# Accuracy of CAD/CAM planned mandibular reconstruction with scapula free flaps in reconstructive head and neck surgery – A single center study

## TL;DR

This study evaluates how accurately CAD/CAM planning predicts the outcome of mandibular reconstruction using scapula free flaps in head and neck surgery.

## Contribution

The study is the first to assess the transfer precision of CAD/CAM in mandibular reconstruction with scapula free flaps.

## Key findings

- Postoperative deviations in scapular segment positioning were observed, with larger deviations in the posterior region.
- Mean segmental gap width deviations were statistically significant at all four measurement sites.
- Soft tissue bulk around the scapula may hinder the accuracy of cutting guides in CAD/CAM-assisted reconstructions.

## Abstract

For the first time, this study explores the transfer precision of computer aided design and computer aided manufacturing (CAD/CAM) in mandibular reconstruction with scapula free flaps (SFF).

Patients undergoing mandibular reconstruction with SFF and patient-specific 3D-printed titanium plates were retrospectively analyzed. To assess planning accuracy, pre- and postoperative datasets were overlaid using 3D Slicer software. Specific measurements included the fit at the mandibular resection site, intersegmental gap width between scapula and mandible, scapular segment positioning, and condylar alignment. Osseous union was evaluated via postoperative computed tomography (CT) or cone beam CT, and plate-related complications were recorded clinically.

In total 15 patients (mean age 65.7 ± 10.0 years) were included. Postoperative scapular segment positioning showed deviations in the posterior region (basal: 6.78 ± 3.34 mm; crestal: 7.61 ± 3.40 mm) and anterior region (basal: 3.54 ± 1.52 mm; crestal: 4.49 ± 2.51 mm). Mean deviation in segmental gap width at the anterior and posterior sites ranged from 0.55 to 1.72 mm and was statistically significant at all four measurements. Mean contralateral condyle position differed significantly from the initial position (8.56 ± 4.31 mm). Intraoral plate exposure occurred in three patients without adjuvant radiotherapy; no extraoral exposure, fractures, or loosening were observed. Non-union between the mandible and SFF was found in three patients.

As with other CAD/CAM-assisted mandibular reconstructions, SFF exhibit discrepancies between virtual planning and the postoperative outcome. The surrounding soft tissue bulk of the scapula can hinder the precise fit of cutting guides, potentially compromising reconstruction accuracy and contributing to increased deviations. Despite these challenges, CAD/CAM-assisted SFF demonstrates acceptable surgical predictability.