RESUMO
INTRODUCTION: Component position and overall limb alignment following total knee arthroplasty (TKA) have been shown to influence implant survivorships and clinical outcomes. While most surgeons utilize standard x-ray imaging for preoperative joint assessments, computer tomography scans (CT), coupled with automated digital analyses have been shown to provide additional surgical and clinical benefits. However, to date, a postoperative CT measurement protocol has not been reported for robotic-arm assisted TKA (RATKA). Therefore, the purpose of this paper was to assess the validity of a novel, vector-based CT alignment measurement protocol. Specifically, we compared: 1) final versus planned component alignment and placement; 2) inter-observer reliability; and 3) intra-observer reliability. MATERIALS AND METHODS: The CT-based technique utilized mathematical models to calculate prosthetic alignments from anatomical landmarks. To assess the models, 30 CT scans from multiple centers were collected on RATKA patients at six weeks postoperatively and analyzed using the proposed technique. Results were compared to the surgeons' preoperative plans for accuracy. Analyses were performed on the same protocol to determine inter-observer reliability. These analyses were repeated 30 days later to assess for intra-observer variability. RESULTS: The mean measurement errors compared between final versus planned component positions and alignments were: 0.79±1.48o varus in overall limb alignment (p=0.004); 0.34±1.20o varus (p=0.121); and 0.35±1.15o varus (p=0.17) for femoral and tibial varus/valgus alignment; 0.71±1.77o flexion (p=0.18) and 0.38±1.88o posterior (p=0.41) for femoral flexion and tibial slope. There was strong reproducibility between observers. Correlation analyses showed low variabilities, with slopes between 0.8 to 1.0 and all R>0.8. CONCLUSION: As robotic technologies become widely available in orthopaedic surgery, it is critical to have tools, such as CT protocols, which can quantitatively verify operative decisions concerning limb alignment and component placement. This study described a novel, vector-based, CT alignment measurement protocol for RATKA which has not previously been defined. The method demonstrated excellent accuracy to plan and low intra- and inter-observer variability. This is a valuable analysis tool for RATKA studies where component accuracy is assessed using postoperative CT images.
