Computer-simulation training for knee and shoulder arthroscopic surgery.

PURPOSE: The aim of this systematic review was to identify the evidence as to whether computer simulators are able to teach and assess arthroscopic skills in a valid and reliable manner and whether these skills can be transferred to the operating theater.
METHODS: Multiple databases were searched by use of strict eligibility criteria to obtain the best evidence available. The studies were critically appraised by independent reviewers with narrative data synthesis.
RESULTS: The systematic literature search found 83 studies, of which 9 were included within this review. All of the studies were performed to validate the use of computer simulation, with only 1 study showing the effect on surgical performance within the operating theater. Five studies investigated knee arthroscopy and four investigated shoulder arthroscopy, with all showing high levels of internal validity and consistency for computer simulation. Outcome measures that appear to be able to discriminate skill level and experience include time to completion for tasks, the distance and path traveled by the probe, and the number of probe collisions.
CONCLUSIONS: Knee and shoulder arthroscopy computer simulators with force feedback, haptic technology, and computer-generated outcome data showed high levels of internal consistency and reliability. The studies also showed improvement of skill levels for inexperienced participants. The evidence suggests that knee arthroscopy simulator training may result in improved performance within the operating theater. Further studies are required to show the transfer and predictive validity of computer simulation within the operating theater to enable it to become established as a valid training tool. LEVEL OF EVIDENCE: Level III, systematic review. Copyright (c) 2010 Arthroscopy Association of North America. Published by Elsevier Inc. All rights reserved.