Improving resident performance in knee arthroscopy: A prospective value assessment of simulators and cadaveric skills laboratories

Background: Cadaveric skills laboratories and virtual reality simulators are two common methods used outside of the operating room to improve residents' performance of knee arthroscopy. We are not aware of any head-to-head comparisons of the educational values of these two methodologies. The purpose of this prospective randomized trial was to assess the efficacy of these training methods, compare their rates of improvement, and provide economic value data to programs seeking to implement such technologies. Methods: Orthopaedic surgery residents were randomized to one of three groups: control, training on cadavera (cadaver group), and training with use of a simulator (simulator group). Residents completed pretest and posttest diagnostic knee arthroscopies on cadavera that were timed and video-recorded. Between the pretest and posttest, the control group performed no arthroscopy, the cadaver group performed four hours of practice on cadavera, and the simulator group trained for four hours on a simulator. All tests were scored in a blinded, randomized fashion using the validated Arthroscopy Surgical Skill Evaluation Tool (ASSET). The mean improvement in the ASSET score and in the time to complete the procedure were compared between the pretest and posttest and among the groups. Results: Forty-five residents (fifteen per group) completed the study. The mean difference in the ASSET score from the pretest to the posttest was 20.40 (p = 0.776) in the control group, 14.27 (p = 0.002) in the cadaver group, and 11.92 (p = 0.096) in the simulator group (p = 0.015 for the comparison among the groups). The mean difference in the testcompletion time (minutes:seconds) from the pretest to the posttest was 0:07 (p = 0.902) in the control group, 3:01 (p = 0.002) in the cadaver group, and 0:28 (p = 0.708) in the simulator group (p = 0.044 for the comparison among groups). Residents in the cadaver group improved their performance at a mean of 1.1 ASSET points per hour spent training whereas those in the simulator group improved 0.5 ASSET point per hour of training. Conclusions: Cadaveric skills laboratories improved residents' performance of knee arthroscopy compared with that of matched controls. Residents practicing on cadaveric specimens improved twice as fast as those utilizing a high-fidelity simulator; however, based on cost estimation specific to our institution, the simulator may be more cost-effective if it is used at least 300 hours per year. Additional study of this possibility is warranted.