Simulation of Anterior Cruciate Ligament Reconstruction in a Dry Model

Abstract

Background: As the demand increases for demonstration of competence in surgical skill, the need for validated assessment tools also increases. Hypothesis/Purpose: The purpose of this study was to validate a dry knee model for the assessment of performance of anterior cruciate ligament reconstruction (ACLR). The hypothesis was that the combination of a checklist and a previously validated global rating scale would be a valid and reliable means of assessing ACLR when performed by residents in a dry model. Study Design: Controlled laboratory study. Methods: All residents, sports medicine staff, and fellows were invited to perform a hamstring ACLR using anteromedial drilling and Endobutton fixation on a dry model of an anterior cruciate ligament. Previous exposure to knee arthroscopy and ACLR was recorded. A detailed surgical manuscript and technique video were sent to all participants before the study. Residents were evaluated by staff surgeons with task-specific checklists created by use of a modified Delphi procedure and the Arthroscopic Surgical Skill Evaluation Tool (ASSET). Each procedure (hand movements and arthroscopic video) was recorded and scored by a fellow blinded to the year of training of each participant. Results: A total of 29 residents, 5 fellows, and 6 staff surgeons (40 participants total) performed an ACLR on the dry model. The internal reliability (Cronbach alpha) of the test when using the total ASSET score was very high (>0.9). One-way analysis of variance for the total ASSET score and the total checklist score demonstrated a difference between participants based on year of training ( P < .001). Post hoc analysis demonstrated a significant difference in global ratings and checklist scores between junior residents (postgraduate year [PGY] 1-3) and senior residents (PGY 4 and 5) and senior residents and fellows ( P < .05). A significant difference was seen between fellows and staff on the global rating ( P < .05) but not on the checklist scores ( P > .05). A good correlation was seen between the total ASSET score and prior exposure to knee arthroscopy (0.73) and ACLR (0.65). The interrater reliability (intraclass correlation coefficient) between the examiner ratings and the blinded assessor ratings for the total ASSET score was very high (>0.8). Conclusion: The results of this study provide evidence that the performance of an ACLR in a dry model is a reliable method of assessing a resident’s knowledge of the steps and instrumentation required, and the method shows evidence of validity. Clinical Relevance: These models may be used to ensure a minimal level of competence before residents and fellows perform ACLR in the operating room.