Latissimus dorsi tendon transfer for irreparable posterosuperior rotator cuff tears. Factors affecting outcome

BACKGROUND: The purpose of this study was to evaluate the clinical results of latissimus dorsi tendon transfer in patients with an irreparable posterosuperior rotator cuff tear to help determine which patient and anatomic factors affect clinical outcome. METHODS: Fourteen patients with a latissimus dorsi tendon transfer were clinically evaluated with use of the PENN (University of Pennsylvania) shoulder score as well as with quantitative measurement of isometric muscle strength and the range of motion of both shoulders at a minimum of twenty-four months postoperatively. The anatomic results were evaluated with postoperative magnetic resonance imaging and electromyography. RESULTS: Nine patients were satisfied with the outcome, had significant clinical improvement, and reported that they would have the operation again under similar circumstances. The other five patients were dissatisfied with the result and had significantly worse PENN scores, active elevation, and objective measures of strength. Eight of the nine patients with a good clinical result were male, and four of the five with a poor result were female. Patients with a good clinical result had had significantly better preoperative function in active forward flexion and active external rotation compared with the patients with a poor result. The magnetic resonance imaging demonstrated healing of the tendon to the greater tuberosity in twelve patients and equivocal healing in two. There was no significant atrophy of any of the transferred muscles. Electromyography demonstrated clear activity in the transferred latissimus muscle during humeral adduction in all fourteen patients, some electrical activity with active forward elevation in only one patient, and some electrical activity with active external rotation in six of the nine patients with a good clinical result. None of the patients with a poor clinical result demonstrated electrical activity of the transferred muscle with active forward flexion or external rotation. CONCLUSIONS: Synchronous in-phase contraction of the transferred latissimus dorsi is a variable finding following the surgical treatment of irreparable posterosuperior rotator cuff tears, but when it is present it is associated with a better clinical result. Preoperative shoulder function and general strength influence the clinical result. Female patients with poor shoulder function and generalized muscle weakness prior to surgery have a greater likelihood of having a poor clinical result.