Abstract
OBJECTIVE: This systematic review aimed to investigate the association between dietary inflammatory potential and liver cancer to provide evidence regarding scientific dietary health education. DESIGN: Systematic review and meta-analysis. SETTING: A comprehensive literature review was conducted to identify case-control or cohort studies that involved dietary inflammation index (DII)/empirical dietary inflammation pattern (EDIP) and liver cancer in PubMed, EMBASE, Cochrane, and Web of Science databases. Using a combination of DII/EDIP and liver cancer as the search terms, the associations between DII/EDIP and liver cancer were then assessed. PARTICIPANTS: Three case-control studies and two cohort studies were brought into the meta-analysis, with 225 713 enrolled participants. RESULTS: Meta-analysis of categorical variables showed that DII/EDIP in the highest category increased the risk of liver cancer compared to DII/EDIP in the lowest category (relative risk (RR) = 2·35; 95 % CI 1·77, 3·13; P = 0·000) and with low heterogeneity across studies (I(2) = 40·8 %, P = 0·119). Meta-analysis of continuous variables showed that significant positive association between liver cancer and DII/EDIP scores (RR = 1·24; 95 % CI 1·09, 1·40; P = 0·001), and no heterogeneity (I² = 0·0 %, P = 0·471). Stratified according to the study design, there was a significant positive association between liver cancer and DII/EDIP scores in both cohort studies (RR = 2·16; 95 % CI 1·51, 3·07; P = 0·000) and case-control studies (RR = 2·75; 95 % CI 1·71, 4·41; P = 0·000). CONCLUSION: The higher the DII/EDIP score, the higher the risk of liver cancer. This finding may have prominent implications for the general population.