Abstract
BACKGROUND: Rather than the absolute dose of diuretic or urine output, the primary signal of interest when evaluating diuretic responsiveness is the efficiency with which the kidneys can produce urine after a given dose of diuretic. As a result, we hypothesized that a metric of diuretic efficiency (DE) would capture distinct prognostic information beyond that of raw fluid output or diuretic dose. METHODS AND RESULTS: We independently analyzed 2 cohorts: (1) consecutive admissions at the University of Pennsylvania (Penn) with a primary discharge diagnosis of heart failure (n=657) and (2) patients in the Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE) data set (n=390). DE was estimated as the net fluid output produced per 40 mg of furosemide equivalents, then dichotomized into high versus low DE based on the median value. There was only a moderate correlation between DE and both intravenous diuretic dose and net fluid output (r(2)≤0.26 for all comparisons), indicating that DE was describing unique information. With the exception of metrics of renal function and preadmission diuretic therapy, traditional baseline characteristics, including right heart catheterization variables, were not consistently associated with DE. Low DE was associated with worsened survival even after adjusting for in-hospital diuretic dose, fluid output, in addition to baseline characteristics (Penn: hazards ratio [HR], 1.36; 95% confidence interval [CI], 1.04-1.78; P=0.02; ESCAPE: HR, 2.86; 95% CI, 1.53-5.36; P=0.001). CONCLUSIONS: Although in need of validation in less-selected populations, low DE during decongestive therapy portends poorer long-term outcomes above and beyond traditional prognostic factors in patients hospitalized with decompensated heart failure.