RAPTOR: The Innovation for Making Long Leg Standing Radiography for Total Knee Arthroplasty from Conventional Radiography

Abstract

Purpose: To evaluate the reliability and validity of femoral anatomical-mechanical angle (fAMA), hip knee ankle angle (HKA), and overlap of long leg standing radiography (LLSR) obtained using a Rapid Orthoroentgenography Making Machine (RAPTOR) compared with a standard X-ray generator. Methods: This observational study was conducted between July 2021 and August 2021, including patients diagnosed with primary knee osteoarthritis that underwent preoperative LLSR for total knee replacement. Three orthopedic surgeons blindly evaluated LLSR (fAMA, HKA, overlap of the femoral shaft) twice within one-month using the Visio program. Intra- and interobserver reliability and validity were analyzed. Results: Three evaluators assessed 30 LLSRs. The intraobserver agreement levels were -0.951–1.062° for fAMA, -10.338–11.076° for HKA, and -0.418–0.418 mm for overlap of RAPTOR, while for the standard X-ray generator the agreement levels were -1.359–1.114° for fAMA, 11.844–12.467° for HKA, and 0 mm for overlap. The intraclass correlation was 0.55–0.99 for all RAPTOR measurements and 0.56–0.99 for standard X-ray generator. The interobserver’s levels of agreement were -1.441–1.175° for fAMA, -7.453–7.475° for HKA, and -0.681–0.637 mm for overlap of RAPTOR, whereas those of the standard X-ray generator were -1.149–1.424° for fAMA, -4.789–6.171° for HKA, and 0 mm for overlap. The intraclass correlation was 0.69–0.97 for all RAPTOR measurements and 0.71–0.95 for the fAMA and HKA standard X-ray generator measurements. The mean and 95% limits of agreement of the comparison between RAPTOR and standard X-ray generator were -0.131° (-1.187, 0.925) for fAMA, -0.126° (-4.724, 4.471) for HKA, and 0.363 (-) mm for overlap. Only overlap was significantly different between the two methods (p=0.0243). Intraclass correlations between the two radiographic methods were 0.75 (0.63, 0.88) for fAMA and 0.93 (0.89, 0.97) for HKA. Conclusions: Estimation of fAMA, HKA, and overlap had moderate to excellent reliability and inter- and intra-rater reliabilities in both RAPTOR and standard X-ray generator. Only overlap was different between the two methods.