Abstract
Background
Accuracy of implantation is a recognized prognostic factor for the long-term survival of TKA. The purpose of this study was to analyze the accuracy of component orientation and post-operative alignment of the leg following CT-based navigation-assisted TKA and to compare these parameters with those of a conventional surgical technique.
Methods
We retrospectively compared the alignment of 130 total knee arthroplasties performed with a CT-based navigation system with that of 130 arthroplasties done with a conventional alignment guide system. The knee joints were evaluated using full-length weight-bearing antero-posterior and lateral radiographs.
Results
The mean hip-knee-ankle angle, the frontal femoral component angle and the frontal tibial component angle were 180.7° (normal angle: 180.0°), 88.8° (90.0°) and 90.6°(90.0°), respectively, for the navigation-assisted arthroplasties and 180.9°(180.0°), 89.8°(90.0°) and 89.3°(90.0°), respectively, for conventional arthroplasties. The mean lateral femoral component angle and the femoral tibial component angle were 0.99° and 89.9°, respectively, for the navigation group and 2.62° and 88.5°, respectively, for the conventional group. All pre-operative leg axes of 10 outliers (HKA<177 or HKA>183) in the navigation group were over 193°, while in the conventional group, 23 outliers’ data were scattered.
Conclusions
Our retrospective study with randomly assigned cases (consecutive patients in two separate hospitals) demonstrates significant improvements in component positioning with the CT-based navigation system compared to the conventional alignment guide system. Furthermore, we found that when analyzing cases within each group with pre-operative hip-knee-ankle angles lower and equal 192°, no outliers were found in the navigation group indicating a high level of alignment accuracy in this group. However, in cases with pre-operative hip-knee-ankle angles larger or equal 193°, outliers were found in both groups and no significant difference between the two groups was observed (p = 0.24). A detailed analysis of the outlier cases in the navigation group revealed that the femoral component was placed in the varus position. We thought that pre-operative underestimation of osteophytes of the medial femoral condyle might have led to a lateral shift of the femoral component during its intra-operative placement and was one of the contributing factors causing lower alignment accuracy.
