The trusted source for
healthcare information and
Synopsis: A retrospective, nonrandomized study of full-thickness rotator cuff tears was undertaken to evaluate the efficacy of nonoperative management. Modest improvements in shoulder function and comfort were noted when treating patients without surgery; furthermore, the overall responsiveness of shoulder function to nonoperative intervention was poor.
Source: Goldberg BA, et al. Outcome of nonoperative management of full-thickness rotator cuff tears. Clin Orthop. 2001;382:99-107.
This study documents the functional outcome associated with nonoperative treatment of full-thickness rotator cuff tears treated by the shoulder service at the University of Washington. This study used contemporary measures of shoulder function (The Simple Shoulder Test) and health status (SF-36), giving Goldberg and colleagues the opportunity to evaluate both function and health in a modern outcome study. Forty-six patients were selected consecutively based on the following inclusion criteria: 1) full-thickness cuff tear diagnosed on a secondary radiographic study (MR, arthrogram, or ultrasound); 2) absence of a Worker’s Compensation claim; 3) minimum 1-year follow-up; and 4) election of nonoperative management by the patient. Treatment involved stretching and strengthening exercises for the remaining intact rotator cuff muscles. At an average follow-up of 2.5 years, 59% of patients experienced improvement, 30% noted worsening, and 11% remained unchanged. The ability to sleep on the affected side and the ability to place the hand behind the head were significantly improved by nonoperative management. Patients who improved were more likely to have a rotator cuff tear of the dominant extremity, had a lower average initial shoulder function score (on the Simple Shoulder Test), and had more difficulty tucking their shirt behind their back. Interestingly, several general health parameters significantly worsened during the follow-up, suggesting that the patients were of declining health.
Controversy exists over clinical decision-making in the management of full-thickness rotator cuff tears. Goldberg et al note only "modest improvement in shoulder function and comfort when patients were treated without surgery; the overall responsiveness of shoulder function to nonoperative intervention was poor." In addition to the nonuniform documentation of the cuff tear, the study is retrospective and nonrandomized whereby patients elected nonoperative care. Although a randomized, prospective design is ideal, in the present study, one would expect a bias for the best possible results with nonoperative treatment as the patients chose this type of management. In contrast, the improvement with rehabilitation was modest. In my opinion, it is unclear whether the results would have been better than a control population. The finding of declining health status is interesting and may indicate some concern by the patients of the safety of undergoing a general anesthetic, in effect prompting the selection of nonoperative care.
I found useful the identifying features of the patient who was more likely to improve with rehabilitation: those with difficulty tucking in their shirts, dominant extremity involvement, and those with poorer initial functional scores. Goldberg et al note that difficulty with tucking in one’s shirt is related to internal rotation deficits or posterior capsular tightness and may be a separate phenomena from a cuff tear, thus making the patient more responsive to a rehabilitation program using stretching techniques. Relating this study to my clinical experience underlines the following points: those patients with initial stiffness and poorer function are more commonly improved with nonoperative management. A randomized trial is still needed to address the comparison of operative and nonoperative treatment options, and exercises vs. no treatment at all.