiometabolic risk (CMR) traits (diabetes, hypertension, dyslipidemia, or an elevated TG: HDL ratio). Sensitivity analyses were also done by adding the group of heavy drinkers (>14 drinks/week females; > 21 drinks/week males) back into the study sample and repeating some primary analyses.
METHODS: The mDASH score was derived from a semi-quantitative food frequency questionnaire collected at examination visit 2 in all three cohorts (Gen III, Omni 2, and NOS). Measures of hepatic steatosis and fibrosis were derived from a VCTE machine called FibroScan, at the third examination visit in all cohorts. Subjects included 2573 participants (1390 females and 1183 males), aged 24 and older, who were free of all cancers except non-melanoma skin, and were not heavy drinkers (>14 drinks/week females; >21 drinks/week males). The mDASH score was classified into tertiles and dichotomized (mDASH score 8.0-21.5 vs. 22.0-39.5) to evaluate the association between mDASH and measures of hepatic steatosis and fibrosis. Multivariable logistic regression models were used to estimate odds ratios (OR) and 95% confidence intervals (CIs) for each of four NAFLD stages (no hepatic steatosis or fibrosis, hepatic steatosis without fibrosis, hepatic fibrosis without steatosis, and hepatic steatosis with fibrosis) for all subjects and females and males separately. Multivariable linear regression models were used to assess the continuous association between mDASH and measures of CAP and log-transformed LSM as indicators of hepatic steatosis and fibrosis, respectively, for all subjects. These multivariable regression models were also used to explore whether the effects of mDASH on CAP (using linear models) or the occurrence of steatosis (using logistic models) were modified by other factors. All analyses were adjusted for confounding by age, sex, education, energy intake, cigarettes per day, and alcohol drinking status. Further adjustments for confounding by BMI, WsHtR, and CMR were also done. Factors found not to confound the effects of mDASH on NAFLD were excluded from the final models.
RESULTS: Overall, 729 individuals met the VCTE criteria for hepatic steatosis (122 of whom had concurrent fibrosis and 607 without any fibrosis) and 96 met the VCTE criteria for fibrosis without steatosis. 1748 individuals had neither steatosis nor fibrosis. Individuals in the middle and highest categories of mDASH had 35% lower odds of any hepatic steatosis (95% CIT2: 0.52-0.81; 95% CIT3:0.51-0.83; p-trend: 0.0008) compared to individuals in the lowest category of mDASH. When stratifying by sex, males had a higher prevalence of steatosis than females. Females in tertile 3 (vs. tertile 1) had 40% lower odds (95% CI: 0.41-0.87) of steatosis versus 37% lower odds (95% CI: 0.45-0.88) for males in tertile 3. However, when further adjusting for BMI, WsHtR, or CMR, mDASH was more protective against steatosis in males than females. Individuals in tertile 2 and tertile 3 (vs. tertile 1) of mDASH had 34% and 29% lower odds (95% CIT2: 0.52-0.84 and 95% CIT3: 0.55-0.92; p-trend: 0.0144) of steatosis without fibrosis. Those in the highest category of mDASH had 56% lower odds of steatosis with fibrosis while those in the middle category of adherence to mDASH had 44% lower odds of steatosis with fibrosis when compared to those in the lowest category of mDASH (95% CIT3: 0.26-0.75; 95% CIT2: 0.35-0.88; p-trend: 0.0023). All odds ratios were attenuated when further adjusting for BMI, WsHtR, or CMR. No association was found between mDASH adherence and the odds of hepatic fibrosis without steatosis. We then examined the associations between mDASH and the continuous measures of steatosis (CAP) and fibrosis (LSM) using linear regression models. For every 5-point increase in mDASH, the CAP measurement was 4.95 dB/m (SE: 1.15; p<0.0001) lower while the log-transformed LSM measure was 0.03 kPa (SE: 0.01; p=0.0007) lower. The associations between mDASH adherence and steatosis were modified by some other factors. Individuals with a low BMI or waist circumference and high adherence to mDASH had the greatest protective effects compared to those with higher BMI or waist circumference, respectively, and low adherence to mDASH (ORLOW BMI+HIGH MDASH 0.13, 95% CI: 0.10-0.17; ORLOW WC+HIGH MDASH: 0.12, 95% CI: 0.09-0.16). There was no statistical interaction between mDASH and these anthropometric measures of body fat. However, overweight individuals (higher BMI) with higher mDASH scores had a 23% lower risk of steatosis than overweight individuals with lower mDASH scores. Thus, mDASH had an independent beneficial effect on risk of steatosis among overweight participants. The effects of mDASH were modified by physical activity. Individuals with higher physical activity levels and greater adherence to mDASH had 48% lower odds of steatosis (95% CI: 0.39-0.68) compared to those with lower mDASH adherence and lower physical activity levels. Even those with lower activity levels had a 27% lower risk of steatosis associated with adherence to the mDASH eating pattern. Higher mDASH adherence was also protective in individuals with light or moderate alcohol consumption. Those with higher mDASH adherence and moderate alcohol consumption (>3 but ≤14 drinks/week for females; >3 but ≤21 drinks/week for males) had 44% lower odds of steatosis (95% CI: 0.41-0.76). High mDASH adherence was also protective in current, former, and nonsmokers, with the greatest protective association being seen in nonsmokers (OR: 0.62; 95% CI: 0.39-1.00). Beneficial effects of mDASH among current and former smokers were weaker and non-statistically significant. Finally, those participants with no cardiometabolic dysfunction were found to have lower relative odds of steatosis, particularly with higher adherence to mDASH. However, even those with cardiometabolic dysfunction seemed to benefit from higher adherence to mDASH (vs. those with cardiometabolic dysfunction with lower adherence to mDASH) (OR: 0.75; 95% CI: 0.59-0.96). When no cardiometabolic traits were present, individuals who were highly adherent to mDASH, had 78% lower odds of steatosis (95% CI: 0.16-0.29).
CONCLUSIONS: In this large prospective cohort study, higher adherence to an mDASH dietary pattern was associated with lower odds of hepatic steatosis with or without fibrosis in females and males aged 24 years or older who were not heavy drinkers (>14 drinks/week in females; >21 drinks/week in males). These beneficial effects were apparent even among certain higher-risk individuals, including those with a higher BMI, less physical activity, and prevalent cardiometabolic dysfunction.
Description
2024