Background: The utilization of peritoneal dialysis (PD) has been growing rapidly in the last decade, and there is a need to identify predictors and associated factors for the prognosis of patients on PD. Lipid‐related indices are recently found to reflect metabolic disturbance in normal population, but their relevance in the PD population has not been explored. Methods: We recruited 140 incident PD patients. Their baseline body anthropometry, bioimpedance spectroscopy parameters, and nutrition indices were measured. Survival and hospitalization were analyzed by multivariable Cox and linear regression models. Results: There were significant internal correlations between TyG index, LAP, and VAI but not conicity index. Patients with higher TyG index were more likely to have diabetes, higher Charlson’s comorbidity score, waist circumference, waist‐to‐hip ratio, and subscapular skin fold. They also had higher total cholesterol, triglyceride, low‐density lipoprotein, fasting glucose, HbA1C, and HOMA‐IR. All four adiposity indices showed modest but significant correlations with the Charlson’s comorbidity score. Higher quartiles of the adiposity indices were associated with worse patient survival by univariate analysis, but this effect was not independent when adjusted for other clinical confounders. The adiposity indices, except the conicity index, had modest but statistically significant correlations with hospital admission and duration of hospitalization, with the TyG index showing the most prominent impact ( β = 0.19, p < 0.0001 and β = 0.29, p = 0.0003, respectively). Conclusions: The lipid‐related indices reflect metabolic disturbance and adiposity in PD patients, and their prognostic significance deserves further studies. Our result suggests that the TyG index may be the most promising prognostic indicator among the four indices.
Yang et al. (Wed,) studied this question.