Background: Serum levels of pro-B-type natriuretic peptide (BNP) and N-terminal (NT) proBNP are measured at admission to assess likelihood of acutely decompensated heart failure (ADHF). Pro-BNP, released by myocardium due to cardiac wall distention, is cleaved into the active BNP and inactive NT-proBNP forms. Elevated NT-proBNP levels on initial presentation are a reliable marker of ADHF. However, the prognostic significance of NT-proBNP levels measured on admission remains unknown. With better understanding of how admitting NT-proBNP levels impacts readmission rates, length of stay, and mortality, further prospective studies with specific interventions can be developed to reduce all-cause readmissions, shorten length of stay, and reduce mortality.

Methods: In this retrospective study, we evaluated heart failure with reduced ejection fraction (HFrEF) admissions from 2017-2018 with a focus on 30, 60, and 90-day all-cause readmissions, length of stay (LOS), and in-hospital mortality rate, that are predicted by NT-proBNP levels measured on admission. Using the HCA Healthcare Enterprise Data Warehouse, adult patients age 18 to 75 were selected using admission ICD-10 codes for HFrEF. Dialysis patients were excluded. Our search of 90 hospitals yielded 21,445 patients who were stratified into quartiles depending on their admission NT-proBNP levels: group 1 (10,500 pg/ml).

Results: Readmission Rates:The 60-day all cause readmission was significantly (p=0.047) higher in group 4 (NT-proBNP >10,500) compared to group 1 (adjusted odds ratio (OR) = 1.116, p = 0.013) or group 2 (adjusted OR = 1.111, p = 0.014). The 90-day all cause readmission for group 4 was also significantly higher when compared to group 1 (adjusted OR = 1.105, p = 0.021).Length of Stay:Elevated NT-proBNP concentrations were associated with a significantly longer LOS (p <0.0005). Pairwise comparisons and estimates for adjusted LOS showed a positive linear association between higher NT-proBNP groups and longer LOS. Group 1 had a median LOS of 4.67 days, which progressively increased to 7.03 days for group 4. Mortality:Higher inpatient mortality rates were associated with elevated NT-proBNP levels. The mortality rate was 0.9% in group 1 compared to a 4.7% mortality rate in group 4. Adjusted OR for mortality increased with increasing levels of NT-proBNP. Most notably, group 4 had a 4.789 adjusted OR of mortality compared to group 1 (p <0.0005) and 3.013 adjusted OR (p <0.0005) compared to group 2.

Conclusions: Based on the analysis of n = 21,445 cases, admitting NT-proBNP levels were associated with significantly higher 60-day all-cause readmission, longer LOS, and increased mortality. These findings suggest that measuring NT-proBNP levels at admission may provide an indication of patient outcomes. Prospective studies with targeted strategies can be developed to reduce readmissions, shorten LOS, and reduced mortality based on admission NT-proBNP levels.