Improved Management of Patients With Congestive Heart Failure - IMPROVE-CHF
The goal of the study was to evaluate the effect on clinical management and costs using N-terminal proB-type natriuretic peptide (NT-proBNP) evaluation among patients presenting to the emergency department (ED) with dyspnea and suspected acute decompensated heart failure (ADHF).
Patients Enrolled: 500
Mean Follow Up: 60 days
Mean Patient Age: Mean age, 71 years
Presenting to the ED with dyspnea and suspected acute heart failure
Age <18 years, overt cause of dyspnea, serum creatinine >250 µmol/L, or acute MI
Health care costs through 60 days for NT-proBNP guided care versus usual care
NT-proBNP samples were drawn in all patients included in the trial, at which time the ED physician provided a clinical diagnosis. Patients were randomized to usual care without revealing the NT-proBNP levels (n = 254) or physician knowledge of NT-proBNP results (n = 246) for use with clinical management. Patients who were admitted to the hospital also had NT-proBNP evaluated at 72 hours (n = 285). The trial was conducted at multiple Canadian hospitals.
Dyspnea at rest was present in 55% of patients. Approximately one-third of patients had a history of heart failure or left ventricular dysfunction (37%), 16% had prior cerebrovascular events, 31% had chronic obstructive pulmonary disease, 32% prior myocardial infarction (MI), and 27% diabetes. Final diagnosis was heart failure in 230 patients (46%).
Addition of NT-proBNP to clinical evaluation was associated with higher discriminatory ability compared with clinical evaluation alone for the ED diagnosis of heart failure (C-statistic 0.90 vs. 0.83, p < 0.001). Use of NT-proBNP alone had a C-statistic of 0.86. Median NT-proBNP was significantly higher in those with a final diagnosis of ADHF compared to without a diagnosis of ADHF (3697 pg/ml vs. 212 pg/ml, p < 0.001). Duration of ED visit was shorter in those patients randomized to NT-proBNP guided care (median 5.6 hours vs. 6.3 hours, p = 0.03). There was no difference in initial hospitalization (57% vs. 58%), median length of stay (6 days vs. 7 days), or intensive care unit admission (8.5% vs. 9.9%) for NT-proBNP and usual care, respectively. There was also no difference in 60-day mortality (5.5% vs. 4.4%, p = 0.58).
Total median health care costs through 60 days were significantly lower among those randomized to NT-proBNP guided care compared with usual care ($5180 vs. $6129 US dollars, p = 0.02). Median cost for the initial ED visit, including the cost of the NT-proBNP test, was $1813 for NT-proBNP and $1982 for usual care (p = 0.10).
Among patients with dyspnea and suspected acute heart failure presenting to the ED, use of NT-proBNP to guide diagnosis and care was associated with lower costs through 60 days compared with usual care without use of NT-proBNP.
In addition to a reduction in costs, use of NT-proBNP was associated with a shorter length of time in the ED, although no difference in hospital admission or length of stay. Addition of NT-proBNP to clinical evaluation was also associated with higher discriminatory ability compared with clinical evaluation alone. Results are similar to those reported in the BASEL trial, which demonstrated reductions in total costs with use of a BNP assay to diagnose heart failure among patients presenting to the ED with acute dyspnea.
Moe GW, Howlett J, Januzzi JL, Zowall H. N-terminal pro-B-type natriuretic peptide testing improves the management of patients with suspected acute heart failure: primary results of the Canadian prospective randomized multicenter IMPROVE-CHF study. Circulation 2007;115:3103-10.
Presented by Dr. Gordon Moe at the American Heart Association Annual Scientific Sessions, Chicago, IL, November 2006.
Keywords: Pulmonary Disease, Chronic Obstructive, Myocardial Infarction, Intensive Care Units, Heart Failure, Peptide Fragments, Health Care Costs, Emergency Service, Hospital, Dyspnea, Ventricular Dysfunction, Left, Diabetes Mellitus, Length of Stay, Natriuretic Peptide, Brain
< Back to Listings