Abstract
Introduction
Roentgen stereophotogrammetric analysis (RSA) is currently the gold standard to measure early prosthetic migration which can predict aseptic loosening. However, RSA has some limitations such as the need for perioperative placed markers and exposure to X-radiation during follow up. Therefore, this study evaluates if low field MRI could be an alternative for RSA. Low field MRI was chosen because it is less hampered by metal artifacts of the prosthesis than high field MRI.
Methods
3D models of both the tibial component of a total knee prosthesis (Genesis II, Smith and Nephew) and the porcine tibia were made. The tibial component was implanted in the tibial bone. Consequently, 17 acquisitions with the low field MRI scanner (Esaote G-scan 0.25T) in transverse direction with a 2D PD weighted metal artifact reducing sequence PD-XMAR (TE/TR 10/1020ms, slice thickness 3mm, FOV 180×180×120 mm³, matrix size 224×224) were made. The first five acquisitions were made without repositioning the cadaver, the second twelve after slightly repositioning the cadaver within limits that are expected to be encountered in a clinical setting. Hence, in these 17 acquisitions no prosthetic-bone motions were induced.
The scans were segmented and registered with Mimics. Virtual translation and rotation of the prosthesis with respect to the bone between two scans were calculated using a Procrustes algorithm. The first five scans without repositioning were used to calculate the measurement error, the following twelve to calculate the precision of low field MRI to measure prosthetic migration. Results were expressed as the maximum total point motion, mean error and 95% CI and expressed in boxplots.
Results
The error of the method to measure the prosthetic position without repositioning has a mean translation between 0.09 and 0.22mm with a 95%CI between 0.30 and 0.46mm. The mean rotation was between 0.02° and 0.11° with a 95%CI between 0.18° and 0.32° with a MTPM of 0.45mm.
The precision of low field MRI to measure migration with repositioning has a mean translation between 0.02 and 0.12mm with a 95%CI between 1.16mm and 1.86mm. The mean rotation was between 0.01° and 0.15° with a 95%CI between 1.78° and 3.26° with a MTPM of 2.35mm. The overall registration error was largest in the distal-proximal direction.
Discussion
At the moment the low field MRI technique is not as accurate as this gold standard RSA. The accuracy of RSA varies between 0.05 and 0.5 mm for translation and 0.15 ° to 1.15 ° for rotation (95% confidence intervals). However, results are comparable with markerless RSA studies. The largest measurement error was found in the distal-proximal direction, which can be explained by the through-plane resolution of 3 mm, which is larger than the in-plane resolution of 0.8×0.8 mm². Future research should focus on improving resolution in the distal-proximal direction which would improve the precision. Moreover, an actual migration study should be performed to proof the true value of this low field MRI base markerless and X-radiation free alternative to measure prosthetic migration.
