Masanobu Suzuki ^1* Ryosuke Watanabe ¹ Akira Nakazono ¹ Yuji Nakamaru ¹ TAKAYOSHI SUZUKI ¹ Shogo Kimura ¹ Kotaro Matoba ² Manabu Murakami ³ Dominik Hinder ⁴ Alkis J. Psaltis ⁴ Akihiro Homma ¹ Peter J. Wormald ⁴

• ¹ Department of Otolaryngology Head and Neck Surgery, Faculty of Medicine and Graduate School of Medicine, Hokkaido University, Sapporo, Hokkaidō, Japan
• ² Department of Forensic Medicine, Faculty of Medicine and Graduate School of Medicine, Hokkaido University, Sapporo, Hokkaidō, Japan
• ³ Center for Medical Education and International Relations, Graduate School of Medicine, Hokkaido University, Sapporo, Hokkaidō, Japan
• ⁴ Department of Surgery–Otorhinolaryngology Head and Neck Surgery, Central Adelaide Local Health Network and the University of Adelaide, Adelaide, Australia

Introduction Traditionally formal assessment of surgical skills has not been part of a surgeon’s accreditation process. The widely adopted apprentice model of “on-the-job training” does create additional risk for the patients. In the past surgical training has used cadavers, but these are expensive, require dedicated wet-lab facilities and are in increasingly short supply. In many countries religious and cultural practices also preclude cadaveric use. Recent 3D-printed technology allows mass reproduction of high-fidelity 3D models. In this study, we examined the utility of 3D sinus models compared to cadaver dissection for surgical skill assessment for endoscopic sinus surgery (ESS). Materials and Methods A total of 17 otolaryngologists performed Endoscopic Sinus Surgery (ESS) on 3D printed sinus models and then repeated these procedures on cadavers. Their surgical performance was assessed with the Objective Structured Assessment of Technical Skills (OSATS) score for ESS and time was taken to complete an ESS procedure. Their performance on the 3D models and cadavers was compared. Results There were no significant differences in the OSATS score between 3D models and cadavers (50.41±13.31 vs. 48.29±16.01, p=0.36). There was a strong positive correlation between the score in 3D models and those in cadavers (r=0.84, p<.001). No significant differences were found in time for a mini-ESS (21:29±0:10 vs. 20:33±0:07, p=0.53). There were positive correlations between 3D models and cadavers in time taken for a mini-ESS (r=0.55, p=0.04). Conclusion The surgical performance on the 3D models was comparable to that on cadavers. This supports the utility of the 3D models as an inexhaustible alternative for cadavers in ESS surgical skill assessment.