Valgus Stress Radiograph Can Predict Alignment Change of Medial Mobile‐Bearing Unicompartmental Knee Arthroplasty

OBJECTIVE: Finding reliable tools to predict alignment change after medial mobile‐bearing unicompartmental knee arthroplasty (UKA) can help surgeons avoid under‐ or over‐correction. This prospective study aimed to investigate whether the parameters related to medial collateral ligament tension on valgus stress radiograph can predict the alignment change of medial mobile‐bearing UKA and establish a prediction model. METHODS: This study prospectively included the patients undergoing medial mobile‐bearing UKA for knee osteoarthritis from November 2018 and April 2021. Patients took valgus stress radiograph and MRI preoperatively and took full‐length weight‐bearing anterior–posterior radiograph of the lower extremity preoperatively and postoperatively. The medial joint space width (MJSW) on valgus stress radiograph, area of femoral and tibial osteophyte on MRI, medial extrusion distance (MED) of the meniscus on MRI, and the change in hip‐knee‐ankle angle (∆HKAA) were measured. Factors influencing ∆HKAA were analyzed by correlation analysis. Univariable and multivariable linear regression analysis was performed to establish a prediction model of ∆HKAA. RESULTS: One hundred and seven knees were included. On average, the preoperative HKAA was 170.84° ± 3.73°, and UKA corrected the alignment to 175.16° ± 3.21° postoperatively (p < 0.001), with ∆HKAA of 4.33° ± 1.93°. Correlation analysis showed that ∆HKAA correlated with MJSW (r = 0.628, p < 0.001), MED (r = 0.262, p < 0.001), and tibial osteophyte area (r = 0.235, p < 0.001). The prediction model for ∆HKAA was obtained by multivariable linear regression: ∆HKAA = −2.003 + 0.947 × MJSW(mm) + 1.838 × total osteophyte area(cm(2)). CONCLUSION: Valgus stress radiographic MJSW and osteophyte area are correlated to the alignment change of medial mobile‐bearing UKA. The prediction model for HKAA change is: ∆HKAA = −2.003 + 0.947 × MJSW(mm) + 1.838 × total osteophyte area(cm(2)).