An MRI-Derived Formula for Estimating the Native Joint Line Position in the Presence of Distal Femoral Bone Loss

Background In the presence of distal femoral condyle bone loss, estimation and restoration of the joint line (JL) position can be guided by extraarticular bony landmarks with the aid of mathematical formulas that rely on the innate correlations between periarticular measurements. To prevent JL elevation, the formula should incorporate the thickness of distal femoral articular cartilage. The aim of this study was to derive a formula to estimate native JL position. Methods One hundred and fifty knee magnetic resonance imaging (MRI) studies belonging to 150 patients were chosen from a database of scans. Multiple periarticular measurements were taken. Based on the strongest correlation between measurements, linear regression analysis was used to derive a regression equation to estimate the JL position. This formula was then tested to determine its accuracy and reliability in estimating the JL. Results Using the Pearson correlation test, the strongest correlation was identified to be between adductor tubercle to joint line distance (ATJL) and transepicondylar width (TEW) with r = 0.723, p <.001. Using linear regression analysis, the following regression equation was obtained: ATJL in millimetres = 0.53 (TEW in millimetres) + 2.4mm. This formula estimated the JL within 4 mm of the native JL in 86% of measured knees and within 8 mm in 100% of measured knees. The mean difference between calculated ATJL and measured ATJL was 2.43 mm with a standard deviation of 1.94 mm. Conclusion The current formula (ATJL = 0.53(TEW) + 2.4mm) reliably estimates native JL distance from the adductor tubercle (AT) to within a clinically significant range, using femoral TEW.