Abstract
BACKGROUND: Perioperative hypoglycemia during digestive endoscopy poses significant risks, with reported incidence rates of 6.75-18.70%. Traditional models focus on metabolic factors, while psychosocial stressors (anxiety, poor sleep quality) may exacerbate risk through neuroendocrine dysregulation. This study aims to develop a novel nomogram (DREAM-Hypo Model) integrating psychosocial, clinical, and metabolic predictors for hypoglycemia risk stratification. METHODS: We prospectively enrolled 610 patients undergoing therapeutic digestive endoscopy at The First Affiliated Hospital of Ningbo University, China (Jan-Jun 2024). Participants were randomly allocated to training and validation cohorts (7:3 ratio). Predictors were identified via univariate and multivariate logistic regression. Model performance was assessed using ROC curves, calibration plots, and decision curve analysis (DCA), with internal validation via 500 bootstrap resamples. METHODS: We prospectively enrolled 610 patients undergoing therapeutic digestive endoscopy at The First Affiliated Hospital of Ningbo University, China (Jan-Jun 2024). Participants were randomly allocated to training and validation cohorts (7:3 ratio). Predictors were identified via univariate and multivariate logistic regression. Model performance was assessed using ROC curves, calibration plots, and decision curve analysis (DCA), with internal validation via 500 bootstrap resamples. RESULTS: The overall hypoglycemia incidence was 14.92%. Six independent predictors were identified: fasting time, anxiety, sleep quality, male gender, diabetes, and somatostatin use (all p<0.05). The nomogram showed strong discrimination (training cohort AUC: 0.830, 95% CI: 0.795-0.865; validation cohort AUC: 0.809, 95% CI: 0.752-0.866). DCA confirmed clinical utility, with superior net benefits at thresholds of 25-30%. CONCLUSION: The DREAM-Hypo nomogram effectively integrates metabolic, procedural, and psychosocial factors to predict perioperative hypoglycemia risk, supporting individualized risk stratification and targeted interventions. TRIAL REGISTRATION: ClinicalTrials.gov, NCT07132476. Registered on August 19, 2025.National Medical Research Registration and Filing Information System, MR-3-25-035244. Registered on May 7, 2025.