Background: In the airline industry, training is costly and operator error must be avoided. Therefore, virtual reality (VR) is routinely used to learn manual and technical skills through simulation before pilots assume flight responsibilities. In the field of medicine, manual and technical skills must also be acquired to competently perform invasive procedures such as flexible fiberoptic bronchoscopy (FFB). Until recently, training in FFB and other endoscopic procedures has occurred on the job in real patients. We hypothesized that novice trainees using a VR skill center could rapidly acquire basic skills, and that results would compare favorably with those of senior trainees trained in the conventional manner.
Methods: We prospectively studied five novice bronchoscopists entering a pulmonary and critical care medicine training program. They were taught to perform inspection flexible bronchoscopy using a VR bronchoscopy skill center; dexterity, speed, and accuracy were tested using the skill center and an inanimate airway model before and after 4 h of group instruction and 4 h of individual unsupervised practice. Results were compared to those of a control group of four skilled physicians who had performed at least 200 bronchoscopies during 2 years of training. Student's t tests were used to compare mean scores of study and control groups for the inanimate model and VR bronchoscopy simulator. Before-training and after-training test scores were compared using paired t tests. For comparisons between after-training novice and skilled physician scores, unpaired two-sample t tests were used.
Results: Novices significantly improved their dexterity and accuracy in both models. They missed fewer segments after training than before training, and had fewer contacts with the bronchial wall. There was no statistically significant improvement in speed or total time spent not visualizing airway anatomy. After training, novice performance equaled or surpassed that of the skilled physicians. Novices performed more thorough examinations and missed significantly fewer segments in both the inanimate and virtual simulation models.
Conclusion: A short, focused course of instruction and unsupervised practice using a virtual bronchoscopy simulator enabled novice trainees to attain a level of manual and technical skill at performing diagnostic bronchoscopic inspection similar to those of colleagues with several years of experience. These skills were readily reproducible in a conventional inanimate airway-training model, suggesting they would also be translatable to direct patient care.