Background: Extended Reality (XR) has demonstrated significant improvements in multiple surgical specialties by reducing operative times and enhancing surgical outcomes. Nevertheless, its usability for perforator visualization in deep inferior epigastric perforator (DIEP) breast reconstruction remains largely unexplored.

Objective: The primary aim of this study was to assess the usability and potential added value of XR in DIEP flap breast reconstruction to enhance the three-dimensional understanding of the deep inferior epigastric artery and its perforators. The secondary objective of this study was to determine if the weight of DIEP flaps can be accurately estimated through the virtual model required for XR visualization.

Methods: This multicenter pilot study involved plastic surgeons and patients undergoing DIEP flap breast reconstruction at three Dutch medical centers. Abdominal computed tomography angiography (CTA) images were used to create virtual 3D models. Surgeons utilized the HoloLens 2 (Microsoft Corporation) to compare intraoperative findings regarding perforator caliber, location within the flap, and the intramuscular course of perforators against CTA. Usability was assessed via the System Usability Scale (SUS). The SUS, perforator characteristics, and the added value of XR in addition to CTA for preoperative planning of DIEP flap breast reconstruction were scored using a 5-point Likert scale with a minimum score of 1 and a maximum score of 5. Furthermore, DIEP flap weight estimations were derived from the segmentations of the DIEP flap required for the virtual model and compared with a two-dimensional method for DIEP flap weight predictions extracted from literature.

Results: This study presents the preliminary results. Up until now, five patients and surgeons have been included. SUS scores were 66 for CTA and 70 for XR, meaning both visualization methods have ‘Good’ usability. XR scored slightly higher compared to CTA regarding perforator caliber with median scores of 3 for CTA and 4 for XR. Location within the flap yielded median scores of 4 for CTA and 5 for XR. The median score for the intramuscular course was 4 for both CTA and XR. The median scores for utilizing XR in addition to CTA for preoperative planning of DIEP flap breast reconstruction and recommending it to a colleague were both 4. DIEP flap weight prediction derived from the virtual model compared to true flap weight showed a mean difference of 333 grams and a mean percentage difference of 18.9%, whereas the two-dimensional method showed a mean difference of 937 grams and a mean percentage difference of 44.5%.

Conclusion: The preliminary results from this study indicate that XR technology, using the HoloLens 2, is a feasible and potentially advantageous tool for preoperative planning in DIEP flap breast reconstruction. The usability of XR was found to be comparable to CTA, with slightly better performance in visualizing perforator characteristics. The positive feedback from participating surgeons suggests that XR could enhance the understanding of the DIEA and its perforators, potentially improving surgical outcomes. Accurate prediction of DIEP flap weight based on the virtual model using the methodology employed in this study is currently not feasible.