Abstract
Introduction
Total knee arthroplasty (TKA) reliably improves pain and function in patients with knee osteoarthritis (OA), though a substantial percentage of patients remain unsatisfied. Reasons include the presence of complications, persistent pain, and unmet expectations. The aim of this study was to determine whether the sequential addition of accelerometer-based navigation of the distal femoral cut and sensor-assisted soft tissue balancing changed complication rates, radiographic alignment, or patient-reported outcomes (PROs) compared to TKA performed with conventional instrumentation.
Methods
This retrospective cohort study included 371 TKAs in 319 patients. All surgeries were performed by a single surgeon in sequential fashion using a measured resection technique with a goal of mechanical alignment. The historical control group, utilizing intramedullary guides for distal femoral resection and surgeon-guided soft tissue balancing, was compared to group 1 (accelerometer-based navigation for distal femoral resection, surgeon-guided balancing) and group 2 (navigated femoral resection, sensor-guided balancing). Primary outcome measures were PROMIS scores including physical function computerized adaptive test (PF CAT), and the Global 10 health assessment (including physical, mental, and pain scores), and Knee Injury Osteoarthritis and Outcome Score (KOOS), measured preoperatively and at 6 weeks and 12 months postoperatively. Radiographic measurements included component position and overall mechanical alignment of the limb and were made at 6 weeks by a single examiner from hip to ankle standing films. Charts were reviewed for pre- and postoperative ROM at 6 weeks, polyethylene insert morphology, and postoperative hematocrit change. Complications were recorded, including manipulation under anesthesia and reoperation. Our study was powered to detect a difference of 1 standard deviation in PF CAT score with 100 patients. Statistical analysis was performed by a statistician including t-tests, multivariate regression, and time series plot analyses.
Results
There were 194 patients in the control group, 103 in group 1, and 74 in group 2. There was no difference in baseline patient demographics. Patients in group 2 had higher baseline mental health subscores than control and group 1 patients (53.2 vs 50.2 vs 50.2, p=0.04). There were no differences in 6-week and one-year postop PF CAT, physical or mental subscores, pain scores, or KOOS scores (all p>0.05). There were 8 total complications in the control group (4.1%), 4 in group 1 (3.8%), and 1 in group 2 (1.4%) (p>0.4). The postoperative mechanical axis of the limb was within 3 degrees of neutral in 71.6% of control patients, 74.8% in group 1, and 85.1% in group 2 (p=0.1). There was no difference in femoral component coronal alignment between groups (p=0.91), though controls had a small but significantly higher degree of flexion in the sagittal plane (6.5 degrees) than groups 1 and 2 (5.4 degrees in both, p=0.003). There was no difference in postoperative ROM or blood loss.
Conclusions
The sequential addition of imageless navigation of the distal femoral cut and sensor-guided ligament balancing did not confer any benefit to short term PROs, radiographic outcomes, or complication rates over conventional techniques. While overall mechanical alignment of the limb was improved in groups 1 and 2 compared to controls, this did not reach statistical significance. The additive costs of navigation and soft-tissue balancing technologies may not be justified.
For figures, tables, or references, please contact authors directly.
