Background: Randomized controlled trials (RCT) are limited by their generalizability to the broader nontrial population. To provide a context for Acute Study of Nesiritide in Decompensated Heart Failure (ASCEND-HF) trial, we designed a complementary registry to characterize clinical characteristics, practice patterns, and in-hospital outcomes of acute heart failure patients.
Methods and results: Eligible patients for the registry included those with a principal diagnosis of acute heart failure (ICD-9-CM 402 and 428; ICD-10 I50.x, I11.0, I13.0, I13.2) from 8 sites participating in ASCEND-HF (n=697 patients, 2007-2010). Baseline characteristics, treatments, and hospital outcomes from the registy were compared with ASCEND-HF RCT patients from 31 Canadian sites (n=465, 2007-2010). Patients in the registry were older, more likely to be female, and have chronic respiratory disease, less likely to have diabetes mellitus: they had a similar incidence of ischemic HF, atrial fibrillation, and similar B-type natriuretic peptide levels. Registry patients had higher systolic blood pressure (registry: median 132 mm Hg [interquartile range 115-151 mm Hg]; RCT: median 120 mm Hg [interquartile range 110-135 mm Hg]) and ejection fraction (registry: median 40% [interquartile range 27-58%]; RCT: median 29% [interquartile range 20-40 mm Hg]) than RCT patients. Registry patients presented more often via ambulance and had a similar total length of stay as RCT patients. In-hospital mortality was significantly higher in the registry compared with the RCT patients (9.3% versus 1.3%,P<0.001), and this remained after multivariable adjustment (odds ratio 6.6, 95% CI 2.6-16.8, P<0.001).
Conclusions: Patients enrolled in a large RCT of acute heart failure differed significantly based on clinical characteristics, treatments, and inpatient outcomes from contemporaneous patients participating in a registry. These results highlight the need for context of RCTs to evaluate generalizability of results and especially the need to improve clinical outcomes in acute heart failure.
Clinical trial registration: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00475852.