Comparing Feedback Techniques in Facial Reconstruction Using 3D-Printed Models

Introduction: 3D-printed facial models have been validated as a realistic and affordable tool for teaching facial reconstruction. Further study is needed on how to best employ this tool. This study aims to assess the role of feedback techniques on confidence and quality of facial plastic repairs.

Methods: A 3D-printed facial model with a nasal ala defect was provided to 18 subjects ranging from medical students to otolaryngology physicians in training. Subjects were tasked to design and repair the defect immediately after a lecture on bilobe flaps. Three feedback techniques were utilized in the process: immediate live (group 1), delayed (group 2), and none (group 3). Subjects were randomized into one of three groups. All groups repeated the same repair on the contralateral ala independently approximately six weeks later. Pre- and post-exercise Likert surveys were completed, and all flap repairs were graded by a facial plastic surgeon.

Results: Groups 1 and 2 reported moderate improvement in confidence following completion of exercise one versus those without feedback (75% vs 0%, p=0.013). There was no difference among groups after exercise one on time to completion (p=0.227), total score (p=0.102) and overall flap appearance (p=0.299). While there was a significant improvement in confidence for group 1 and a trend in group 2 following exercise one (p=0.008 and p=0.072, respectively) when compared to baseline, there was no significant change in confidence following exercise two in any group when compared to exercise one (p=0.987). When compared to exercise one, there was no difference among groups between time to completion (p=0.242), total score (p=0.412), and overall appearance (p=0.717).

Conclusions: Use of 3D-printed facial models can improve trainee confidence in performing bilobe flaps when feedback is implemented following a didactic session. This study suggests that there is limited objective improvement when a task is repeated without guidance regardless on how the material was originally delivered. This suggests that facial reconstruction warrants guided exercises, which can be facilitated with 3D models.