Simulation-based training for thoracoscopic lobectomy: a randomized controlled trial

Abstract

Video-assisted thoracic surgery is gradually replacing conventional open thoracotomy as the method of choice for the treatment of early-stage non-small cell lung cancers, and thoracic surgical trainees must learn and master this technique. Simulation-based training could help trainees overcome the first part of the learning curve, but no virtual-reality simulators for thoracoscopy are commercially available. This study aimed to investigate whether training on a laparoscopic simulator enables trainees to perform a thoracoscopic lobectomy. Twenty-eight surgical residents were randomized to either virtual-reality training on a nephrectomy module or traditional black-box simulator training. After a retention period they performed a thoracoscopic lobectomy on a porcine model and their performance was scored using a previously validated assessment tool. The groups did not differ in age or gender. All participants were able to complete the lobectomy. The performance of the black-box group was significantly faster during the test scenario than the virtual-reality group: 26.6min (SD 6.7min) versus 32.7min (SD 7.5min). No difference existed between the two groups when comparing bleeding and anatomical and non-anatomical errors. Simulation-based training and targeted instructions enabled the trainees to perform a simulated thoracoscopic lobectomy. Traditional black-box training was more effective than virtual-reality laparoscopy training. Thus, a dedicated simulator for thoracoscopy should be available before establishing systematic virtual-reality training programs for trainees in thoracic surgery.