Most international acute heart failure trials have failed to show benefit with respect to key end points. The impact of site enrollment and protocol execution on trial performance is unclear.Methods and Results—
We assessed the impact of varying site enrollment volume among all 7141 acute heart failure patients from the ASCEND-HF trial (Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure). Overall, 398 sites enrolled ≥1 patient, and median enrollment was 12 patients (interquartile range, 5–23). Patients from high enrolling sites (>60 patients/site) tended to have lower ejection fraction, worse New York Heart Association functional class, and lower utilization of guideline-directed medical therapy but fewer comorbidities and lower B-type natriuretic peptide level. Every 10 patient increase (up to 100 patients) in site enrollment correlated with lower likelihood of protocol noncompletion (odds ratio, 0.93; 95% confidence interval [CI], 0.89–0.98). After adjustment, increasing site enrollment predicted higher risk of persistent dyspnea at 6 hours (per 10 patient increase: odds ratio 1.02; 95% CI, 1.01–1.03) but not at 24 hours (odds ratio, 0.99; 95% CI, 0.98–1.00). Higher site enrollment was independently associated with lower risk of 30-day death or rehospitalization (per 10 patient increase: odds ratio, 0.98, 95% CI, 0.96–0.99) but not 180-day mortality (hazard ratio, 0.99; 95% CI, 0.98–1.01). The influence of increasing site enrollment on clinical end points varied across geographic regions with strongest associations in Latin America and Asia-Pacific (all interaction P<0.01).Conclusions—
In this large, acute heart failure trial, site enrollment correlated with protocol completion and was independently associated with trial end points. Individual and regional site performance present challenges to be considered in design of future acute heart failure trials.Clinical Trial Registration—
URL: http://www.clinicaltrials.gov. Unique identifier: NCT00475852.