Prediction of recurrent heart failure hospitalizations and mortality using the echocardiographic Killip score

Abstract

Aim: Examine the performance of a simple echocardiographic "Killip score" (eKillip) in predicting heart failure (HF) hospitalizations and mortality after index event of decompensated HF hospitalization. Methods: HF patients hospitalized at our facility between 03/2019–03/2021 who underwent an echocardiography during their index admission were included in this retrospective analysis. The cohort was divided into 4 classes of eKillip according to: stroke volume index (SVI) < 35ml/m2 > and E/E' ratio < 15 >. An eKillip Class I was defined as SVI ≥ 35ml/m2 and E/E' ≤ 15 and was used as reference. Results: Included 751 patients, median age 78.1 (IQR 69.3–86) years, 59% men, left ventricular ejection fraction 45 (IQR 30–60)%, brain natriuretic peptide levels 634 (IQR 331–1222)pg/ml. Compared with eKillip Class I, a graded increase in the combined endpoint of 30-day mortality and rehospitalizations rates was noted: (Class II: HR 1.77, CI 0.95–3.33, p = 0.07; Class III: HR 1.94, CI 1.05–3.6, p = 0.034; Class IV: HR 2.9, CI 1.64–5.13, p < 0.001 respectively), which overall persisted after correction for clinical (Class II: HR 1.682, CI 0.9–3.15, p = 0.105; Class III: HR 2.104, CI 1.13–3.9, p = 0.019; Class IV: HR 2.74, CI 1.54–4.85, p = 0.001 respectively) or echocardiographic parameters (Class II: HR 1.92, CI 1.02–3.63, p = 0.045; Class III: HR 1.54, CI 0.81–2.95, p = 0.189; Class IV: HR 2.04, CI 1.1–3.76, p = 0.023 respectively). Specifically, the eKillip Class IV group comprised one-third of the patient population and persistently showed increased risk of 30-day HF hospitalizations or mortality following multivariate analysis. Conclusion: A simple echocardiographic score can assist identifying high-risk decompensated HF patients for recurrent hospitalizations and mortality.