Sir,

Virtual reality surgical simulation training improves resident performance as measured by the simulator itself and wet-lab performance.1, 2 However, there is little published data showing that simulators improve ‘real life’ resident surgery performance.3

We performed a retrospective comparative consecutive study of resident surgical outcomes at our institution (approved by the Institutional Review Board of Henry Ford Health System, Detroit, MI, USA) before and after the introduction of the Eyesi surgical simulator (VRmagic, Mannheim, Germany). The first 50 phacoemulsification cases of 20 residents (5 residents completing residency per academic year, 2007–2010) were studied. Simulator-group residents were required to complete at least 6 h of simulator training. All 20 residents received significant wet-lab training (24 h structured course, including 4 h of wet-lab phaco), and surgical case supervision and coaching. Both the simulator and the wet-lab were easily accessible for residents who wanted additional training. Outcome measures were the incidence of posterior capsule tears and operating time.

The nonsimulator and simulator groups each comprised 500 cases with 40 and 35 posterior capsule tears, respectively. Analysis of the first 25 cases of each resident yielded complication rates of 8.8% and 10.0% for the nonsimulator and simulator groups, respectively, and 7.2% and 3.6% (P=0.11) for cases 26 through 50, respectively. The percentage of long cases (defined as >40 min) for cases 10 through 50 was 42.3% and 32.4% (P=0.005) for the nonsimulator and simulator groups, respectively (Figure 1).

Figure 1
figure 1

Bar graph of the percentage of resident cases with surgical time greater than 40 min. Each 10 cases for each of the 10 residents in the two groups (∼100 cases per group) are compared using the χ2 test.

Full size image

Comment

As the simulator residents performed better for their later cases, it appears that simulator training was beneficial for trainees who have commenced ‘live surgery’ training. Possible reasons for the nonsimulator residents performing better for their initial cases could be more attending intervention or conversely, superior natural resident ability. Because of the retrospective data collection, we were not able to differentiate between these possibilities. Other study limitations were that variation in resident and patient characteristics could not be accounted for. Study strengths include consistent surgical technique (divide and conquer), instrumentation, and surgical instructors. Our study supports the suggestion that the addition of virtual reality surgical training to an established surgical training program slightly shortens the learning curve for the first 50 phacoemulsification cases.