Association between triglyceride glucose-body mass index and long-term adverse outcomes in individuals with heart failure: a retrospective cohort study.

Background: Insulin resistance (IR) plays a pivotal role in the pathogenesis and progression of heart failure (HF), and may be exacerbated in advanced stages of HF, thereby perpetuating a deleterious cycle. The triglyceride glucose-body mass index (TyG-BMI), an established surrogate marker for IR, has been associated with adverse cardiovascular events. However, its prognostic significance in individuals diagnosed with HF has not been fully elucidated. The specific aim is to evaluate the association between the TyG-BMI index and long-term all-cause mortality and HF-rehospitalization in patients with heart failure, and to examine whether right ventricular dysfunction (RVD) modifies this association. Methods: This retrospective cohort study included 1,644 participants hospitalized with HF at Fuwai Central China Cardiovascular Hospital, of whom 850 (51.7%) had HF with preserved ejection fraction (HFpEF). Participants were stratified into tertiles based on TyG-BMI values. The primary composite endpoint included all-cause mortality and HF-related rehospitalization. Multivariable Cox proportional hazards models and restricted cubic spline analyses were employed to assess associations, with survival probabilities visualized using Kaplan-Meier estimates. Results: Over a median follow-up period of 52 months, 415 individuals died from all causes, and 479 experienced HF-related rehospitalization. TyG-BMI exhibited a U- or J-shaped association with long-term outcomes: both the lowest and highest TyG-BMI tertiles carried significantly higher risks of all-cause mortality and HF rehospitalization compared with the intermediate tertile, indicating that risk was not merely elevated at low values but followed a non-linear U- or J-shaped curve (p < 0.001). Restricted cubic spline analysis demonstrated an inverse J-shaped relationship between TyG-BMI and the composite endpoint, with inflection points at 193.10 for mortality and 222.30 for rehospitalization. Individuals in the lowest TyG-BMI tertile demonstrated a markedly elevated risk of the composite outcome (hazard ratio [HR]: 2.41, 95% confidence interval [CI]: 1.96-2.96, p < 0.001). These associations remained consistent in the HFpEF subgroup (p for interaction > 0.05). Additionally, TyG-BMI was strongly associated with all-cause mortality among individuals with concomitant right ventricular dysfunction (RVD) (HR: 2.34, 95% CI: 1.55-3.54, p < 0.001). Conclusion: A U-shaped association was observed: both very low and very high TyG-BMI were associated with increased all-cause mortality and HF rehospitalization. These findings support the potential utility of TyG-BMI as a prognostic biomarker for risk stratification. Furthermore, TyG-BMI appears to modify the relationship between RVD and adverse clinical outcomes, with obesity potentially exerting a modulatory effect.