Robotic Versus Electromagnetic Navigational Bronchoscopy for Pulmonary Lesion Assessment. Results from the RELIANT Randomized Trial

摘要

Abstract Rationale: Robotic assisted bronchoscopy has emerged as an alternative to electromagnetic navigational bronchoscopy for patients undergoing bronchoscopic biopsy of a peripheral pulmonary lesion. Observational data suggest comparable diagnostic yields. However, randomized controlled trials are lacking, in part, because conventional study designs are poorly suited to assess new devices within a rapidly evolving technological landscape. Methods In a pragmatic, investigator-initiated, single-center, cluster randomized noninferiority multiple-crossover trial, patients scheduled to undergo advanced bronchoscopy were randomly assigned to robotic assisted or electromagnetic navigational bronchoscopy. The trial was conducted in two operating rooms at Vanderbilt University Medical Center. Each OR was randomly assigned each day to either robotic assisted or electromagnetic navigational bronchoscopy, with each OR day representing one cluster. Allocation was concealed until randomization. All patients provided written informed consent. The trial was approved by the Institutional Review Board (221255), registered at ClinicalTrials.gov (NCT 05705544) before initiation, and was overseen by an independent data and safety monitoring board. The primary outcome was diagnostic yield defined as the proportion of cases yielding lesional tissue per consensus statement from the American Thoracic Society and the American College of Chest Physicians. Secondary and safety endpoints included procedure duration and complications. The primary outcome was analyzed following a modified intention to treat approach using a generalized linear mixed model that was covariate adjusted. Analysis of the secondary and safety outcome followed a similar approach. Results Between March 2023 and April 2024, 411 patients were enrolled and analyzed, 203 in the robotic assisted group and 208 in the electromagnetic navigational bronchoscopy group. The groups were balanced with regard to patient and lesion characteristics. Lesional tissue was obtained in 158 of 203 cases in the robotic assisted group (77.8%), and 157 of 208 cases in the electromagnetic navigational group (75.5%) with a p-value for non-inferiority = 0.007. The median duration of the procedure was 37 minutes in the robotic assisted group and 32 minutes in the electromagnetic group (difference, 5 minutes; 95% confidence interval 2.0 to 7.7). The most common complication was pneumothorax, which occurred in 4 patients in the robotic assisted group and 6 in the electromagnetic navigational group. Conclusions Among patients undergoing bronchoscopy to biopsy a peripheral pulmonary lesion, the diagnostic yield of robotic assisted bronchoscopy was non-inferior to that of electromagnetic navigation bronchoscopy. As the first pragmatic, cluster randomized trial, RELIANT serves as a paradigm for future trials evaluating devices within a rapidly evolving technological landscape.