Comparison of Routine Knee Magnetic Resonance Imaging at 3 T and 7 T.

OBJECTIVE

The aim of this study was to compare quantitative and semiquantitative parameters (signal-to-noise ratio [SNR] and diagnostic confidence) from a standard knee magnetic resonance imaging (MRI) examination with comparable sequence protocols and acquisition times at 3 T and at 7 T.

MATERIALS AND

METHODS

Forty patients experiencing knee pain of unknown etiology underwent comparable MR protocols with standard turbo-spin echo and short tau inversion recovery sequences of the knee joint (5 sequences) at 3 T and 7 T. For quantitative analysis, SNR was determined using these 5 sequences and 3 additional morphological sequences. For a semiquantitative assessment of diagnostic confidence, a diagnostic confidence score (DCS) was assigned, using a 10-point scale. Two experienced radiologists who specialized in musculoskeletal imaging and who were blinded to the field-strength independently assessed 22 potential pathological findings, in total, in 4 anatomically defined areas in the knee joint and rated their diagnostic confidence.

RESULTS

In quantitative analysis, all sequences provided higher voxel-volume-adjusted SNR values at 7 T compared with that at 3 T. In semiquantitative analysis, summed DCS values for potential pathological findings in each of the 4 anatomically defined areas were higher at 7 T compared with that at 3 T. There was a statistically significant improvement in the DCS for both readers at 7 T for the diagnosis and exclusion of focal or diffuse grade I or II cartilage defects in the patellar cartilage. For 8 potential pathological findings, a statistically significant difference between the 2 field-strengths could be observed for 1 reader only. For the residual 13 potential pathological findings, there was no statistically significant difference observed. The percentage of concordant ratings was 84.6% at 3 T and 85.4% at 7 T.

CONCLUSIONS

Ultra-high-field MRI at 7 T improved the overall diagnostic confidence in routine MRI of the knee joint compared with that at 3 T. This is especially true for small joint structures and subtle lesions. Higher spatial resolution was identified as the main reason for this improvement.