Evaluating Multiple Metabolic Indicators to Predict Gastric Intestinal Metaplasia Risk

Abstract Metabolic syndrome is highly associated with gastric cancer (GC) formation, although the reliability of individual indices for predicting IM (intestinal metaplasia) risk remains inconsistent. This retrospective cohort study applied univariate and multivariate analyses using Python and its statistical packages to analyze the relationships between multiple metabolic indicators and IM, including the Atherogenic Index of Plasma (AIP), the Triglyceride-Glucose Index (TyG), and levels of fasting (TC, AC: Fasting) blood glucose (AC), postprandial blood glucose (PC), total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL), low-density lipoprotein (LDL), and very-low-density lipoprotein (VLDL).Our analysis of the metabolic indicators revealed that TyG and AIP were not predictors of IM. However, across all ages and genders, LDL was a significant predictor of IM. Moreover, we found that the accuracy associated with certain metabolic indicators of IM can vary according to age and gender. More specifically, HDL was a significant indicator of IM in young males, while TC was significant in young females. Additionally, for middle-aged individuals, PC was a significant indicator in males, while AC was significant in females. In elderly males, LDL, VLDL, and TyG were significant indicators, while TC and LDL were significant in elderly females. Furthermore, the AUC of elder individuals (> 60%) was significantly higher compared to young individuals (54.7%, males; 56.5%, females) and middle-aged individuals (53.6%, males; 52.5%, females). By conducting a comprehensive analysis of multiple metabolic indicators, our study reveals that significance varies according to gender and age, although LDL is a significant predictor of IM across all groups.