Abstract Background The triglyceride glucose (TyG) index serves as a dependable marker for insulin resistance and has shown a significant co
Abstract Background The triglyceride glucose (TyG) index serves as a dependable marker for insulin resistance and has shown a significant correlation with the severity of acute pancreatitis (AP). However, no research exists regarding the association between the TyG index and the development of acute respiratory failure (ARF) in AP. This study assesses the association between TyG index and ARF in patients with AP. Methods Retrospective cohort analysis was conducted with the MIMIC-IV 2.2 critical care data. The endpoint focused on ARF during hospitalization. Statistical analysis encompassed univariate and multivariate logistic regressions, alongside restricted cubic spline (RCS) analysis to explore potential nonlinear associations. Receiver operating characteristic (ROC) curve analysis was employed to identify the optimal TyG index cutoff, leading to the classification of patients into Low TyG and High TyG groups. Propensity score matching (PSM) and inverse probability of treatment weighting (IPTW) were subsequently applied to minimize the influence of confounding factors, thereby further clarifying the relationship between the TyG index and ARF in patients with AP. Results A total of 758 patients were involved in this study, the incidence of ARF was 21.64%. Logistic regression analyses demonstrated a significant association between the TyG index and the incidence of ARF in patients with AP. The RCS model illustrated a nonlinear relationship between a higher TyG index and an increased risk of ARF. The cutoff value of TyG index was 9.099 for ARF in patients with AP based on the ROC curve analysis. Furthermore, following PSM and IPTW, multivariate logistic regression analysis indicated that the High TyG group exhibited a significantly higher risk of ARF compared to the Low TyG group (P
