Background:

The most effective strategy for administration of loop diuretics in patients hospitalized with acute decompensated heart failure (ADHF) still needs to be determined. Delivery of optimal dosing for rapid and safe diuresis must include an understanding of the pharmacokinetics of loop diuretics and the complexities of health care delivery. This study aimed to evaluate the effects of a nurse–driven protocol for the administration of loop diuretics to patients admitted with ADHF.

Methods:

We performed a retrospective cohort study comparing the use of a diuretic protocol to usual diuretic dosing for patients hospitalized with ADHF at a large academic cardiac transplant center in Chicago from September 2010 to August 2011. An interdisciplinary team developed the diuretic protocol to allow quick titration by nursing staff of furosemide or bumetanide (bolus or continuous infusion) with a goal of 100–250 cc of urine output per hour. The protocol required frequent closed loop communication between nurses and physicians with aggressive electrolyte monitoring. After extensive physician and nurse training using simulation, we implemented the protocol in August 2010. The patient’s admitting physician determined use of the protocol. Using an electronic data warehouse, we searched medical records to identify all patients admitted to a telemetry floor with the primary diagnosis of heart failure who received two doses of intravenous loop diuretic in a 24–h period. We compared patients who received the diuretic protocol versus those who received standard diuretic therapy. Comparisons were made between groups for total weight loss, length of stay (LOS), 30–day readmissions, in–hospital mortality, and changes in creatinine. We adjusted data for covariates including systolic blood pressure, total days receiving diuretics, Charlson Score (severity of illness), use of inotropes, ICU transfer, LOS, and levels of sodium, creatinine and brain natriuretic peptide.

Results:

There were significant differences in total weight loss (mean 7.7 kg, SD = 10.4 vs 2.7 kg, SD = 6.0; p < 0.001) and LOS (mean 12.5 days, SD = 26.0 vs 7.2 days, SD = 7.3; p < 0.001) between diuretic protocol (n = 74) and usual care groups (n = 580), respectively. There were no between group differences for 30–day readmission, mortality, or changes in creatinine. After adjusting for covariates, patients receiving the diuretic protocol lost 3.7 kg more than patients receiving usual care (p < 0.001), while differences in LOS, 30–day readmission, mortality, and change in creatinine were non–significant between groups.

Conclusions:

The use of a diuretic protocol is associated with increased total weight loss compared to usual care in patients admitted with ADHF and appears to be as safe as usual care. Further study is required to investigate these relationships.