Time-dependent changes after latissimus dorsi transfer: tenodesis or tendon transfer?

BACKGROUND: Transfer of the latissimus dorsi tendon to the posterosuperior part of the rotator cuff is an option in active patients with massive rotator cuff tears to restore shoulder elevation and external rotation. However, it is unknown whether this treatment prevents progression of cuff tear arthropathy. QUESTIONS/PURPOSES: The purpose of this study was to determine whether the observed improvement in shoulder function in the early postoperative period with latissimus dorsi tendon transfer for irreparable rotator cuff tears will be permanent or will deteriorate in the midterm period (at 1-5 years after surgery).
METHODS: During a 6-year period, we performed 11 latissimus dorsi tendon transfers in 11 patients for patients with massive, irreparable, chronic tears of the posterosuperior part of the rotator cuff (defined as > 5 cm supraspinatus and infraspinatus tendon tears with Goutallier Grade 3 to 4 fatty infiltration on MRI), for patients who were younger than 65 years of age, and had high functional demands and intact subscapularis function. No patients were lost to followup; minimum followup was 12 months (median, 33 months; range, 12-62 months). The mean patient age was 55 years (median, 53 years; range, 47-65 years). Shoulder forward elevation, external rotation, and Constant-Murley and American Shoulder and Elbow Surgeons scores were assessed. Pain was assessed by a 0- to 10-point visual analog scale. Acromiohumeral distance and cuff tear arthropathy (staged according to the Hamada classification) were evaluated on radiographs.
RESULTS: Shoulder forward elevation, external rotation, Constant-Murley scores, and American Shoulder and Elbow Surgeons scores improved at 6 months. However, although shoulder motion values and Constant-Murley scores remained unchanged between the 6-month and latest evaluations, American Shoulder and Elbow Surgeons scores decreased in this period (median, 71; range, 33-88 versus median, 68; range, 33-85; p = 0.009). Visual analog scale scores improved between the preoperative and 6-month evaluations but then worsened (representing worse pain) between the 6-month and latest evaluations (median, 2; range, 0-5 versus median, 2; range, 1-6; p = 0.034), but scores at latest followup were still lower than preoperative values (median, 7; range, 4-8; p = 0.003). Although acromiohumeral distance values were increased at 6 months (median, 8 mm; range, 6-10 mm; p = 0.023), the values at latest followup (median, 8 mm; range, 5-10 mm) were no different from the preoperative ones (mean, 7 mm; range, 6-9 mm; p > 0.05). According to Hamada classification, all patients were Grade 1 both pre- and postoperatively, except one who was Grade 3 at latest followup.
CONCLUSIONS: The latissimus dorsi tendon transfer may improve shoulder function in irreparable massive rotator cuff tears. However, because the tenodesis effect loses its strength with time, progression of the arthropathy should be expected over time. Nevertheless, latissimus dorsi tendon transfer may help to delay the need for reverse shoulder arthroplasty for these patients. LEVEL OF EVIDENCE: Level IV, therapeutic study. See Instructions for Authors for a complete description of levels of evidence.