Outcomes With Ultrafiltration Among Hospitalized Patients With Acute Heart Failure (from the National Inpatient Sample)

Ali Yazdanyar, Department of Emergency and Hospital Medicine, Lehigh Valley Hospital-Cedar Crest, Allentown, Pennsylvania; Morsani College of Medicine, University of South Florida, Tampa, Florida. Electronic address: a.yazzzd@gmail.com.
Julien Sanon, Department of Emergency and Hospital Medicine, Lehigh Valley Hospital-Cedar Crest, Allentown, Pennsylvania.
Kevin Bryan Lo, Department of Medicine, Einstein Medical Center, Philadelphia, Pennsylvania.
Amogh M. Joshi, Department of Medicine, Lehigh Valley Health Network, Allentown, Pennsylvania.
Emilee Kurtz, Department of Medicine, Lehigh Valley Health Network, Allentown, Pennsylvania.
Mohammed Najum Saqib, Division of Nephrology, Department of Medicine, Lehigh Valley Health Network, Allentown, Pennsylvania.
Nauman Islam, Department of Medicine/Cardiology, Lehigh Valley Health Network, Allentown, Pennsylvania.
Mahek K. Shah, Sidney Kimmel College of Medicine/Cardiology, Thomas Jefferson University, Philadelphia, Pennsylvania; Sidney Kimmel College of Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania.
Adam Feldman, Department of Medicine/Cardiology, Tower Health/Reading Hospital, Reading, Pennsylvania.
Anthony Donato, Sidney Kimmel College of Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania; Department of Medicine, Tower Health/Reading Hospital, Reading, Pennsylvania.
Janani Rangaswami, Sidney Kimmel College of Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania; Department of Medicine/Nephrology, Einstein Medical Center, Philadelphia, Pennsylvania.

Abstract

Acute heart failure (HF) management is a complex and often involves a delicate balance of both cardiac and renal systems. Although pharmacologic diuresis is a mainstay of the pharmacologic management of decompensated HF, ultrafiltration (UF) represents a nonpharmacologic approach in the setting of diuretic resistance. We conducted a cross-sectional analysis of the 2009 through 2014 hospitalization data from the National Inpatient Sample. The study population consisted of hospitalizations with a discharge Diagnosis Related Groups of HF who were older than 18 years of age, did not have end-stage kidney disease, acute kidney injury and had not undergone hemodialysis or hemofiltration. There were 6,174 hospitalizations which included UF among the 7,799,915 hospitalizations for HF. Hospitalizations which included UF were among patients significantly younger in age (68.1 ± 1.0 vs 73.8 ± 0.1 years), male (61.9% vs 47.7%), and with higher prevalence of co-morbid conditions including chronic kidney disease (58% vs 31%), diabetes mellitus (53% vs 42%), and higher rates of co-morbidity (Charlson comorbidity score ≥2, 92% vs 80%). All-cause mortality was significantly higher among hospitalizations which included an UF (4.68% vs 2.24%). Hospitalizations with UF had a longer mean length of stay (6.2 vs 4.3 days, p <0.01) average total charges ($42,035 vs 24,867 USD, p <0.01) as compared with those without UF. Hospitalizations with UF were associated with a greater adjusted odds of all-cause mortality (odds ratio: 3.36, [95% confidence interval 1.76,6.40]), greater than DRG-level target length of stay (odds ratio, 2.46; [95 confidence interval 1.65,3.67]), and a 72% increase in the average hospital charges. In conclusion, hospitalizations which included UF identified a subgroup of HF patients with more co-morbid conditions who are at higher risk of mortality and increased resource burden in terms of length of stay and costs. These findings also highlight that the need for UF may identify patients who are most likely to benefit from a multidisciplinary cardiorenal approach to alter the trajectory of their disease.