Background: Accurate treatment decisions in acute ischemic stroke hinge on nuanced knowledge of neurovascular anatomy. Yet, many trainees still miss key findings, highlighting the need for stronger neurovascular education. We hypothesized that supplementing traditional lectures with 3D-printed models may improve retention and recognition of intracranial arterial anatomy in trainees. Methods: We conducted a single-institution pilot trial comparing traditional lecture-based instruction with adjunctive 3D model use in teaching neurovascular anatomy ( Figure 1 ). Neurology residents, radiology residents, and medical students were randomized to either control or 3D groups during two sessions (2024, 2025). Participants completed pre- and post-intervention assessments testing vessel identification on various imaging modalities and in various orientations. Mixed-effects models were used to analyze score changes, with fixed effects for time (pre vs. post-lecture), instructional modality (control vs. 3D), learner type, and interaction terms. Likert scale surveys were offered to assess trainee satisfaction. Results: Both groups demonstrated significant improvement from pre- to post-test (F=40.3, p<0.0001). The 3D group showed a greater mean percent score increase (20% ± 3%) compared with the control group (12% ± 4%), although the difference did not reach statistical significance (p=0.088) ( Figure 2 ). Testing the time×trainee type interaction indicated that the improvement from pre- to post-test in both groups was driven by neurology residents, who exhibited significantly greater improvements (p<0.0001) whereas radiology residents and medical students showed smaller, nonsignificant group differences ( Figure 3 ). Amongst neurology residents, there was no difference between 3D and control groups (p=0.50). Overall, 86% of participants rated the models as “somewhat” to “very” useful. Conclusions: In this pilot study, adjunctive 3D-printed models were well-received and associated with numerically greater score improvements, though no significant advantage over standard instruction was observed. Neurology residents demonstrated the largest pre- to post-test improvements, but gains were comparable between 3D and control groups. Future work should refine quiz assessments and model integration, with attention to tailoring instruction and model use to training level and specialty, to optimize their role in neurovascular education.
Molaie et al. (Thu,) studied this question.