Randomized Clinical Trials in Cerebrovascular Neurosurgery: Analysis of Discontinuation and Nonpublication Prevalence and Trends.

Abstract

Premature randomized clinical trial (RCT) discontinuation and nonpublication have been shown to frequently compromise advancements in medicine. However, the extent to which these problems exist in RCTs of cerebrovascular disease is not well understood. This study aims to investigate the prevalence of discontinuation and nonpublication among cerebrovascular clinical trials and explore characteristics associated with each. We searched Clinicaltrials.gov using selected keywords to identify relevant RCTs. Trials generated in our search were screened to exclude unrelated trials, active trials, and trials completed within the previous three years. We analyzed included RCTs to identify the publication status or reason for discontinuation. Of the 421 included RCTs, 245 (58.2%) were completed and 176 (41.8%) were discontinued. Trials investigating medical devices and government-funded trials had the lowest rates of discontinuation. Participant accrual difficulty was the most common reason for discontinuation identified (12.5%), though 52.8% of the discontinued trials did not provide a reason for discontinuation. There were 226 published trials (53.7%) and 195 unpublished trials (46.3%). Unpublished trials included 135 of the 176 discontinued trials (76.7%) and 60 of 245 completed trials (24.5%). Trial discontinuation was significantly associated with nonpublication. Trials funded by the National Institutes of Health were less likely to be unpublished. Premature discontinuation was common among cerebrovascular trials in neurosurgery and frequently attributed to difficulties with patient enrollment. Nonpublication was especially common among discontinued trials. Our findings may highlight opportunities to improve research efficiency and potentially improve outcomes for patients with cerebrovascular diseases.