Abstract
BACKGROUND: Shoulder arthroscopy is associated with a steep learning curve. Virtual reality (VR) arthroscopic simulator training has recently gained prominence as a promising training modality for shoulder arthroscopy. However, there is a lack of high-quality evidence regarding its efficacy. Thus, in this study we perform a systematic review to investigate the effects of VR simulator in shoulder arthroscopy training. METHODS: We performed a systematic search of four databases (PubMed, Embase, Scopus, and Cochrane Library). Studies that compared outcomes pre- and post-VR training as well as outcomes between VR groups and control (non-VR/low fidelity) groups were included. Primary outcomes between VR and control consisted of time to completion and arthroscopic evaluation score. Additionally, pre- versus post-training improvement in arthroscopic evaluation score was evaluated. RESULTS: Seven studies were included in this systematic review. Total cohort size was 143, with 93 in the VR group and 50 in the control group. Studies that compared time to completion in VR and control groups all showed significantly shorter time to completion in the VR group, with individual study standardized mean difference (SMD) ranging from -0.24 to -1.37. In terms of arthroscopic evaluation score, several studies showed pre- to post-training improvement in (individual SMD range, 0.32 to 2.32) but no difference between VR and control groups (individual SMD range, -0.02 to 0.82). CONCLUSIONS: In this study we found that VR training results in shorter task completion time for trainees, whereas there are no differences in arthroscopic evaluation scores between VR versus control groups. Hence, VR simulator training improves arthroscopic efficiency for orthopedic trainees, with limited improvement in quantitative skills. Level of evidence: II.