Integrating Virtual Reality and Patient-Specific 3D Models for Staged Craniopagus Twins Separation: Evolution of a Surgical Planning Pipeline Across 3 Consecutive Cases

BACKGROUND AND OBJECTIVES: Craniopagus twins (CPT) represent one of the rarest forms of conjoined twinning, and their surgical separation demands meticulous preoperative planning. We describe the evolution of an integrated surgical planning pipeline combining patient-specific 3-dimensional (3D) printed models, tissue-mimicking materials (TMM), and immersive virtual reality (VR) across 3 consecutive CPT cases at a single institution, and discuss the emerging role of artificial intelligence-driven segmentation as the next translational step. METHODS: Three CPT cases were managed at the Ribeirão Preto Medical School (University of São Paulo, Brazil). Cases 1 and 2 were completed separations using an incremental prototyping pipeline: computed tomography/MRI-based segmentation, virtual modeling, fused deposition modeling 3D printing with polylactic acid, and silicone rubber or latex covering as TMM. Case 3 continued immersive VR visualization using a Meta Quest 3 headset, enabling the multidisciplinary team to rehearse surgical steps in a simulated operating room. RESULTS: Thirteen patient-specific models were produced across the 3 cases over approximately 36 months. The pipeline matured progressively: Case 1 established the core workflow; case 2 refined TMM application and skin-surface calculations; cases 2 and 3 integrated VR-based immersive briefings. Separation was achieved in all 3 pairs; in case 3, however, both twins died intraoperatively from hemorrhage and cardiogenic shock with refractory circulatory collapse, despite anatomically accurate planning models. Team feedback indicated that VR enhanced spatial understanding of the shared vascular anatomy and improved preoperative coordination vs flat-screen visualization. CONCLUSION: The progressive integration of VR into an established 3D printing pipeline represents a meaningful advance in CPT separation planning. Artificial intelligence-driven segmentation tools hold promise for automating the bottleneck step, potentially making this pipeline replicable across institutions with fewer specialized resources.