Effect of case-specific 3D-printed models on surgical performance in cadaveric dissection—a randomized controlled trial

PURPOSE: To investigate if training with case-specific 3D-printed temporal bone (TB) models improves trainees’ mastoidectomy performance in cadaveric dissection compared with training on a standard, generic-anatomy 3D-printed TB model. METHODS: In this randomized controlled trial, 22 novice ORL residents performed a cadaveric mastoidectomy with posterior tympanotomy after training two procedures either on 3D-printed TB models of their specific cadaveric bone (CM, n = 11), or on 3D-printed TB with generic anatomy (GM, n = 11). Blinded experts assessed final-product performance using a modified Welling Scale. Secondary outcomes included correlation between performance on 3D-printed models and cadavers, questionnaires on model quality, and usefulness for planning. RESULTS: On their first 3D-printed model, the CM group performed better than the GM group (13.1 vs. 10.9 points; p < 0.01). In their second 3D-printed procedure, scores were similar (15.8 vs. 15.7; p = 0.9) as both groups improved significantly from the first to the second procedure (p < 0.001). In subsequent cadaver dissection, the CM group scored lower than the GM group (13.2 vs. 14.4; p < 0.001). Correlation between performance on 3D-printed models and cadavers was not significant in the CM group (r = 0.45; p = 0.16) but very strong in the GM group (r = 0.77; p < 0.01). Questionnaire responses indicated high perceived realism and educational value in both groups. CONCLUSION: Training with case-specific 3D-printed TB models did not benefit or correlate with cadaveric mastoidectomy performance in novice residents. This supports the use of generic-anatomy, educational 3D-printed TB models in the early training stages and highlights the importance of aligning simulation fidelity with learner experience. TRIAL REGISTRATION: Not applicable.