Abstract
Aims
Our aim was to assess the dynamic changes of pulmonary congestion (PC) through variations of sonographic B-lines, in addition to conventional clinical, biohumoral and echocardiographic findings, to improve prognostic stratification of patients admitted for acute heart failure with reduced and preserved ejection fraction (HFrEF, HFpEF).
Methods
In this multicenter, prospective, observational study, lung ultrasound was performed in all patients at admission and before discharge by trained investigators, blinded to clinical findings and outcomes.
Results
We enrolled 208 consecutive patients admitted for acute heart failure (125 HFrEF, 83 HFpEF, mean age 75.9±11.7 years, 36% females, mean ejection fraction 38%). After 180-day follow-up, 38 composite endpoint events occurred (cardiovascular deaths or HF re-hospitalisations). In a multivariate model, B-lines at discharge had independent prognostic value in the overall population together with NT-proBNP, moderate-to-severe mitral regurgitation (MR) and inferior vena cava diameter at admission. When dividing the population in HFrEF and HFpEF, B-lines at discharge was the only independent parameter to predict events in all subgroups. At ROC analysis, a cut-off of B-lines>15 at discharge displayed the highest accuracy in predicting adverse events (AUC=0.80, p<0.0001). The identification of patients unable to halve B-lines during hospitalization (ΔB-lines%), in addition to B-lines >15 at discharge, improved event classification (integrated discrimination improvement=4%, p=0.01; continuous net reclassification improvement=22.8%, p=0.04).
Conclusions
The presence of residual subclinical sonographic PC at discharge predicts adverse events in the whole spectrum of acute HF patients, independently of conventional biohumoral and echocardiographic parameters. The dynamic evaluation of pulmonary decongestion during hospital stay can further improve patient risk stratification.
Funding Acknowledgement
Type of funding source: None
A HIGH-PERFORMANCE MACHINE LEARNING MODEL TO PREDICT 90-DAY ACUTE HEART FAILURE READMISSION AND DEATH IN HEART FAILURE WITH PRESERVED EJECTION FRACTION
