Abstract
Objective
Computer-assisted minimally invasive total knee arthroplasty (CAMI-TKA) has gained increasing interest from orthopaedic surgeons due to its advantages in improving accuracy of component placement combined with benefits in postoperative recovery due to a smaller incision. However, long-term clinical and radiographic outcomes are lacking. The purpose of the present study is to compare the long-term radiographic features and functional outcomes between patients who underwent CAMI-TKA and those who underwent conventional TKA.
Methods
One hundred and eight patients who were randomized to undergo CAMI-TKA or conventional TKA during 2004 and 2005 were contacted by phone for a prospective follow-up review. Patients who have passed away or declined to participate in the study were excluded. Patients were asked to return to the hospital for clinical and functional assessments, long-leg and knee roentgenograms. Baseline characteristics were compared to account for potential confounders and multivariate statistical analysis applied to account for any differences in baseline characteristics.
Results
As shown in Figure 1, a total of 101 patients (93.52%) were contacted, and 69 patients (63.89%) returned to hospital for assessments and investigations. By the time of this study, two patients from the Conventional TKA group had undergone revision TKA, one due to infection and one due to aseptic loosening. The average follow up time was 9.07 years (8.51–9.61 years). Subsequent comparison was carried out between 37 patients from the conventional TKA group and 30 from the CAMI-TKA group. Both groups had similar pre-operative demographics, clinical and functional assessments except for the Function Score component of the Knee Society Score (Conventional=50 vs. CAMI=55, p=0.049). At follow-up, the Short Form-36 and Knee Society Scores were comparable between the two groups. However, patients from CAMI group reported a significantly higher Oxford Knee Score compared to those from conventional group (p=0.013). No significant intergroup differences were found in mechanical knee alignment and component placement angle in the coronal views. In the sagittal views, the femoral components demonstrated a more extensional configuration in the conventional group, in contrary to a more flexional configuration in the CAMI group (1.0° extension vs. 1.5° flexion, p<0.001). There also existed a significant difference in sagittal tibial component angles, where the conventional group had a steeper posterior slope compared to the CAMI group (5.1° vs. 2.5°, p=0.002). Four knees from CAMI-TKA group and 1 knee from Conventional TKA group were found to have non-progressive radiolucencies between the components and bone cuts, without statistical or clinical significance. No other patients demonstrated any migrating or shifting of the prosthesis that could be construed as possible failure in either group.
Conclusions
Computer-assisted minimally invasive total knee arthroplasty provided similar clinical, functional, and radiographic outcomes compared with conventional total knee arthroplasty after an average of 9 years follow-up. This technique can be employed to exploit its short-term advantages without compromising long-term clinical and radiographic outcomes.
