Discontinuation and non-publication of atrial fibrillation clinical trials: A retrospective analysis of 538 trials.

Background: Atrial fibrillation (AF) clinical trials are essential for informing evidence-based care, yet many trials are discontinued or remain unpublished. This contributes to research inefficiency and reduced transparency. This study evaluated characteristics associated with discontinuation and non-publication of AF clinical trials registered on ClinicalTrials.gov. Methods: We performed a retrospective analysis of interventional AF trials registered between 1999 and 2025 on ClinicalTrials.gov. Trials were categorized as completed or discontinued, and as published or unpublished. Trial characteristics were extracted, and multivariable logistic regression was used to identify predictors of discontinuation and non-publication. Results: Among 538 AF trials, 423 trials (78.6 percent) were completed and 115 trials (21.4 percent) were discontinued. Only 153 trials (36.2 percent) were published. The estimated median time from primary completion to publication was 18.0 months (95% confidence interval, 14.0-21.0 months). Trials enrolling 100 participants or more had significantly lower odds of both discontinuation and non-publication. Multicenter trials showed higher odds of discontinuation but lower odds of non-publication. The other characteristics studied were not significant predictors. Conclusions: A substantial proportion of AF trials are discontinued or remain unpublished, revealing ongoing gaps in research transparency. Larger sample sizes support higher completion and publication rates, while a multicenter design increases discontinuation risk but improves dissemination. These findings underscore the need for strategies that strengthen trial feasibility and reporting practices.