The predictive value of the triglyceride-glucose index for five-year all-cause mortality differs between patients with acute de novo heart failure and those with acute decompensated chronic heart failure

Introduction
De novo heart failure (DNHF) and acute decompensated chronic heart failure (ADCHF) represent two distinct subtypes of acute heart failure (AHF).

Aim
To investigate whether the predictive value of the triglyceride-glucose index (TyG) for 5-year all-cause mortality differs between patients with DNHF and those with ADCHF.

Material and methods
1160 patients out of 1740 hospitalized AHF patients were recruited and were classified into DNHF (n = 536) and ADCHF (n = 624) groups. All participants were followed up for a 5-year period, with the primary endpoint being all-cause mortality.

Results
DNHF patients had lower TyG levels than ADCHF patients (p < 0.001). Higher TyG levels were associated with increased 5-year mortality in ADCHF patients (log-rank p < 0.001), but not in DNHF patients (log-rank p = 0.97). TyG acted as an independent predictor of mortality in ADCHF patients (adjusted HR = 2.72 [95% CI: 1.93–3.85], p < 0.001), but not in DNHF patients (unadjusted HR = 1.15 [95% CI: 0.68–1.94], p = 0.61). In ADCHF patients, TyG had a good predictive power for 5-year mortality (AUC = 0.852; 95% CI: 0.821–0.884) and was superior to EFFECT risk of death score and NT-proBNP. TyG was not significantly associated with 5-year all-cause mortality in DNHF patients (non-linear p = 0.19). However, in ADCHF patients, when the TyG value exceeded 10.36, the risk of all-cause mortality in non-diabetic AHF patients was significantly increased (HR increased 2.96, 95% CI: 2.48–3.52, non-linear p = 0.024).

Conclusions
Among non-diabetic patients, TyG index was significantly associated with 5-year all-cause mortality in ADCHF patients but not in DNHF patients. Differentiating the predictive value of TyG between DNHF and ADCHF patients could enhance our understanding and refine treatment strategies for individuals with AHF.