Background: A newly developed dietary inflammatory index (DII) to evaluate the inflammatory potential of diets was published recently. Many studies have investigated the link between diet-related inflammation and human cancer risk, but the results remain controversial.
Objective: We sought to determine the dose-response relation between DII and human cancer risk based on published epidemiologic literature.
Design: To summarize evidence, we performed a dose-response meta-analysis to investigate the association between DII and cancer incidence. We systematically searched PubMed, Embase, Web of Science, and the Cochrane library up to 5 November 2017. After data extraction, pooled RRs were calculated and dose-response analyses were performed using a restricted cubic spline model with 4 knots. Subgroup analyses, sensitivity analyses, and tests for publication bias were also performed.
Results: In all, 44 high-quality studies with 1,082,092 participants were included. The results showed that an elevated DII (continuous-RR: 1.13; 95% CI: 1.09, 1.16; category DIIhighest vs lowest-RR: 1.58; 95% CI: 1.45, 1.72) independently indicated higher cancer risk except for lung cancer and Australian studies. A linear dose-response relation between DII and overall cancer risk was found, with an 8.3% increase in the risk of cancer per DII score. The pooled RR of DII and cancer risk was 1.86 (95% CI: 1.63, 2.13) from 30 case-control studies but was lower in 14 prospective cohorts (RR: 1.29; 95% CI: 1.19, 1.40). The sensitivity analysis and Egger's test supported the main results.
Conclusions: Our analysis indicated that higher DII is significantly correlated with cancer risk. More prospective studies with large sample sizes, involving more ethnic groups and different cancer types, are required in the future. This review was registered with PROSPERO as CRD42017077075.