Development and Validation of a Prediction Model for Microvascular Complications of Type 2 Diabetes Based on Inflammation-Metabolism Composite Indicators

Objective This study evaluated the predictive value of novel inflammatory and metabolic composite indices—including NLR, MLR, SIRI, SII, AIP, TyG, and TyG-BMI—for microvascular complications of type 2 diabetes mellitus (DMC) and developed a corresponding risk prediction model. Methods Clinical data from 964 hospitalized T2DM patients at the First Affiliated Hospital of Xinjiang Medical University (September 2023–September 2025) were retrospectively analyzed and classified according to the presence of DMC. Group differences in demographic and clinical parameters were compared. LASSO regression was used to identify key predictors, after which the dataset was divided into training and validation sets (7:3 ratio). A multivariable Logistic regression nomogram was constructed in the training set and evaluated using ROC curves, calibration curves, and decision curve analysis (DCA). Result (i) The DMC group showed significantly higher NLR, MLR, SIRI, and SII, and significantly lower AIP, TyG, and TyG-BMI compared with the non-DMC group (all P < 0.05), indicating heightened systemic inflammation and metabolic disturbances.(ii) Multivariate analysis identified hypertension,urea nitrogen,albumin, alanine aminotransferase, osteocalcin, parathyroid hormone, SIRI, and TyG-BMI as independent predictors of DMC.(iii) The LASSO-Logistic model achieved an AUC of 0.80 (95% CI: 0.76–0.83) in the training set and 0.73 (95% CI: 0.67–0.78) in the validation set, outperforming SIRI or TyG-BMI alone. Calibration curves demonstrated good agreement between predicted and observed risks, and DCA confirmed favorable clinical utility across multiple thresholds. Conclusion Inflammatory and metabolic composite indices play a significant role in identifying DMC risk. SIRI and TyG-BMI were confirmed as key independent predictors. The LASSO-Logistic nomogram demonstrated reliable discrimination and clinical applicability, offering an effective tool for early detection and risk stratification in T2DM patients.