Effect of Medial and Lateral Posterior Condylar Offset on Knee Flexion after Mobile-bearing Total Knee Arthroplasty.

after Mobile-bearing Total Knee Arthroplasty +Ishii, Y; Noguchi, H; Takeda, M; Sato, J +Ishii Orthopaedic & Rehabilitation Clinic, Gyoda, Saitama, JAPAN [email protected] INTRODUCTION: Postoperative studies using X-ray analyses have reported contradictory findings regarding whether PCO had an effect on flexion after total knee arthroplasty (TKA), since correctly recognizing which condyle reflects the results of the TKA may be difficult with conventional radiographic techniques. To address these issues, it is crucial for surgeons to assess the change in each condyle. A threedimensional (3D) lower extremity alignment assessment system (Knee CAS; LEXI, Inc., Tokyo, Japan) has been developed which combines data from computed radiography (CR) and computed tomography (CT) and enables the detection of changes in each condyle. The purpose of this study was to analyze the correlation between the changes in each PCO and the postflexion angle in posterior cruciate ligament-retaining (PCLR) and -sacrificing (PCLS) mobile-bearing prostheses. METHODS: The local institutional review board approved this study. All patients provided informed consent. We evaluated 138 knees from 124 patients who underwent TKA with the LCS Total Knee System (DePuy, Warsaw, IN). Sixty-six knees received meniscal bearing-type (MB; PCLR) prostheses, and 72 knees received rotating platform-type (RP; PCLS) prostheses (Table 1). The two prosthesis designs had the same geometry in the coronal plane; however, the PCLR design had nonconstrained anteroposterior (AP) and rotational movement, and the PCLS design has only nonconstrained rotational movement. The LCS femoral component has an anatomical articulating surface, and the radii of curvature decrease posteriorly. The flexion angle was assessed with a standard hand-held goniometer preoperatively and 12 months postoperatively. Computed tomographic evaluation of the condylar offset: We used a quantitative 3D technique developed by Sato et al. for the assessment of changes in both the medial and lateral femoral condylar offsets. This assessment required CT scanning of each patient’s femur and tibia before surgery. In addition, biplanar CR images of the lower extremities were acquired before and after TKA. The biplanar CR images were downloaded to a PC with the 3D lower extremity alignment assessment system (Knee CAS; LEXI, Inc., Tokyo, Japan). The maximum spatial errors of this procedure were 0.5 mm when determining distance. We measured the maximum thicknesses of the medial and lateral posterior condyles from the edge of the condyle to a line tangent to the posterior cortex of the femoral shaft (Fig 1a, 1b). Spearman’s rank correlation coefficient was used to evaluate relationships between changes in each PCO and the post-flexion angle. RESULTS: In PCLR meniscal-bearing knees, the mean medial PCO was 26.3 ± 2.9 mm preoperatively and 26.3 ± 4.2 mm postoperatively. The difference between preand postoperative offsets was 0.0 ± 3.7 mm. The mean lateral PCO was 25.3 ± 2.4 mm preoperatively and 28.9 ± 3.3 mm postoperatively. The difference between preand postoperative offsets was 3.7 ± 3.7 mm. The maximum flexion angle was 120 ± 17° preoperatively and 114 ± 14° postoperatively. The difference was -6 ±15°. There were no significant correlations between the changes in the PCOs and the post-flexion angles (post-flexion angle vs. change in medial condyle: R = 0.122, p = 0.325; post-flexion angle vs. change in lateral condyle: R = -0.063, p = 0.610). In PCLS rotating platform knees, the mean medial PCO was 26.1 ± 2.3 mm preoperatively and 25.7 ± 3.5 mm post-operatively. The difference between preand postoperative offsets was -0.4 ± 3.3 mm. The mean lateral PCO was 25.0 ± 2.4 mm preoperatively and 28.3 ± 4.2 mm postoperatively. The difference between preand postoperative offsets was 3.3 ± 4.1 mm. The maximum flexion angle was 114 ± 21° preoperatively and 112 ±16° post-operatively. The difference was -2 ±19°. There were no significant correlations between the changes in the PCOs and the post flexion angles (post-flexion angle vs. change in medial condyle: R = 0.020, p = 0.864; post-flexion angle vs. change in lateral condyle: R = -0.217, p = 0.068). DISCUSSION: The most important finding of this study with CT-based evaluations was that we found no statistical correlation between the changes in medial and lateral PCO after surgery and the changes in knee flexion in both mobile-bearing PCLR and PCLS prostheses. Therefore, for the current prosthesis designs, changes in PCO were not a significant factor for the degree of knee flexion achieved after TKA. Recently, with the relation to the evaluation of PCO, Ishii et al. reported that changes in PCO based on X-ray evaluations showed no significant correlations with the changes observed in the CT-evaluated medial and lateral PCOs. There are three possible main reasons for this discrepancy. First, the femoral condyles are asymmetric in shape and dimension before surgery. Second, during surgery, it is usually necessary to remove an asymmetric portion of bone from the posterior femoral condyles to equalize the length of the soft tissue and properly align the rotation of the femur, particularly for a rectangular flexion gap. Third, a different magnification of the medial condyle from that of the lateral condyle might be induced by X-ray evaluations. The magnification of the lateral condyle was always less than that of the medial condyle because the lateral side was always closer to the film plate when taking lateral radiographs or performing videofluoloscopic procedures. Most studies corrected for the discrepancy in magnitude between preand postoperative radiographs, but not between the medial and lateral condyles because of the limitations of radiographic evaluations. In both current mobile-bearing PCLR and PCLS prostheses, the in vivo kinematics and flexion angle were shown to have similar results, although the PCLR design has non-constrained AP and rotational movement and the PCLS design has only non-constrained rotational movement. Both designs showed approximately 1 mm of anteroposterior movement between 0 and 90 of flexion. Thus, both prosthesis designs showed almost the same positioning of the femoral component on the tibial component during knee flexion. Although we did not perform a kinematic study of both designs in this study, we recognized that the current PCLR prosthesis designs, with a 10 posterior tibial slope, had less of an effect on the non-constrained AP movement and thus did not significantly affect the paradoxical anterior movement with flexion as effectively as did the PCLS design, which has only non-constrained rotational movement. In conclusion, several factors have been recognized to influence knee flexion after TKA. In this study, we investigated the influence of medial and lateral PCO. We found that changes in individual posterior condylar offsets were not correlated with knee flexion one year after TKA with the current prosthesis designs. SIGNIFICANCE: Changes in PCO with CT-based evaluations were not a significant factor for the degree of knee flexion achieved after TKA for the current prosthesis designs. REFERENCES: 1. Sato T. J Arthroplasty 2004;19:620-8. 2. Sato T. J Arthroplasty 2007;22:560. 3. Ishii Y. J Arthroplasty 2011;26:255. 4. Clarke HD. Surgery of the knee. Vol. 1. New York: Churchill Livingstone, 2001:13. 5. Insall JN. Surgery of the knee. Vol. 2. New York: Churchill Livingstone, 2001:1553. 6. Stiehl JB. Am J Knee Surg 2000;13:13. Table 1. Details of the patients in both groups PCLR PCLS Knees/Patients 66/66 72/72 Gender (male/female) 10/56 12/60 Average age (years) (S.D.) 72 (6) 74 (6) BMI (SD) (kg/m) (S.D.) 26 (4) 27 (4) Mean posterior slope () (SD) 10 (2) 10 (2) Mean coronal alignment () (SD) 6 (3) 6 (3) HSS score 92 (2) 90 (4)