Comparison of CT-based migration analysis against radiostereometric analysis for assessing two-year migration in an uncemented medial pivot tibial knee implant

Aims: Radiostereometric analysis (RSA) is the gold standard for migration analysis of orthopaedic implants, but CT-based RSA has shown comparable results in an increasing number of joints. In this study, we compared the method in an uncemented medial pivot tibial implant. Our primary outcome measure was maximum total point motion (MTPM). Secondary measures were segmental and feature point migration parameters, range of motion (ROM), Forgotten Joint Score (FJS-12), Knee Injury and Osteoarthritis Score (KOOS), effective radiation doses, and time usage. Methods: A total of 31 patients were prospectively included to receive an uncemented medial pivot knee implant using mechanical alignment (MA) with conventional instruments. Migrations were measured up to 24 months with marker-based RSA, model-based RSA, and CT-based RSA. Clinical data were measured using a goniometer and patient-reported outcome measures ((PROMs); FJS-12 and KOOS). Results: Migrations at 24 months for MTPM showed a mean of 1.09 mm (95% CI 0.78 to 1.40) for marker-based RSA, 1.17 mm (95% CI 0.85 to 1.49) for model-based RSA, and 0.74 mm (95% CI 0.45 to 1.04) for CT-based RSA. MTPM for CT-based RSA was lower than both RSA methods (p < 0.001). Differences increased between the methods at all timepoints, but migrations stabilized for all methods. Time usage (min:sec) for performing RSA radiographs was 17:02 and 4:10 for CT. FJS-12 score increased from 12.3 at baseline to 70.4 at 24 months; and KOOS increased for all sub-scales. ROM increased by 12.4° (95% CI 5.6 to 19.3). Conclusion: For clinical importance, the uncemented GMK Sphere shows a stable migration pattern using all methods and can therefore be considered safe. CT-based RSA shows lower migration values of an uncemented medial pivot tibial implant up to 24 months compared with conventional RSA.