MRI evaluation of the knee with non-ferromagnetic external fixators: cadaveric knee model.

Abstract

The advent of MRI-compatible external fixation devices has made the use of MRI possible in patients who have been treated with external fixation. However, although there have been multiple studies determining the safety of MRI scans with external fixator devices, there are no studies determining the artifact effect these devices can have on the MRI image. The purpose of our study was to evaluate the effect of two popular brands (Stryker and Synthes) of MRI-compatible external fixators on the diagnostic capacity of a knee MRI. We hypothesize that (1) MRI images would have higher noise due to the presence of an external fixator and (2) images of high diagnostic capacity will be obtainable in the presence of each external fixator spanning the knee. Using seven cadaveric knees, a study was performed to analyze MRI images taken in the presence each external fixator. Scans taken with no external fixator present served as controls. Signal-to-noise ratios (SNRs) were measured at five anatomic structures. These structures were compared as a quantitative measure of image quality. A qualitative analysis was also performed using a five-point grading scale to assess the influence of metal artifact on the quality of the images. Each scan was graded by three blinded musculoskeletal radiologists focusing on six key anatomic structures. A reduction in SNR was identified on the external fixator group compare to the control groups at the patella tendon, MM and PCL. Qualitative scoring by three expert radiologists showed no difference in ability to identify the six key anatomic landmarks between the Stryker, Synthes and control images. Although the presence of external fixation devices does increase the noise artifact in MRI scans, patients treated with these external fixators can undergo MRI of local structures with high likelihood of obtaining diagnostic quality images.