Shift Angle and Medial Retraction of the Deep Layer on Magnetic Resonance Imaging Are Associated With Intraoperative Reparability in Arthroscopic Rotator Cuff Repair

Abstract

Purpose To determine whether magnetic resonance imaging‐based measurements of the deep layer, specifically the shift angle and medial retraction, can predict intraoperative reparability of large‐to‐massive rotator cuff tears. Methods This retrospective study included patients who underwent arthroscopic repair of large‐to‐massive rotator cuff tears between April 2015 and September 2019, with a minimum follow‐up of 12 months. The shift angle was defined as the anteroposterior extent of the deep‐layer tear on sagittal magnetic resonance imaging and medial retraction as the distance from the lateral edge of the deep‐layer stump to the cartilage margins. Inter‐ and intraobserver reliabilities and predictive accuracy for reparability were assessed. Results A total of 177 shoulders were analyzed, of which 149 were reparable and 28 were irreparable. The irreparable group showed larger shift angles (114.3° ± 14.5° vs 82.6° ± 24.0°, P < .001) and greater medial retraction (33.9 ± 5.2 vs 25.5 ± 6.1 mm, P < .001) than the reparable group. Receiver operating characteristic analysis showed good diagnostic performance for predicting irreparability for both the shift angle (area under the curve, 0.861; cutoff, 98.0°) and medial retraction (area under the curve, 0.870; cutoff, 29.8 mm), comparable to or better than superficial tear‐size measurements. Measurement reliability for both parameters was substantial to excellent (interobserver intraclass correlation coefficient = 0.862 and 0.769; intraobserver intraclass correlation coefficient = 0.896 and 0.889, respectively). Conclusions Magnetic resonance imaging‐based assessment of the deep layer using the shift angle and medial retraction showed good diagnostic performance and substantial reliability for predicting intraoperative reparability. Level of Evidence Level III, retrospective cohort study.