Radiographs are not sufficient for evaluation of component fit in subtle knee pain after total knee arthroplasty

Abstract

Purpose To determine the component fit by radiography or computed tomography after total knee arthroplasty and the relation of imaging with clinical examination of residual knee pain. Methods The study was conducted in 172 patients with residual knee pain after total knee arthroplasty. The patients were examined to determine whether they experienced pain upon palpation at nine regions surrounding the tibial and femoral components, and the results were noted. The Knee Society Clinical Rating System and The Western Ontario and McMaster Universities Arthritis Index pain scale score forms were completed for all patients. Radiologic evaluation was performed using computed tomography and anteroposterior, lateral, and oblique radiographs to determine component overhang/underhang status at these nine regions. Overhang, underhang, and cortical fit groups were created based on the position of the component at the bone margin. A statistical relationship was sought between the clinical scores and the values measured to determine which imaging method showed the best correlation with clinical scores. Consistency of CT and Rx measurements was compared using the McNemar-Bowker test. Comparisons between groups were made using Student's t test for normally distributed data, and the Mann-Whitney U test. Results Computed tomography and radiographic measurements were similar in the medial, anterior, and lateral tibial regions. However, no similarities were observed in the anteromedial, anterolateral, posteromedial, and posterolateral tibial regions, and in the distal-medial and distal-lateral aspects of the femur. Statistical relationships among decreased clinical scores, pain with palpation, and the presence of overhang/underhang were only observed in the medial tibial region for imaging using radiography. A statistically significant relationship was observed in the medial, posteromedial, and posterolateral tibial regions, and in the distal-medial region of the femur for imaging based on computed tomography. Conclusions Radiography could only aid in assessing the component fit in the anteromedial, medial, and lateral regions of the tibia in patients with residual knee pain following knee arthroplasty, but it was not sufficient in comparison with computed tomography in six other regions.