Abstract
Introduction
Instability, loosening, and patellofemoral pain belong to the main causes for revision of total knee arthroplasty (TKA). Currently, the diagnostic pathway requires various diagnostic techniques such as x-rays, CT or SPECT-CT to reveal the original cause for the failed knee prosthesis, but increase radiation exposure and fail to show soft-tissue structures around TKA. There is a growing demand for a diagnostic tool that is able to simultaneously visualize soft tissue structures, bone, and TKA without radiation exposure. MRI is capable of visualising all the structures in the knee although it is still disturbed by susceptibility artefacts caused by the metal implant. Low-field MRI (0.25T) results in less metal artefacts and offers the ability to visualize the knee in weight-bearing condition. Therefore, the aim of this study is to investigate the possibilities of low field MRI to image, the patellofemoral joint and the prosthesis to evaluate the knee joint in patients with and without complaints after TKA.
Method
Ten patients, eight satisfied and two unsatisfied with their primary TKA, (NexGen posterior stabilized, BiometZimmer) were included. The patients were scanned in sagittal, coronal, and transversal direction on a low field MRI scanner (G-scan Brio, 0.25T, Esaote SpA, Italy) in weight-bearing and non-weight-bearing conditions with T1, T2 and PD-weighted metal artefact reducing sequences (TE/TR 12–72/1160–7060, slice thickness 4.0mm, FOV 260×260×120m3, matrix size 224×216). Scans were analysed by two observers for:
-
- Patellofemoral joint: Caton-Descamps index and Tibial Tuberosity-Trochlear Groove (TT-TG) distance.
-
- Prosthesis malalignment: femoral component rotation using the posterior condylar angle (PCA) and tibial rotation using the Berger angle.
Significance of differences in parameters between weight-bearing and non-weight-bearing were calculated with the Wilcoxon rank test. To assess the reliability the inter and intra observer reliability was calculated with a two-way random effects model intra class correlation coefficient (ICC). The two unsatisfied patients underwent revision arthroplasty and intra-operative findings were compared with MRI findings.
Results
In the satisfied group, a significant difference was found between TT-TG distance in non-weight-bearing and weight-bearing condition (p=0.018), with a good interrater reliability ICC=0.89. Furthermore, differences between weight-bearing and non-weight-bearing were found for the CD ratio, however, not significant (p=0.093), with a good interrater reliability ICC=0.89. The Berger angle could be measured with an excellent interrater reliability (ICC=0.94). The PCA was hard to assess with a poor interrater reliability (ICC=0.48). For one unsatisfied patient a deviation was found for tibial component rotation, according to the perioperative findings as, ‘malposition of the tibial component’. For the other unsatisfied patient revision surgery was performed due to aseptic loosening in which the MRI showed a notable amount of synovitis.
Conclusion
It is possible to image the patellofemoral joint and knee prosthesis with low field MRI. Patellofemoral measurements and tibial component rotation measurements can reliably be performed. For the two patients with complaints MRI findings were consistent with intra-operative findings. Further research should focus on a larger group of patients with complaints after TKA to verify the diagnostic capacity of low field MRI for peri-prosthetic knee problems.
For any figures or tables, please contact authors directly.
