Collaborative Team Training in Virtual Reality is Superior to Individual Learning For Performing Complex Open Surgery: A Randomized Controlled Trial

Abstract

Objective: To assess whether multiplayer immersive Virtual Reality (iVR) training was superior to single-player training for the acquisition of both technical and nontechnical skills in learning complex surgery.

Background: Superior teamwork in the operating room (OR) is associated with improved technical performance and clinical outcomes. iVR can successfully train OR staff individually; however, iVR team training has yet to be investigated.

Methods: Forty participants were randomized to individual or team iVR training. Individually trained participants practiced alongside virtual avatar counterparts, whereas teams trained live in pairs. Both groups underwent 5 iVR training sessions over 6 weeks. Subsequently, they completed a real-life assessment in which they performed anterior approach total hip arthroplasty surgery on a high-fidelity model with real equipment in a simulated OR. Teams performed together, and individually trained participants were randomly paired up. Videos were marked by 2 blinded assessors recording the 'Non-Operative Technical Skills for Surgeons, Oxford NOn-TECHnical Skills II and Scrub Practitioners' List of Intraoperative Non-Technical Skills' scores. Secondary outcomes were procedure duration and the number of technical errors.

Results: Teams outperformed individually trained participants for nontechnical skills in the real-world assessment (Non-Operative Technical Skills for Surgeons: 13.1±1.5 vs 10.6±1.6, P = 0.002, Non-TECHnical Skills II score: 51.7 ± 5.5 vs 42.3 ± 5.6, P = 0.001 and Scrub Practitioners' List of Intraoperative Non-Technical Skills: 10 ± 1.2 vs 7.9 ± 1.6, P = 0.004). They completed the assessment 33% faster (28.2 minutes ± 5.5 vs 41.8 ± 8.9, P < 0.001), and made fewer than half the number of technical errors (10.4 ± 6.1 vs 22.6 ± 5.4, P < 0.001).

Conclusions: Multiplayer training leads to faster surgery with fewer technical errors and the development of superior nontechnical skills.

FIGURE 1

The equipment set up for the multiplayer mode. The images in the top right and left demonstrate the view through the headsets for the surgeon (right) and scrub technician (left). The image below shows the hardware (headset and motion-tracked controllers) being used in a team training session.

FIGURE 2

The setup for the real-world assessment using the distributed simulation. A, Demonstrates the overall setup with equipment, personnel, and model. B, Shows a participant broaching the femur during the simulated operation.

FIGURE 3

Column scatter plots demonstrating the nontechnical performance in the real-world assessment for the 3 measured scores: (A) NOTSS (B) NOTECHS-II, and (C) SPLINTS, for team (T) and solo (S) trained participants. The central horizontal line within the box shows the mean. The whiskers demonstrate the SD. Significant P values (<0.05) are indicated.

FIGURE 4

Column scatter plots demonstrating the technical performance metrics in the real-world assessment: (A) technical error count and (B) procedure duration in minutes, for team (T) and solo (S) trained participants. The central horizontal line within the box shows the mean. The whiskers demonstrate the SD. Significant P values (<0.05) are indicated.