The association between dietary inflammatory index (DII) scores and c-reactive protein (CRP) and nonalcoholic fatty liver disease (NAFLD) in a general population cohort

Abstract

Background and aims: Although there is extensive literature showing the ability of the dietary inflammation index (DII®) to predict concentrations of plasma inflammatory markers, few studies are testing the association between DII scores and nonalcoholic fatty liver disease (NAFLD). Considering the high prevalence of NAFLD and its complications, we conducted a validation study of DII scores and examined its association with NAFLD in the general adult population of Iran.

Methods: This cross-sectional study was conducted on 3110 adult participants in the Amol Cohort Study (AmolCS) who underwent abdominal ultrasonography to diagnose NAFLD. DII and energy-adjusted DII (E-DII™) scores were computed using data from a valid semi-quantitative 168-item food frequency questionnaire (FFQ). Multivariable logistic regression adjusting for socio-demographic, lifestyle, and health-related factors was used to assess association.

Results: The EDII was associated with CRP inflammatory biomarker. Participants in the highest, i.e., most pro-inflammatory tertile had the highest odds of NAFLD by ultrasound in all models [fully adjusted model: OR (95 % CI) _tertile3vs.1:1.54 (1.05-2.05); P_trend = 0.04, and 1.63 (1.19-2.21); P_trend = 0.03 in women and men, respectively]. The highest tertile had the highest OR for NAFLD by fatty liver index (FLI) only in men [fully adjusted model OR (95 % CI) _tertile3vs.1: 1.77 (1.15-2.71); P_trend = 0.01]. Similar results were also obtained for NAFLD by hepatic steatosis index (HSI) in women [fully adjusted model: OR (95 % CI) _tertile3vs.1: 1.70 (1.12-2.58); P_trend = 0.03]. The results of the fully adjusted multivariable model of liver markers and NAFLD status, stratified by gender and abdominal obesity, revealed that the highest tertiles had the highest OR for NAFLD by ultrasound and NAFLD by FLI only in men without abdominal obesity [fully adjusted model: OR (95 % CI) _tertile3vs.1: 1.83 (1.17-2.84); P_trend = 0.03, and, respectively]. NAFLD by FLI tended to increase strongly with tertile E-DII scores in men without abdominal obesity in crude and three adjusted models [full-adjusted model: OR (95 % CI) _tertile3vs.1: 3.64 (1.56-8.46); P_trend = 0.005]. By contrast, women with abdominal obesity in the highest tertile had the highest OR for NAFLD by ultrasound in all models [full-adjusted model: OR (95 % CI) _tertile3vs.1: 1.67 (1.07-2.62); P_trend = 0.02].

Conclusions: Our results suggest that diet plays a role in regulating inflammation. Additionally, we observed an inflammatory diet predicts the risk of NAFLD in Iranian adults. However, longitudinal studies are required in order to further substantiate the utility of the DII in the development of more effective dietary interventions among populations at risk of chronic disease.

Keywords: Diet; Dietary inflammatory index; Inflammation; Iranian cohort; NAFLD.