Purpose

Immersive VR (iVR) simulators provide accessible, low cost, realistic training adjuncts in time and financially constrained residency programs. Over recent years, iVR has been applied in many surgical education contexts, including anatomy instruction, preoperative planning, intraoperative communication, and the topic of our review, surgical skills training. With increasing interest in this technology in research and surgical training programs, clarity on the effect of global skill training should be provided. This systematic review examines the current literature on the effectiveness of iVR for surgical skills acquisition in medical students, residents, and staff surgeons.

Methods

A literature search was performed on MEDLINE, EMBASE, CENTRAL, Web of Science and PsycInfo for primary studies published between January 1, 2000 and May 13, 2020 on the use of iVR to develop technical surgical skills. Two reviewers independently screened titles, abstracts, and full texts, extracted data, and assessed quality and strength of evidence using the Medical Education Research Quality Instrument (MERSQI) and the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework. Results were qualitatively synthesized, and descriptive statistics were calculated. Where possible, standardized mean differences (SMD) were calculated using a random effects model.

Results

The literature search yielded 8,939 citations, with 13 articles included for qualitative synthesis. The mean MERSQI score was 11.9±1.6 out of 18. In total, 242 participants, of whom 164 were residents, completed training in: orthopedic surgery (8/13), obstetrics and gynecology (2/13), non-specific surgical skills (2/13), and general surgery (1/13). Three articles described pre-post studies, while seven randomized (RCT) and two non-randomized controlled trials compared iVR with no simulator training, and one RCT compared iVR with conventional VR. Immersive Immersive VR-trained groups performed 18% to 43% faster on procedural time to completion compared to control (pooled SMD =-1.11 [95% CI=-1.66 to -0.57, I2 =1%, p<0.0001]). Immersive VR training also resulted in greater task completion, with the iVR group in one study scoring 14 points higher on a 30-point task-specific checklist for total hip arthroplasty (p<0.001). Accuracy was improved as well. In two RCTs on pedicle screw placement (PSP), iVR groups placed significantly more successful grade I/A screws compared to controls (89.6% vs. 60.4%, and 69.6% vs. 55.4%). Studies that assessed performance with global rating scales demonstrated mixed results. The strength of evidence for the outcomes ‘time to completion’ and ‘implant placement accuracy’ were rated ‘moderate’ on the GRADE framework, whereas the strength of ‘task-specific checklist’ and ‘global rating scale’ outcomes was rated ‘low.’

Conclusion

Evidence for incorporation of iVR into surgical training programs is supported by multiple high-quality studies demonstrating improved procedural times, task completion, and accuracy, positive user ratings, and cost-effectiveness. The current body of evidence is limited by scale and heterogeneity of outcome reporting. Thus, the rising enthusiasm for iVR should be paralleled by further high-quality studies to assess its long-term impact on learner and patient-important outcomes.