Arthroscopic Shoulder Simulation Studies Reveal Improvements in Performance Metrics Without Proven Transferability to the Operating Room: A Systematic Review

• Published in: Arthroscopy
• Main Authors: DeClercq, Madeleine Grace, Pfennig, Mitchell T., Gannon, James, Oshikoya, Olamide, Perry, Bradley, Dunne, Kevin F., Wiater, J. Michael
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 02-09-2024
• Subjects: arthroscopic; orthopaedic education; Shoulder; simulation
• ISSN: 0749-8063; 1526-3231; 1526-3231
• DOI: 10.1016/j.arthro.2024.08.020

To evaluate the use of shoulder arthroscopic simulation in orthopaedic surgery trainees.
A literature search was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, using PubMed, Medline (Ovid), and EMBASE library databases. Inclusion criteria were experimental studies reporting pre- and post-test results of shoulder arthroscopic simulation in orthopaedic trainees (studies reporting results of comparison between groups not within the groups were excluded). Participant demographics, type of simulator training, simulator tasks assessed, and performance outcome measures were systematically reviewed. Each performance outcome measure was graphically represented in a Forest plot with point estimates of the incidence of performance outcomes with corresponding 95% confidence intervals and I2.
Fifteen studies met inclusion criteria with a total of 353 participants. The most common procedures simulated were diagnostic shoulder arthroscopy (n = 9 [60%]), arthroscopic Bankart repairs (n = 3 [20%]), and rotator cuff repairs (n = 2 [13%]). Simulations primarily used virtual reality (60%) and benchtop models (40%). The primary outcomes measured were time to task completion and Arthroscopic Surgical Skill Evaluation Tool scores. Time to task completion improved significantly with training (range 13-439 seconds pretest to 8-253.29 seconds post-test), with substantial heterogeneity across studies (I2 = 87%). ASSET scores improved in 60% of the studies (ranging from 14-20.9 pretest to 17.9-28.5 post-test), with low heterogeneity (I2 = 20%). In addition, both camera and probe distances decreased after simulation use, whereas the 14-point anatomic checklist showed no pre- to post-test differences.
Arthroscopic simulation training benefits technical skills in shoulder arthroscopy, but the quality, assessment, and validity of these protocols vary. The translation of simulation training into the operating room has yet to be conclusively demonstrated.
Level IV, systematic review of Level I-IV studies.