Long-term Results of Arthroscopic Rotator Cuff Repair: A Follow-up Study Comparing Single-Row Versus Double-Row Fixation Techniques

Background: Arthroscopic rotator cuff repair (RCR) with suture anchor-based fixation techniques has replaced former open and mini-open approaches. Nevertheless, long-term studies are scarce, and lack of knowledge exists about whether single-row (SR) or double-row (DR) methods are superior in clinical and anatomic results. Purpose: To analyze long-term results after arthroscopic RCR in patients with symptomatic rotator cuff tears and to compare functional and radiographic outcomes between SR and DR repair techniques at least 10 years after surgery. Study Design: Cohort study; Level of evidence, 3. Methods: Between 2005 and 2006, 40 patients with a symptomatic full-thickness rotator cuff tear (supraspinatus tendon tear with or without a tear of the infraspinatus tendon) underwent arthroscopic RCR with either an SR repair with a modified Mason-Allen suture-grasping technique (n = 20) or a DR repair with a suture bridge fixation technique (n = 20). All patients were enrolled in a long-term clinical evaluation, with the Constant score (CS) as the primary outcome measure. Furthermore, an ultrasound examination was performed to assess tendon integrity and conventional radiographs to evaluate secondary glenohumeral osteoarthritis. Results: A total of 27 patients, of whom 16 were treated with an SR repair and 11 with a DR repair, were followed up after a mean +/- SD period of 12 +/- 1 years (range, 11-14 years). Five patients underwent revision surgery on the affected shoulder during follow-up period, which led to 22 patients being included. The overall CS remained stable at final follow-up when compared with short-term follow-up (81 +/- 8 vs 83 +/- 19 points; P = .600). An increasing number of full-thickness retears were found: 6 of 22 (27%) at 2 years and 9 of 20 (45%) at 12 years after surgery. While repair failure negatively affected clinical results as shown by the CS (P < .05), no significant difference was found between the fixation techniques (P = .456). In general, progressive osteoarthritic changes were observed, with tendon integrity as a key determinant. Conclusion: Arthroscopic RCR with either an SR or a DR fixation technique provided good clinical long-term results. Repair failure was high, with negative effects on clinical results and the progression of secondary glenohumeral osteoarthritis. While DR repair slightly enhanced tendon integrity at long-term follow-up, no clinical superiority to SR repair was found.