Proton Density Fat-Fraction of Rotator Cuff Muscles Is Associated With Isometric Strength 10 Years After Rotator Cuff Repair A Quantitative Magnetic Resonance Imaging Study of the Shoulder

Background: Quantitative muscle fat-fraction magnetic resonance (MR) imaging techniques correlate with semiquantitative Goutallier scores with failure after rotator cuff (RC) repair. Purpose: To investigate the relationship of proton density fat fraction (PDFF) of the RC muscles with semiquantitative MR scores, cartilage T2 relaxation times, and clinical isometric strength measurements in patients 10 years after unilateral RC repair. Study Design: Cross-sectional study; Level of evidence, 3. Methods: Bilateral shoulder MR imaging was performed in 13 patients (11 male, 2 female; age, 72 +/- 8 years) 10.9 +/- 0.4 years after unilateral autologous periosteal flap augmented RC repair (total shoulders assessed, N = 26). Goutallier classification, muscle atrophy, RC tendon integrity, and cartilage defects were determined based on morphological MR sequences. A paracoronal 2D multi-slice multi-echo sequence was used for quantitative cartilage T2 mapping. A chemical shift-encoding-based water-fat separation technique (based on a 6-echo 3D spoiled gradient echo sequence) was used for quantification of the PDFF of RC muscles. Isometric shoulder abduction strength was measured clinically. Mean and SD, Pearson correlation, and partial Spearman correlation were calculated. Results: There were 6 RC full-thickness retears in ipsilateral shoulders and 6 RC full-thickness tears in contralateral shoulders. Isometric shoulder abduction strength was not significantly different between ipsilateral and contralateral shoulders (50 +/- 24 N vs 54 6 24 N; P = .159). The mean PDFF of RC muscles was 11.7% +/- 10.4% (ipsilateral, 14.2% +/- 8.5%; contralateral, 9.2% +/- 7.8%; P = .002). High supraspinatus PDFF correlated significantly with higher Goutallier scores (R = 0.75, P < .001) and with lower isometric muscle strength (R = -0.49, P = .011). This correlation remained significant after adjustment for muscle area measurements and tendon rupture (R = -0.41, P = .048). More severe cartilage defects at the humerus were significantly associated with higher supraspinatus PDFF (R = 0.44; P = .023). Cartilage T2 values did not correlate with muscle PDFF (P > .05). Conclusion: MR imaging-derived RC muscle PDFF is associated with isometric strength independent of muscle atrophy and tendon rupture in shoulders with early and advanced degenerative changes. It therefore provides complementary, clinically relevant information in tracking RC muscle composition on a quantitative level.