Association between different insulin resistance surrogates and all-cause mortality in peritoneal dialysis patients

Abstract

Background: The insulin resistance (IR) is related with poor outcomes in peritoneal dialysis (PD) patients. This study aimed to explore the association between surrogate IR indexes and all-cause mortality in PD patients. Methods: 1736 PD patients were enrolled from March 2010 to December 2021 in multiple PD centers. Cox proportional hazards models and restricted cubic splines (RCS) regression were constructed to assess the relationship between TyG index, TyG-BMI index, TG/HDL-C ratio and all-cause mortality. All three IR indexes were integrated into same model to access the predictive stability. Furthermore, forest plot was draw to display the result of subgroup analysis in PD patients. Results: 378 mortality events were recorded during a median follow-up time of 2098 days. The RCS regression showed a linear association between the TyG index and TyG-BMI index and all-cause mortality (P for non-linearity 0.566, 0.716 respectively), while a non-linear association between TG/HDL-C ratio and all-cause mortality (P for non-linearity 0.022). Multivariable Cox proportional hazards analyses indicated higher levels of TyG index, TyG-BMI index and TG/HDL-C ratio were independent risk factors for all-cause mortality in PD patients (HR 1.587, 95%CI 1.264-1.993; HR 1.449, 95%CI 1.085-1.934; HR 1.424, 95%CI 1.104-1.838, respectively). By integrating three IR indexes into the same model, the TyG index showed the highest predictive stability among them. Forest plot of TyG index showed there were no significant interactions in the subgroups. Conclusion: The TyG index, TyG-BMI index, and TG/HDL-C ratio were significantly associated with all-cause mortality in PD patients, and the TyG index may be the most stable surrogate marker for IR in PD patients among them, and IR was associated with all-cause mortality in PD patients.