Simulation-based models for endovascular and microsurgery training in brain arteriovenous malformations: A systematic review

Abstract

Background

Simulation-based training has become increasingly important in cerebrovascular education; however, available arteriovenous malformation (AVM) models vary in fidelity, accessibility, and educational focus. This systematic review aims to identify, classify, and compare simulation-based and experimental AVM models used for operative neurosurgical and endovascular training.

Methods

A PRISMA-compliant systematic review of PubMed was conducted. English-language studies describing AVM-specific simulation or experimental models for operative/microsurgical and/or endovascular training were included. Extracted data included model type, targeted training domain, procedural tasks, anatomic and flow fidelity, feasibility, and reported educational outcomes. Results were synthesized qualitatively.

Results

Twenty studies met the inclusion criteria. AVM training platforms were categorized into five groups: in vivo animal models, ex vivo biological tissue models, in vitro synthetic flow models, 3D-printed physical models, and digital visualization platforms. Most models emphasized endovascular skill acquisition (n = 13), supporting catheter navigation, embolic delivery, and hemodynamic assessment. Operative microsurgical training was addressed less frequently (n = 3), with ex vivo placenta-based models providing the highest tactile fidelity for vessel handling, dissection, and flow assessment. Visualization and planning platforms (n = 4), including 3D printing and VR/AR, enhanced three-dimensional anatomic understanding but lacked haptic feedback. Educational validation was heterogeneous and relied predominantly on subjective assessment.

Conclusion

Current AVM simulation platforms predominantly support endovascular training, whereas high-fidelity operative microsurgical simulation remains limited. A competency-driven, multimodal training strategy integrating complementary endovascular, microsurgical, and cognitive platforms is warranted. Future efforts should prioritize the development of an operation-focused model and rigorous educational validation aligned with the technical demands of contemporary operative neurosurgery.