Exploring discrepancies between in vivo and simulated correction in 3D-planned distal radius osteotomies: the influence of biological factors

Abstract

Three-dimensional virtual surgical planning (3D VSP) with patient-specific surgical guides (PSSGs) can improve the accuracy of corrective distal radius osteotomy. Simulation on 3D-printed bone models allows controlled evaluation of planned corrections, but such models do not reproduce biological factors, including bone quality and soft-tissue tension. The extent to which simulated corrections reflect in vivo surgical results remains unclear. This exploratory study included 13 patients who underwent corrective distal radius osteotomy using 3D VSP and PSSGs. Simulated osteotomies were performed on patient-specific 3D-printed bone models. Residual correction errors in volar tilt (VT), radial inclination (RI), and radial length (RL) were calculated relative to the 3D VSP for both in vivo and simulated corrections. Discrepancies between in vivo and simulated corrections were assessed, and their associations with age, bone mineral density estimated from Hounsfield units, and planned maximum distraction were explored using Spearman’s rank correlation. Residual correction errors in simulated osteotomies were generally comparable to those observed in vivo. The largest differences between simulated and in vivo corrections were observed for VT (3.2), although the mean difference remained within clinically acceptable limits. A positive correlation was found between RL discrepancy and planned maximum distraction (rₛ = 0.71, 95% CI: 0.27–0.91, p = 0.01), and a negative correlation was found between RL discrepancy and bone mineral density (rₛ = −0.64, 95% CI: −0.88 – −0.13, p = 0.02). No correlations were identified for VT or RI. Simulated distal radius osteotomies on 3D-printed bone models produced correction errors broadly comparable to those achieved in vivo. However, small but systematic discrepancies in VT and RL suggest that biological factors may influence the relation between simulated and clinical outcomes. These findings support the use of simulation models for comparative studies while highlighting the need for cautious interpretation of simulation-based results.