Accuracy of Patient-Specific Instrument for Cylindrical Axis Implementation in Kinematically Aligned Total Knee Arthroplasty

Factors Causing Unintended Sagittal and Axial Alignment Changes in High Tibial Osteotomy: Comparative 3-Dimensional Analysis of Simulation and Actual Surgery

BACKGROUND

Unintended secondary changes in the posterior tibial slope (PTS) and tibial torsion angle (TTA) may occur after medial open-wedge high tibial osteotomy (MOWHTO). In surgical procedures using patient-specific instruments (PSIs), it is essential to reproduce the PTS and TTA that were planned in simulations.

PURPOSE

To analyze the factors causing unintended sagittal and axial alignment changes after MOWHTO.

STUDY DESIGN

Case series; Level of evidence, 4.

METHODS

Overall, 63 patients (70 knees) who underwent MOWHTO using a PSI between June 2020 and June 2023 were retrospectively reviewed. Preoperative and postoperative computed tomography scans were 3-dimensionally reconstructed. Simulated osteotomy was performed so that the weightbearing line could pass through the target point. A PSI gapper was 3-dimensionally printed to fit the posteromedial corner of the osteotomy gap in the simulated HTO model. After MOWHTO using the PSI gapper, the actual postoperative model was compared with the preoperative or simulation model. This assessment included PTS, TTA, hinge axis, and osteotomy-related parameters. Cortical breakage around the lateral hinge was evaluated to assess stability.

RESULTS

The mean PTS and TTA did not change in the simulation. However, significant changes were observed in the actual postoperative PTS and TTA (change, -2.4°± 2.2° and -3.9°± 4.7°, respectively). The PTS was reduced, while the TTA decreased with internal rotation of the distal fragment. The difference in the axial hinge axis angle (AHA) between the simulation and actual surgery was the factor most correlated with the difference in the PTS (r = 0.625; P < .001). In regression analysis, the difference in the AHA was the only factor associated with the difference in the PTS (β = 0.558; P = .001), and there were no factors that showed any significant associations with the difference in the TTA. In subgroup analyses for the change in the TTA, the correction angle and anterior osteotomy angle were significantly higher in the more internal rotation group (P = .023 and P = .010, respectively). The TTA change was significantly higher in the unstable group with lateral cortical breakage (P = .018). The unstable group was more likely to show an internal rotation of ≥5° (odds ratio, 5.0; P = .007).

CONCLUSION

The AHA was associated with a difference in the PTS between the simulation and actual surgery. The change in the TTA was caused by a combination of multiple factors, such as a large correction angle and anterior osteotomy angle, but mainly by instability of the lateral cortical hinge.

Preoperative joint line obliquity, a newly identified factor for overcorrection, can be incorporated into a novel preoperative planning method to optimise alignment in high tibial osteotomy

PURPOSE

The aim of this study was to analyse the factors associated with additional postoperative alignment changes after accurate bony correction by selecting only patients with well-performed bony correction as planned and develop a method of incorporating significant factors into preoperative planning.

METHODS

Among 104 consecutive patients who underwent medial open wedge high tibial osteotomy (MOWHTO) between October 2019 and July 2022, 61 with well-performed bony corrections were retrospectively reviewed. The major criterion for well-performed bony correction was a difference of <1° between the simulated medial proximal tibial angle (MPTA) and the actual postoperative MPTA as measured in three dimensions. Radiographic parameters, such as the joint line convergence angle (JLCA) and joint line obliquity (JLO), were measured preoperatively and postoperatively, utilising standing and supine whole lower extremity anteroposterior, valgus and varus stress radiographs. Multiple linear regression analysis identified the factors affecting alignment changes, and a prediction model was developed. A method for applying this prediction model to preoperative planning was proposed.

RESULTS

Preoperative JLCA on standing (preJLCAstd ), preoperative JLCA on 0° valgus stress radiograph (vgJLCA0 ), and preoperative JLO (preJLO) were significantly correlated with JLCA change (∆JLCA) (p < 0.001, p < 0.001, p = 0.006). The prediction model was estimated as ∆JLCA = 0.493 × (vgJLCA0 ) - 0.727 × (preJLCAstd ) + 0.189 × (preJLO) - 1.587 in. (R = 0.815, modified R2  = 0.646, p < 0.001). The proposed method resulted in a reduced overcorrection rate (p = 0.003) and an improved proportion of acceptable alignments (p = 0.013).

CONCLUSION

PreJLCAstd , vgJLCA0  and preJLO can be used to estimate ∆JLCA. PreJLO was recently identified as a significant factor associated with additional alignment changes. Utilising the proposed preoperative planning and a prediction model with these factors shows promise in calibrating postoperative alignment after MOWHTO.

LEVEL OF EVIDENCE

Level III, retrospective cohort study.