BackgroundCataract surgery requires a high level of dexterity and experience to avoid serious intra- and post-operative complications. Proper surgical training and evaluation during the learning phase are crucial to promote safety in the operating room (OR). This scoping review aims to report cataract surgery training efficacy for patient safety and trainee satisfaction in the OR when using virtual reality simulators (EyeSi [Haag-Streit, Heidelberg, Germany] or HelpMeSee [HelpMeSee foundation, Jersey city, New Jersey, United States]) or supervised surgical training on actual patients programs in residents.MethodsAn online article search in the PubMed database was performed to identify studies proposing OR performance assessment after virtual-reality simulation (EyeSi or HelpMeSee) or supervised surgical training on actual patients programs. Outcome assessment was primarily based on patient safety (i.e., intra- and post- operative complications, OR performance, operating time) and secondarily based on trainee satisfaction (i.e., subjective assessment).ResultsWe reviewed 18 articles, involving 1515 participants. There were 13 using the EyeSi simulator, with 10 studies conducted in high-income countries (59%). One study used the HelpMeSee simulator and was conducted in India. The four remaining studies reported supervised surgical training on actual patients, mostly conducted in low- middle- income countries. Training programs greatly differed between studies and the level of certainty was considered low. Only four studies were randomized clinical trials. There were 17 studies (94%) proposing patient safety assessments, mainly through intraoperative complication reports (67%). Significant safety improvements were found in 80% of comparative virtual reality simulation studies. All three supervised surgery studies were observational and reported a high amount of cataract surgeries performed by trainees. However, intraoperative complication rates appeared to be higher than in virtual reality simulation studies. Trainee satisfaction was rarely assessed (17%) and did not correlate with training outcomes.ConclusionsPatient safety assessment in the OR remains a major concern when evaluating the efficacy of a training program. Virtual reality simulation appears to lead to safer outcomes compared to that of supervised surgical training on actual patients alone, which encourages its use prior to performing real cases. However, actual training programs need to be more consistent, while maintaining a balance between financial, cultural, geographical, and accessibility factors.