The Construction and Validation of a Risk Prediction Model for Perioperative Hypoglycemia in Digestive Endoscopy

Overview

Digestive endoscopic surgery, including diagnosis and treatment interventions, has become an indispensable means of managing various gastrointestinal diseases in modern gastroenterology practice. Perioperative hypoglycemia is a common but not fully recognized complication. The reported incidence rate of different study populations and types of surgery ranges from 6.75% to 17.50%. The clinical consequences of endoscopic hypoglycemia are profound. Recurrent attacks are associated with an increased risk of transient ischemic attacks and other cardiovascular and cerebrovascular events. From the perspective of the healthcare system, hypoglycemia may reduce patients' compliance with the intake of intestinal cleaning agents, leading to inadequate intestinal preparation, affecting the effectiveness of diagnosis and treatment, prolonging endoscopic surgery time, increasing hospitalization days and costs. This study aims to develop and validate a comprehensive nomogram that integrates three key dimensions: traditional metabolic risk factors (diabetes status, fasting time), program variables (infusion rate, somatostatin use), and new psychosocial factors (anxiety, sleep quality). By using machine learning enhanced logistic regression to prospectively collect data from 610 patients from multiple centers, the resulting column chart model provides clinical doctors with a practical tool for personalized risk stratification and targeted intervention measures, such as reducing anxiety protocols or optimizing fluid management for high-risk patients. The purpose of this study is to identify the influencing factors of hypoglycemia during the perioperative period of digestive endoscopy, help identify and screen high-risk populations, provide theoretical basis for the development of intervention measures, reduce the occurrence of hypoglycemia during the perioperative period of endoscopy, improve patient satisfaction, and reduce hospitalization time and costs.

Conditions