Predicting Long-Term Mortality in Patients With Acute Heart Failure by Using Machine Learning

Background: High mortality rates in patients with acute heart failure (AHF) necessitate proper risk stratification. However, risk-assessment tools for long-term mortality are largely lacking. We aimed to develop a machine-learning (ML)-based risk-prediction model for long-term all-cause mortality in patients admitted for AHF.

Methods and results: The ML model, based on boosted a Cox regression algorithm (CoxBoost), was trained with 2704 consecutive patients hospitalized for AHF (median age 73 years, 55% male, and median left ventricular ejection fraction 38%). We selected 27 input variables, including 19 clinical features and 8 echocardiographic parameters, for model development. The best-performing model, along with pre-existing risk scores (BIOSTAT-CHF and AHEAD scores), was validated in an independent test cohort of 1608 patients. During the median 32 months (interquartile range 12-54 months) of the follow-up period, 1050 (38.8%) and 690 (42.9%) deaths occurred in the training and test cohorts, respectively. The area under the receiver operating characteristic curve (AUROC) of the ML model for all-cause mortality at 3 years was 0.761 (95% CI: 0.754-0.767) in the training cohort and 0.760 (95% CI: 0.752-0.768) in the test cohort. The discrimination performance of the ML model significantly outperformed those of the pre-existing risk scores (AUROC 0.714, 95% CI 0.706-0.722 by BIOSTAT-CHF; and 0.681, 95% CI 0.672-0.689 by AHEAD). Risk stratification based on the ML model identified patients at high mortality risk regardless of heart failure phenotypes.

Conclusions: The ML-based mortality-prediction model can predict long-term mortality accurately, leading to optimal risk stratification of patients with AHF.