Relation of dietary insulin index and dietary insulin load to metabolic syndrome depending on the lifestyle factors: Tehran lipid and glucose study

Abstract

Aim: The hypothesis of the effect of the insulinogenic effects of diet on the development of cardiovascular diseases has been proposed, but the findings of previous studies are very contradictory. We investigated the association between dietary insulin index (DII) and dietary insulin load (DIL), and metabolic syndrome (MetS) risk. Another objective was to examine the extent to which lifestyle (physical activity, smoking status, and weight change) and sex influence the relationship between DII, DIL, and MetS risk. Materials and methods: We followed 1915 participants in the Tehran Lipid and Glucose Study. DIL and DII were calculated based on a validated food frequency questionnaire. Weight change was measured, and participants were categorized into > 3% weight loss, weight stable (± 3%), and > 3% weight gain. By joint classification, the association between DII and DIL (< median and ≥ median) and risk of MetS was assessed according to weight change, sex, physical activity levels, and smoking status. Cox proportional hazards models were used to estimate the HRs (95% CI), adjusting for potential confounders. Results: During 8.9 years of follow-up, among 1915 participants, we documented 591 new cases of MetS. DII and DIL were not associated with MetS risk in the crude and adjusted models. However, DIL and DII were associated with weight gain (≥ 3%). In the crude model, DIL and DII were associated with a higher risk of weight gain [HR: 1.74: 95% CI 1.50–2.03, and 1.70 (1.46–1.98), respectively]. These associations remained significant after further adjustment for confounders. The HRs were 1.61 (1.35–1.92) for DIL and 1.64 (1.39–1.93) for DII. Among men, women, participants with low physical activity levels, and smokers, the risk of MetS, independent of DIL and DII, only increased in a participant with weight gain. In women with weight stability, DIL and DII, higher than the median, were positively associated with MetS risk. Conclusion: Our findings suggest that the association between MetS risk and a hyperinsulinemic diet depended on weight change.