Initial Diuretic Strategy During Hospitalization for Heart Failure and Associated Outcomes: Insights from the TREAT-AHF Registry.
BACKGROUND Real-world data regarding diuretic strategies and associated outcomes in patients hospitalized for heart failure (HF) in community health systems are limited. OBJECTIVES Evaluate associations between initial diuretic therapy, markers of decongestion, and clinical outcomes in patients hospitalized for HF. METHODS Patients hospitalized for HF from 2015-2022 across 30 health systems in the U.S. were identified in the Truveta national database. High dose loop diuretics were defined as >2.5X home dose or >160 mg IV furosemide equivalent over 24 hours. Four study groups were defined based on most intensive diuretic strategy used within 48 hours of admission: 1) high dose loop diuretics with adjuvant therapy (thiazide or acetazolamide), 2) high dose loop diuretics alone, 3) low dose loop diuretics with adjuvant therapy, and 4) low-dose loop diuretics alone. Multivariable logistic and linear regression models adjusted for clinical and demographic covariates were developed to evaluate associations between initial diuretic strategies and both in-hospital outcomes (acute kidney injury [AKI]), hemoconcentration, and weight change) and the composite of readmission or death. RESULTS Patients were treated with low dose loop diuretics (N=81,734; 74.9%), high dose loop diuretics (N=17,187; 15.7%), low dose loop diuretics plus adjuvant therapy (N=7,027; 6.4%), and high dose loop diuretics plus adjuvant therapy (N=3,210; 2.9%). Patients treated with more intensive strategies had greater illness severity, including more frequent prior HF hospitalizations and worse kidney function. Adjusted weight loss during hospitalization was greater for patients treated with more intensive strategies (high dose loop with adjuvant: 4.6 lbs [4.0-5.2]; high dose loop alone: 2.1 lbs [1.9-2.4]; low dose loop with adjuvant: 1.4 lbs [1.0-1.8]), as were adjusted odds of AKI. The adjusted odds of 90-day readmission/death were not lower with more intensive initial diuretic strategies. However, odds of 90-day readmission/death were lower for every 5 lbs of achieved weight loss (adjusted odds ratio: 0.97, 95% confidence interval 0.96-0.97). CONCLUSIONS More intensive diuretic strategies were used in sicker patients, improved in-hospital decongestion, but were not associated with improved 90-day outcomes. However, greater weight loss was associated with modestly lower risk of death or readmission. These data highlight the need for prospective studies to evaluate if enhancing decongestion can improve outcomes in patients hospitalized for HF.