Radiographic Risk Factors for Excessive Joint Line Obliquity After Knee Osteotomy for Medial Osteoarthritis: A Phenotype-Based Approach.

Background: In patients undergoing high tibial osteotomy (HTO), an excessive increase in joint line obliquity (JLO) after surgery leads to poor clinical outcomes. Phenotype analysis is a simple and intuitive method to classify knee alignment, including JLO. However, there are no studies investigating phenotype changes before and after surgery in patients undergoing realignment osteotomy.

Purpose: To investigate (1) the preoperative and postoperative phenotype distribution in patients undergoing primary realignment osteotomy for medial compartmental osteoarthritis and (2) the incidence and radiographic risk factors for excessive postoperative JLO.

Study design: Cross-sectional study; Level of evidence, 3.

Methods: A total of 348 knees (320 patients) undergoing primary realignment osteotomy for medial compartmental osteoarthritis with a varus deformity from January 2010 to July 2021 were included. Preoperatively and at 1 year postoperatively, we evaluated coronal-plane alignment on standing scanograms, classifying them into 9 phenotypes based on the modified Coronal Plane Alignment of the Knee classification. We identified changes in the preoperative and postoperative phenotype distribution through scatterplots and analyzed differences in excessive postoperative JLO (arithmetic JLO [aJLO] >4°) based on the preoperative phenotype. Receiver operating characteristic analysis identified preoperative radiographic parameters and their cut-off values as risk factors for excessive postoperative JLO after HTO.

Results: The most common preoperative phenotype was type I (varus and apex distal; n = 149 [42.8%]), and the most common postoperative phenotype was type VI (valgus and apex neutral; n = 187 [53.7%]). The incidence of excessive postoperative JLO was significantly higher in preoperative type IV (varus and apex neutral; 14.5%) than in type I (0.7%), type II (neutral and apex distal; 0.0%), and type V (neutral and apex neutral; 10.4%). In receiver operating characteristic analysis for excessive postoperative JLO, the preoperative mechanical lateral distal femoral angle (mLDFA) and aJLO showed a high area under the curve of 0.916 and 0.914, respectively, with cut-off values of 90.2° and -1.5°, respectively. In the preoperative mLDFA ≥90.2° group, the incidence of excessive postoperative JLO was significantly higher than in the preoperative mLDFA <90.2° group (19.8% vs 0.4%, respectively; P < .001). In the preoperative aJLO ≥-1.5° group, the incidence of excessive postoperative JLO was significantly higher than in the preoperative aJLO <-1.5° group (35.1% vs 1.4%, respectively; P < .001).

Conclusion: Preoperative type IV (varus and apex neutral) had a higher risk of excessive postoperative JLO. Specifically, when the preoperative mLDFA was ≥90.2° or the preoperative aJLO was ≥-1.5°, there was an increased risk of excessive postoperative JLO after isolated HTO.