Markers of Decongestion, Dyspnea Relief and Clinical Outcomes Among Patients Hospitalized With Acute Heart Failure

Study Questions:

Do measures of decongestion in heart failure (HF) correlate with clinical outcomes?


This was a retrospective analysis of the DOSE-AHF (Diuretic Optimization Strategy Evaluation in Acute Heart Failure ) trial. The relationship between HF symptoms (dyspnea assessed via visual analog scale [VAS]) with weight loss, net fluid loss, and percent reduction in N-terminal pro–B-type natriuretic peptide (NT-proBNP) levels at 72 hours was examined in patients (n = 308) hospitalized with acute HF. A composite endpoint of death, rehospitalization, or emergency room visit was also examined at 60 days and is presented as hazard ratios [95% confidence intervals] below.


Median [25th, 75th percentile] patient age was 68 [56, 77] years and mean left ventricular ejection fraction (LVEF) was 35% (27% had an LVEF ≥50%). The median fluid and weight loss at 72 hours was 3.8 liters and 6.5 pounds, respectively, and the median reduction in NT-proBNP was 24%. After 72 hours, dyspnea (assessed via VAS) was modestly correlated with change in NT-proBNP (r = 0.13, p = 0.04), but not net fluid loss (r = 0.07, p = 0.27) or weight loss (r = 0.04, p = 0.54). The adjusted risk of the 60-day composite endpoint was 0.93 [0.88, 1.00] per 1000 ml of fluid loss, 0.92 [0.85, 0.98] per 4 pounds of weight loss, and 0.97 [0.93, 1.02] per 10% reduction in NT-proBNP. There was no association between dyspnea assessment at 72 hours and clinical outcome at 60 days. In those with 0, 1, 2, or 3 of the above markers of decongestion present, survival free of death, HF hospitalization, or emergency department visit was 67%, 64%, 46%, and 38%, respectively (p = 0.05).


While weight loss, fluid loss, and NT-proBNP poorly correlate with dyspnea at 72 hours, each are associated with improved clinical outcomes at 60 days.


A cornerstone of HF management is symptom and volume status assessment. In this small retrospective study, weak correlations were demonstrated between volume assessment, natriuretic peptide changes, and patient symptoms and clinical outcome. While borderline significant p values were noted for the composite endpoints at 60 days, the magnitude of risk reduction despite ample fluid/weight and BNP reductions was small. However, in the era of accountable care, even small reductions in hospitalization may matter.

Clinical Topics: Heart Failure and Cardiomyopathies, Acute Heart Failure, Heart Failure and Cardiac Biomarkers

Keywords: Natriuretic Peptides, Emergency Medical Services, Biological Markers, Body Fluids, Weight Loss, Heart Failure, Dyspnea, Hospitalization

< Back to Listings