Development and Validation of a Preoperative Glycolipid Metabolism-Based Nomogram for Predicting Postoperative Recurrence in Primary Glioma: A Retrospective Cohort Study

Background The high postoperative recurrence rate of glioma poses a significant challenge to patient survival. Metabolic reprogramming, particularly abnormalities in glycolipid metabolism, plays a critical role in tumor progression. While indicators such as the triglyceride-glucose(TyG) index are reliable markers of insulin resistance and metabolic dysregulation with established prognostic value in several cancers, their utility in predicting glioma recurrence remains largely unexplored. Methods This retrospective study included 302 primary glioma patients who underwent resection at Linyi People's Hospital (2016–2024). Patients from one ward (n = 236) were randomly divided into training (n = 142) and internal validation (n = 94) cohorts. An independent external validation cohort (n = 66) was formed from another ward. Key glycolipid markers were analyzed using maximum selection rank statistics to determine optimal cut-offs. Significant markers were integrated via Least Absolute Shrinkage and Selection Operator (LASSO) regression into a composite LASSO-score, which was then combined with clinicopathological variables in a multivariate Cox regression to build a nomogram predicting 1-, 2-, and 3-year recurrence-free survival (RFS). Model performance was evaluated using the Concordance Index (C-index), time-dependent receiver operating characteristic curves(ROC), calibration curves, and decision curve analysis(DCA). Results During a median follow-up of 18.37 months in the training cohort, 97 recurrence events were recorded. Key metabolic markers, including TyG, TG/HDL-C, and TyG-BMI, showed significant associations with RFS (P < 0.05). The LASSO-score, alongside radiotherapy and Ki-67 expression, was identified as an independent predictor of RFS. The nomogram demonstrated consistent predictive accuracy across all cohorts, with area under the curve(AUC) values for predicting 1-, 2-, and 3-year RFS being 0.731, 0.706, and 0.760 in the training cohort; 0.742, 0.700, and 0.727 in the internal validation cohort; and 0.780, 0.892, and 0.926 in the external validation cohort, respectively. Calibration curves showed excellent agreement between predictions and observations, and the model effectively stratified patients into distinct risk groups (P < 0.0001). Conclusion We developed and validated a nomogram incorporating preoperative glycolipid metabolic indicators and key clinical variables that reliably predict postoperative recurrence risk in glioma patients. This practical tool shows promise for facilitating individualized patient management and postoperative surveillance.