Posterosuperior and Anterior Rotator Cuff Repairs Yield Comparable 1‐Year Constant Scores in Routine Clinical Practice: A Propensity Score‐Matched Study From LaTour Group

Purpose To compare clinical outcomes and ultrasound‐assessed healing rates between posterosuperior and anterior rotator cuff repairs using propensity score matching. Methods Consecutive patients undergoing rotator cuff repair between 2015 and 2021 by a single fellowship‐trained shoulder surgeon were included if they had posterosuperior or anterior rotator cuff tears, Goutallier fatty infiltration grade 0 to 2, and a follow‐up of 1 year. Revision surgery, anterosuperior tears, and partial repairs were exclusion criteria. Propensity score matching (1:3) was performed on age, sex, traumatic origin, tendon retraction, and preoperative internal rotation and Constant scores. The primary outcome was the 1‐year Constant score. Secondary outcomes included range of motion, visual analog scale, American Shoulder and Elbow Surgeons, Single Assessment Numeric Evaluation, and ultrasound‐assessed healing (Sugaya classification) at 6 months. Results Of the 1045 screened patients, 639 met the inclusion criteria. After propensity score matching, 53 patients with anterior cuff tears were matched to 146 patients with posterosuperior tears, with excellent covariate balance between groups. Postoperative patient‐reported outcome measures (Constant, American Shoulder and Elbow Surgeons, Single Assessment Numeric Evaluation, and pain) and range of motion at 1 year were similar between groups, and ultrasound‐assessed healing at 6 months did not differ. The proportions of patients achieving or exceeding the minimal clinically important difference were also comparable between groups for all outcome measures, including the Constant score (76% vs 81%), American Shoulder and Elbow Surgeons score (69% vs 66%), pain visual analog scale (92% vs 96%), and Single Assessment Numeric Evaluation score (77% vs 85%) ( P = .569, P = .823, P = .525, and P = .291, respectively). In the adjusted mixed‐effects model, tear location was not statistically associated with the postoperative Constant score ( β = 1.7, 95% confidence interval −3.0 to 6.3; P = .485). Conclusions In this propensity score‐matched cohort, anterior and posterosuperior arthroscopic rotator cuff repairs in patients with limited fatty infiltration yielded comparable 1‐year Constant scores in a real‐world setting. Level of Evidence Level III, retrospective comparative study.