Clinical Report: Shape Restoration Model for Distal Radius Malunion Osteotomy

Overview

This study developed a statistical shape completion model to reconstruct a healthy distal radius using proximal radius data, addressing limitations of contralateral reference reliance in corrective osteotomy planning. The model was trained and validated on 3D CT bone models from 100 patients, demonstrating potential to improve surgical planning accuracy and applicability.

Background

Distal radius fractures are common and often complicated by malunion, leading to pain and functional impairment. Corrective osteotomy is effective but challenging due to complex deformities. Current 3D planning methods rely on the contralateral radius as a reference, which is not always feasible due to bilateral fractures or deformities. Shape completion models offer an alternative by predicting healthy bone shape from partial data, potentially enhancing preoperative planning without additional radiation exposure.

Data Highlights

100 patients with bilateral CT scans of the radius were included after exclusion criteria. The dataset was split into training (80), validation (10), and testing (10) sets. CT scans had slice thicknesses of 0.6 or 0.75 mm, covering the entire radius. Left radii were mirrored to create a uniform dataset of right radii models. A coordinate system was defined based on anatomical landmarks to standardize model alignment. The statistical shape model was developed using the Scalismo library.

Key Findings

• A shape completion model was successfully developed to reconstruct the distal radius from proximal radius data without contralateral reference.
• Approximately 6.5% of patients with distal radius malunion lack a viable contralateral radius for surgical planning.
• 3D planning using contralateral radius improves outcomes but is limited by bilateral fractures or anatomical differences.
• The model uses a standardized coordinate system aligned with anatomical landmarks to ensure reproducibility.
• Training on 80 patient models allowed creation of a robust statistical shape model applicable to clinical cases.

Clinical Implications

This shape completion model can expand the applicability of 3D preoperative planning for distal radius malunion by eliminating the need for contralateral imaging, reducing radiation exposure and planning errors. Surgeons may achieve more accurate osteotomy planning in patients with bilateral deformities or contralateral abnormalities, potentially improving functional outcomes and reducing complications.

Conclusion

The developed shape completion model represents a promising tool for reconstructing healthy distal radius anatomy from partial data, facilitating improved corrective osteotomy planning without reliance on contralateral imaging. Further clinical validation is warranted to confirm its impact on surgical outcomes.

References

1. Meesters et al. 2023 -- 3D Planning Improves Outcomes in Distal Radius Malunion
2. Mauler et al. 2022 -- Shape Completion Models for Radius Reconstruction
3. Lüthi et al. 2018 -- Statistical Shape Modeling Techniques
4. Smees et al. 2019 -- Coordinate Systems for Radius Alignment
5. Centre for Orthopaedic Surgery OCON Data 2018-2024