Diagnostic performance of DIXON sequences on low-field scanner for the evaluation of knee joint pathology.

Background and aim Recently, there has been a growing interest in the use of Dixon sequence for knee MRI in order to save time spent on the scanner, and improving diagnostic utility. Our purpose was to compare the diagnostic performance of Dixon sequence on low-field MRI with the proton-density sequence on high-field MRI. Methods This prospective study included 40 patients who underwent 0.25T knee MRI, using the routine protocol with the addition of a sagittal 4-point Dixon sequence (SPED), and an additional sequence on 1.5T scanner, consisting in a fat-suppressed proton-density fast-spin-echo (FS PD-FSE). Two radiologists independently examined the images, evaluating the anatomic identification score and diagnostic performances of the two sequences. Interreader agreement was evaluated using an intraclass correlation coefficient (ICC). Results Final population counted 34 patients (36 knee MR images) with a mean age of 52.9 years (range, 18-75 years). Interreader agreement was very high except for cartilage injuries at medial femoral condyle and medial tibial plateau (ICC SPED: 0.757, ICC FS PD-FSE: 0.746), even if not statistically significant. There were no significant differences in mean signal-to-noise ratio (SNR), artifacts presence and diagnostic confidence between SPED and PD-FS sequence. Conclusions Dixon sequences on low-field scanner have a comparable diagnostic accuracy to PD-FS sequence obtained on a high field scanner for knee MR imaging. (www.actabiomedica.it).