Abstract: PO1803
Which Loop Is Best? Comparing the Effect of Loop Diuretic Prescribing on Mortality and Heart Failure Readmission
Session Information
- Hypertension and CVD: Clinical, Outcomes, and Trials
November 04, 2021 | Location: On-Demand, Virtual Only
Abstract Time: 10:00 AM - 12:00 PM
Category: Hypertension and CVD
- 1402 Hypertension and CVD: Clinical, Outcomes, and Trials
Authors
- Virkud, Arti Vikas, University of North Carolina at Chapel Hill Gillings School of Global Public Health, Chapel Hill, North Carolina, United States
- Kshirsagar, Abhijit V., University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, North Carolina, United States
- Chang, Patricia, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, North Carolina, United States
- Jonsson Funk, Michele, University of North Carolina at Chapel Hill Gillings School of Global Public Health, Chapel Hill, North Carolina, United States
- Edwards, Jessie K., University of North Carolina at Chapel Hill Gillings School of Global Public Health, Chapel Hill, North Carolina, United States
- Kosorok, Michael R., University of North Carolina at Chapel Hill Gillings School of Global Public Health, Chapel Hill, North Carolina, United States
- Gower, Emily, University of North Carolina at Chapel Hill Gillings School of Global Public Health, Chapel Hill, North Carolina, United States
Background
Loop diuretics are a mainstay of heart failure (HF) management. While furosemide is most commonly prescribed, torsemide and bumetanide are increasingly being prescribed, possibly due to their superior bioavailability. Few trials or real-world evidence studies have compared the effectiveness of prescribing these loop diuretics while adequately addressing critical study design biases.
Methods
We identified beneficiaries initiating the study loop diuretics by using an active comparator, new-user cohort design and Medicare claims data from 2007-2017. We estimated 1-year risks of death and a composite outcome (HF readmission/death) using inverse probability of treatment weighting to adjust for relevant confounders. We calculated a dose equivalency based on furosemide to adjust for disease severity.
Results
We identified 45,310 furosemide, 1,148 torsemide, and 1,630 bumetanide new users. In the total weighted population, 24.3% had a reduced ejection fraction, 27.1% had CKD (> Stage 2), with a mean age of 80.1 years and a mean furosemide dose equivalent of 50.8 mg/mL. The 1-year risk of death across all study loop diuretics was similar (19.9%-20.6%), whereas the risk of the composite outcome was more varied (29.1%-32.0%). The 1-year risk difference (95% CI) of the composite outcome was -2.9% (-6.2, 0.4) for torsemide vs. furosemide and -1.1% (-3.8, 1.6) for bumetanide vs. furosemide.
Conclusion
Among Medicare beneficiaries, the risk of HF readmission/death varies meaningfully with torsemide having a reduced risk compared to other study loop diuretics. This study leverages claims data and causal methodology to generate a less biased and more generalizable estimate than previous studies. While additional trial and real-world evidence studies are needed, this study suggests initial loop diuretic prescription after HF hospitalization may produce long-term differences in risk of death and HF readmission.
Funding
- NIDDK Support