Simulation and skills training in mitral valve surgery

Objective: Limited exposure and visualization and technical complexity have affected resident training in mitral valve surgery. We propose simulation-based learning to improve skill acquisition in mitral valve surgery. Methods: After reviewing instructional video recordings of mitral annuloplasty in porcine and plastic models, 11 residents (6 integrated and 5 traditional) performed porcine model mitral annuloplasty. Video-recorded performance was reviewed by attending surgeon providing audio formative feedback superimposed on video recordings; recordings were returned to residents for review. After 3-week practice with plastic model, residents repeated porcine model mitral annuloplasty. Performance assessments initially (prefeedback) and at 3 weeks (postfeedback) were based on review of video recordings on 5-point rating scale (5, good; 3, average; 1, poor) of 11 components. Ratings were averaged for composite score. Results: Time to completion improved from mean 31 +/- 9 minutes to 25 +/- 6 minutes after 3-week practice (P = .03). At 3 weeks, improvement in technical components was achieved by all residents, with prefeedback scores varying from 2.4 +/- 0.6 for needle angles to 3.0 +/- 0.5 for depth of bites and postfeedback scores of 3.1 +/- 0.8 for tissue handling to 3.6 +/- 0.8 for suture management and tension (P <= .001). Interrater reliability was greater than 0.8. In this sample, composite scores of first-year integrated and traditional residents were lower than those of senior level residents; comparatively, third-year integrated residents demonstrated good technical proficiency. Conclusions: Simulation-based learning with formative feedback results in overall improved performance of simulated mitral annuloplasty. In complex surgical procedures, simulation may provide necessary early graduated training and practice. Importantly, a "passing" grade can be established for proficiency-based advancement. (J Thorac Cardiovasc Surg 2011;141:107-12)