Quantitative assessment of supraspinatus fatty infiltration on a novel MRI section: comparison with conventional methods

Accurate quantification of fatty infiltration (FI) after rotator-cuff tear (RCT) is critical for therapeutic planning. The conventional Goutallier classification, based on the oblique-sagittal “Y-view,” is subjective, shows only moderate inter-observer agreement, and frequently underestimates the degree of infiltration when the tendon is severely retracted. Recently, evaluation at a more medial level—the suprascapular notch (“R-view”)—has been proposed because it better visualizes the retracted muscle belly, yet it has not been combined with quantitative analysis. To determine (1) the agreement between quantitative FI (Fat%) measured on the R-view and the traditional Goutallier grade, and (2) the diagnostic value of this quantitative approach in patients with Patte stage 3 retraction. Patients who underwent arthroscopic rotator-cuff repair (ARCR) between 2022 and 2024 were retrospectively enrolled and stratified by Patte classification. On both the Y-view and the R-view, we measured supraspinatus cross-sectional area (CSA), Fat%, and Goutallier grade. Intraclass correlation coefficients (ICC) were calculated for reliability, receiver operating characteristic (ROC) curves were constructed for diagnostic performance, and the two views/methods were compared. A total of 192 patients were included. CSA was significantly larger on the R-view than on the Y-view (P < 0.05), whereas both Fat% and Goutallier grade were significantly lower (P < 0.05). Fat% and Goutallier grade were strongly correlated on both views (R = 0.781 vs. 0.826, P < 0.001). In Patte stage-3 patients, Fat% derived from the R-view achieved the highest diagnostic performance (Area Under the Curve(AUC), AUC = 0.73); the optimal cutoff was 56%, yielding 86% specificity and a Youden index(YI) of 0.41. The R-view more accurately depicts the true status of the retracted supraspinatus. Combining this imaging plane with quantitative Fat% measurement significantly improves objectivity and reliability in the assessment of FI, especially in cases of severe tendon retraction, and offers surgeons a superior pre-operative imaging strategy.