Determinants of Diuretic Responsiveness and Associated Outcomes During Acute Heart Failure Hospitalization: An Analysis From the NHLBI Heart Failure Network Clinical Trials

Abstract

Background: Poor response to loop diuretic therapy is a marker of risk during heart failure hospitalization. We sought to describe baseline determinants of diuretic response and to further explore the relationship between this response and clinical outcomes.

Methods and results: Patient data from the National Heart, Lung, and Blood Institute Heart Failure Network ROSE-AHF and CARRESS-HF clinical trials were analyzed to determine baseline determinants of diuretic response. Diuretic efficiency (DE) was defined as total 72-hour fluid output per total equivalent loop diuretic dose. Data from DOSE-AHF was then used to determine if these predictors of DE correlated with response to a high- versus low-dose diuretic strategy. At 72 hours, the high-DE group had median fluid output of 9071 ml (interquartile range: 7240-11775) with median furosemide dose of 320 mg (220-480) compared with 8030 ml (6300-9915) and 840 mg (600-1215) respectively for the low DE group. Cystatin C was independently associated with DE (odds ratio 0.36 per 1mg/L increase; 95% confidence interval: 0.24-0.56; P < 0.001). Independently from baseline characteristics, reduced fluid output, weight loss and DE were each associated with increased 60 day mortality. Among patients with estimated glomerular filtration rate below the median, those randomized to a high-dose strategy had improved symptoms compared with those randomized to a low-dose strategy.

Conclusions: Elevated baseline cystatin C, as a biomarker of renal dysfunction, is associated with reduced diuretic response during heart failure hospitalization. Higher loop diuretic doses are required for therapeutic decongestion in patients with renal insufficiency. Poor response identifies a high-risk population.

Keywords: Acute heart failure; congestion; cystatin C; loop diuretic; renal failure.

Figure 1

Consort diagram detailing the pooled study population from the ROSE-AHF and CARRESS-HF cohorts used for the predictors of diuretic efficiency and associated outcomes analyses.

Figure 2

a Scatter plot of net IV furosemide dose equivalent by fluid output in the first 72 hours. b Scatter plot of net IV furosemide dose equivalent by weight change in the first 72 hours.

Figure 2

a Scatter plot of net IV furosemide dose equivalent by fluid output in the first 72 hours. b Scatter plot of net IV furosemide dose equivalent by weight change in the first 72 hours.

Figure 3

(Central Figure) Kaplan-Meier curves for survival stratified by median diuretic efficiency. The KM mortality rate at 60 days is 13.1% in the low DE group and 7.0 % in the high DE group.