Frontiers in Cardiovascular Medicine · Oct 29, 2026
Beyond the trigger: the unknown precipitant as a high-risk phenotype in acute heart failure
By Aletras G, Bachlitzanaki M, Stratinaki M et al.
Background: A substantial proportion of patients with acute heart failure (AHF) present without an identifiable precipitating factor. We compared the clinical profile, echocardiographic characteristics, and mid-term outcomes of AHF patients stratified by precipitating factor category. Methods: We conducted a prospective single-center observational study at Venizelio General Hospital, Heraklion, including 486 consecutive patients hospitalized with AHF between February 2023 and January 2025. Patients were classified into three groups: behavioral ( n = 148), non-behavioral/clinical ( n = 230), and unknown ( n = 108). The primary endpoint was all-cause death at 12 months; secondary endpoints were terminal renal replacement therapy (RRT) and a composite of death, AHF rehospitalization or RRT. Groups were compared by Kaplan–Meier analysis and Cox regression with the behavioral group as reference, and the cumulative incidence of first AHF rehospitalization was estimated with death and terminal RRT as competing events. Results: The unknown group was older and more frail (CFS ≥5), with more advanced chronic kidney disease, more right heart failure and lower TAPSE, higher NT-proBNP and lower admission eGFR. All-cause mortality at 12 months was 41.7% versus 20.3% and 29.6% (log-rank p < 0.001; unknown vs. behavioral HR 2.44, 95% CI 1.54–3.88), with a parallel gradient for the composite endpoint (event-free survival 35.2% vs. 58.8% and 52.6%; HR 1.96, 1.39–2.77). After adjustment for age, admission eGFR, right heart failure, NT-proBNP and frailty, precipitant category no longer improved model fit (likelihood ratio p = 0.066) and the association of the unknown group with mortality was attenuated and no longer significant (adjusted HR 1.43, 0.88–2.31). With death and terminal RRT as competing events, 12-month cumulative incidence of first AHF rehospitalization did not differ (36.1% vs. 29.7% and 26.1%; p = 0.207). ACEi/ARB prescription at discharge was lower in the unknown group (21.2% vs. 39.0% and 44.2%, p < 0.001). Conclusions: In AHF, the absence of an identifiable precipitating factor identifies a high-risk phenotype characterized by advanced multimorbidity, right ventricular dysfunction, renal impairment and worse mid-term outcomes. The difference was driven predominantly by mortality rather than rehospitalization, and the association was attenuated after adjustment for the phenotype-defining variables. The label may therefore serve as an early bedside marker of advanced disease, requiring confirmation in multicenter cohorts.