Dietary inflammatory index and gestational diabetes mellitus in a multicenter cross sectional study in Suzhou

Background Evidence linking dietary inflammatory potential with gestational diabetes mellitus (GDM) has been inconsistent. We examined this association in a multicenter cross-sectional study. Methods Women attending three tertiary hospitals in Suzhou, China, from February to November 2025 completed a food-frequency questionnaire at 24 to 28 gestational weeks, recalling the preceding three months. A 26-parameter Dietary Inflammatory Index (DII) was calculated, and GDM was defined from 75-g oral glucose tolerance test (OGTT) values. Logistic regression adjusted for demographic, socioeconomic, obstetric, behavioral, and hospital covariates. Results The primary analysis included 1,071 women after removal of 23 exact duplicate records; 225 (21.0%) had GDM. Mean age was 29.28 years (SD 4.17), prepregnancy body mass index was 22.42 kg/m² (SD 3.28), and DII was 0.71 (SD 1.13). Each one-SD higher DII was associated with higher adjusted GDM odds (OR 1.26, 95% CI 1.06 to 1.51; P = 0.011). The adjusted quartile trend was not significant (P = 0.162), and there was no evidence of nonlinearity (P = 0.888). Among 803 women completing the questionnaire one to three days before OGTT, the estimate was attenuated and imprecise (OR 1.20, 95% CI 0.97 to 1.48). Conclusion Higher continuous DII was associated with GDM, but quartile and sensitivity analyses limited the consistency of the findings. Overlapping exposure and outcome windows and measurement limitations precluded causal or predictive interpretation.