Perforator Mapping for Head and Neck Reconstructive Surgery: a Novel Personalized Approach using Magnetic Resonance Angiography based 3D Models and 3D-Printing

A. F. De Geer^1,2*I. Mulder¹L. C. Ter Beek³A. S. te Boekhorst⁴B. I. Plakké⁴L. H.E. Karssemakers¹R. Dirven^1,5P. J.F.M. Lohuis¹T. J.M. Ruers^6,7F. J. Siepel²P. K. de Koekkoek-Doll⁴M. J.A. Van Alphen¹W. H. Schreuder^1,8

• ¹Verwelius 3D Lab, Department of Head and Neck Surgery and Oncology, Netherlands Cancer Institute - Antoni van Leeuwenhoek, Amsterdam, Netherlands
• ²Robotics and Mechatronics, University of Twente, Enschede, Netherlands
• ³Department of Medical Physics, Netherlands Cancer Institute - Antoni van Leeuwenhoek, Amsterdam, Netherlands
• ⁴Department of Radiology, Netherlands Cancer Institute - Antoni van Leeuwenhoek, Amsterdam, Netherlands
• ⁵Department of Otorhinolaryngology and Head and Neck Surgery, Radboud university medical center, Nijmegen, Netherlands
• ⁶Department of Surgery, Netherlands Cancer Institute - Antoni van Leeuwenhoek, Amsterdam, Netherlands
• ⁷Faculty of Science and Technology, University of Twente, Enschede, Netherlands
• ⁸Department of Oral and Maxillofacial Surgery, Amsterdam Medical University Center and Academic Center for Dentistry Amsterdam (ACTA), Amsterdam, Netherlands

Background: Reconstruction of large head and neck defects in oncologic patients often requires free vascularized tissue flaps. Successful flap design and elevation depend on accurate preoperative identification of perforator vessels. Preoperative Magnetic Resonance Angiography (MRA) could offer detailed insights into perforator course, caliber, origin, and main pedicle length, and thus is expected to surpass conventional handheld Doppler. This study introduces a novel approach for perforator mapping in reconstructive head and neck surgery that integrates MRA with 3D modelling and 3D-printing Methods: The proposed workflow comprises four steps: 1) acquisition of contrast-enhanced MRA, 2) construction of a 3D anatomical model, 3) design and 3D-printing of a patient-specific perforator guide, and 4) transfer of perforator locations from the model to the patient's skin using the guide. To illustrate the clinical feasibility and potential utility of this approach, an initial cohort of patients undergoing perforator flap surgery for oncologic head and neck reconstruction was included. Flap types included fibula free flap (FFF), anterolateral thigh flap (ALT), and medial sural artery perforator flap (MSAP). Intraoperative findings were compared with the 3D models, and surgeons evaluated the models' usability for virtual planning of flap design and elevation using a five-point Likert scale questionnaire.Results: Ten patients were included: three FFF, two ALT, and five MSAP cases. In FFF and ALT patients, all perforators intraoperatively used for flap elevation were successfully visualized on MRA and represented in the 3D models. In MSAP patients, small-caliber perforators were not consistently visible. The mean absolute difference between pedicle lengths measured in the 3D models and intraoperatively was 1.0 cm (SD 0.9 cm). The usability questionnaire yielded an average score of 4.2 out of 5, suggesting the potential of MRA-based 3D models for virtual surgical flap planning.Conclusions: This is the first study to combine preoperative MRA with 3D modelling and 3Dprinting for perforator mapping in head and neck reconstruction. The workflow offers a radiationfree, patient-specific planning tool that may enhance surgical precision and support personalized flap design in complex oncological cases.