J-shaped relationship of the triglyceride-glucose index with all-cause mortality in initial-hemodialysis patients in China：a multicentre, retrospective cohort study.

Introduction: The relationship between the triglyceride-glucose (TyG) index and mortality in hemodialysis patients remains uncertain. This study aimed to investigate the correlation between TyG index and all-cause mortality in initial-hemodialysis patients in China.

Methods: 783 patients participated in the study and were grouped into quintiles according to the TyG index. Multivariate Cox models and subgroup analyses were utilized. Nonlinear correlations were explored using restricted cubic splines, and a two-piecewise Cox proportional hazards model was developed around the inflection point.

Results: During a median follow-up of 44 months, 231 (29.50%) patients occurred mortality. Multivariate Cox regression confirmed that both lower and higher TyG indices independently predicted all-cause mortality (all P < 0.05). The predictive value of a high TyG index for all-cause mortality remained consistent across age, sex, BMI, and diabetes subgroups. A restricted cubic spline unveiled a J-shaped relationship between the two variables in initial-hemodialysis patients. A TyG index exceeding 8.83 exhibited a positive correlation with all-cause mortality (HR, 1.78; 95% CI: 1.27-2.46, P < 0.001).

Conclusions: A J-shaped relatonship was identified between the TyG index and all-cause mortality in initial-hemodialysis patients in China. The threshold of 8.83 for all-cause mortality.