|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Twenty patients who had a massive tear (greater than five centimeters) of the rotator cuff that was not amenable to direct tendon-to-bone or tendon-to-tendon repair had reconstruction consisting of transfer of the subscapularis tendon in conjunction with subacromial decompression. At a mean of thirty months (range, twenty-three to seventy months) after the operation, seventeen of the patients were satisfied with the result. Nineteen patients reported a decrease in pain compared with preoperatively. However, nine patients had weakness and discomfort with prolonged or repetitive overhead activities, and two patients had lost active elevation of the shoulder despite substantial relief of pain. Subscapularis transfer is a useful adjunct in the operative treatment of massive tears of the rotator cuff; it facilitates the closure of larger defects that are not amenable to simpler, more traditional reconstructive techniques. However, because there is a risk of the procedure adversely affecting active elevation of the shoulder, it should be used with caution in patients who have full functional elevation preoperatively.
Since Neer's report25 in 1972, a number of studies have supported the value of operative repair of chronic, full-thickness tears of the rotator cuff in combination with subacromial decompression for relief of pain and improvement of function of the shoulder4,8,10,14,17,19,20,31. It has been suggested that the result of operative repair may depend in part on the size of the defect of the rotator cuff5,13,14,16,20,31,35. For larger defects that cannot be repaired with primary tendon-to-bone or tendon-to-tendon suture, a number of alternative operative procedures have been advocated. These have included muscle advancement12,18,24, lateral transposition of the supraspinatus (a so-called interval slide)4, local tendon transfer15,27, and subacromial decompression with débridement of the rotator cuff37. Other operative approaches incorporating synthetic materials34, devascularized fascial grafts3, and allografts32 have largely been abandoned4. Transposition of the proximal half to two-thirds of the subscapularis tendon has been proposed by a number of authors as an alternative method for the reconstruction of larger tears of the rotator cuff, especially those associated with widespread scarring in the surrounding tissue, retraction of the supraspinatus, or loss of available tissue for repair2,4,9,11,26,27,30,31. The purpose of the current study was to review retrospectively our experience with transfer of the subscapularis tendon for the reconstruction of chronic massive tears of the rotator cuff that were not amenable to direct primary repair.
From 1986 through 1991, 190 patients were managed operatively for a tear of the rotator cuff by the senior one of us (S. E. K.). In 161 patients, the operative repair was performed with a direct tendon-to-bone or tendon-to-tendon suture technique. Twenty-nine patients were found to have a massive tear (greater than five centimeters) of the rotator cuff that was not amenable to primary repair. Six of these patients were managed with subacromial decompression and débridement of the rotator cuff alone, as extensive local scarring with loss of the tissue substance or degenerative changes in the subscapularis tendon, or both, precluded the use of this tendon for reconstruction. In the remaining twenty-three patients, the proximal half to two-thirds of the subscapularis tendon was transferred from its site of insertion on the lesser tuberosity of the humerus for use in closure of the defect of the rotator cuff. Twenty of these patients had a sufficient duration of follow-up, and were available, for this study. Fifteen patients were men and five were women; the mean age at the time of the operation was sixty-one years (range, thirty-six to seventy-two years). The right shoulder was involved in thirteen patients and the left, in seven; the dominant limb was involved in all but six patients. Four patients had had a repair of the rotator cuff of the contralateral shoulder, and two had had a subscapularis transfer for reconstruction of a massive defect of the rotator cuff of the contralateral shoulder. These two shoulders were excluded from the study because of the limited duration of follow-up. Pain in the shoulder was the most common symptom at the time of presentation and was the primary indication for the operation in all but two patients. In those two patients, one of whom reported having no preoperative discomfort at all, loss of function was the major symptom. Sixteen patients had pain at night. Eleven patients described the pain as continuous and necessitating frequent use of narcotic analgesics; six also described the pain as continuous but used analgesics only occasionally. Two patients had little or no pain at rest, but pain with activity necessitated frequent use of non-prescription analgesics. Preoperatively, two patients reported that they were unable to use the upper extremity at all, five could perform only very light activities, and seven thought that they could perform only light household tasks and activities of daily living. Five patients stated that they were able to perform most household tasks and to drive an automobile without difficulty, and one patient noted only minor restriction in the ability to use the upper extremity above shoulder height. Fourteen patients had received one to five (mean, three) injections of corticosteroids before the index operation. Twelve patients attributed the onset of the symptoms to a traumatic event: six patients described a fall (one had an associated closed dislocation of the shoulder), four were injured while lifting a heavy object, and two were injured in a motor-vehicle accident. Five of the twelve patients had had pain in the involved shoulder before the injury. Six of the twelve patients were injured in the workplace, and all six were receiving Workers' Compensation at the time of the operation. The mean duration from the injury to the operation was thirteen months (range, four to forty-nine months). Six patients had the operation within seven months after the traumatic event. The initial physical examination revealed a painful arc of motion of the involved shoulder when the upper extremity was passively manipulated at shoulder height. Palpable crepitus with manipulation of the shoulder was noted in fifteen patients. There was tenderness to palpation in the region of the greater tuberosity and the anterior subacromial space in sixteen patients, and the acromioclavicular joint was tender to palpation in six. Nine patients were unable to elevate the upper extremity actively above shoulder height. Fifteen patients had weakness of external rotation of the shoulder on testing against the manual resistance of the examiner's hand; atrophy of the infraspinatus fossa was noted in nine of these fifteen patients. Radiographs of the shoulder were made preoperatively for all twenty patients; these included anteroposterior radiographs with the shoulder in internal and external rotation, an axillary radiograph, and a supraspinatus outlet radiograph28. Reactive changes, including sclerosis, rarefaction, and formation of cysts, were seen in the greater tuberosity or on the undersurface of the acromion, or at both locations, in all but one patient. Eight patients had evidence of degenerative changes of the acromioclavicular joint, and anteroposterior radiographs demonstrated prominent acromial and clavicular osteophytes inferiorly in six of these patients. Anteroposterior radiographs demonstrated superior migration of the humeral head in five patients; this was manifested by a break in the line between the calcar of the proximal part of the humerus and the inferior neck of the scapula. Nineteen patients had an arthrogram that demonstrated extravasation of contrast medium into the subacromial space. The only patient who had a normal arthrogram was dependent on axillary crutches because of an above-the-knee amputation. After a work-related injury to the shoulder in 1986, an arthrogram was found to be normal. Because of continued pain, an arthroscopic examination and subacromial decompression was performed three years later. The arthroscopic examination revealed a massive tear of the rotator cuff, and the patient had a reconstruction several months later.
Operative Technique The distal aspect of the clavicle was excised in thirteen patients: because of pain and symptomatic osteoarthrosis of the acromioclavicular joint in six, for decompression of the supraspinatus outlet in two, and to facilitate exposure for mobilization of the rotator cuff in five (patients managed earlier in the study period). When it was necessary to excise the distal aspect of the clavicle, the original split in the deltoid was extended superiorly and medially, again with care taken to preserve a good soft-tissue sleeve of the deltoid anteriorly and of the trapezius posteriorly. The dissection was kept lateral to the coracoclavicular ligaments. The distal one and one-half centimeters of the clavicle was then excised. In two additional patients, who had prominent clavicular osteophytes, only the undersurface of the distal aspect of the clavicle was debrided. Thickened bursal tissue was resected to allow clear definition of the extent of the tear of the rotator cuff and to eliminate the potential for any residual mechanical irritation, or so-called non-outlet impingement27, in the subacromial space. Traction sutures were then placed in the supraspinatus tendon superiorly and in the infraspinatus tendon posteriorly to aid in mobilization of the component tendons of the cuff. Thinned and frayed edges of the torn cuff were debrided conservatively. Superficial adhesions along the undersurface of the acromion, deltoid, and trapezius were divided. The coracohumeral ligament was routinely released, and capsular releases were performed along the undersurface of the rotator cuff by dividing the capsule a few millimeters peripheral to the glenoid labrum. The undersurface of the supraspinatus was then mobilized bluntly, with care taken to avoid the suprascapular nerve. The initial goal of the operation was to achieve a direct tendon-to-bone or tendon-to-tendon repair. The ability to repair the cuff depended on the quality and mobility of the soft tissues as well as on the size of the tear. If a primary repair could not be achieved with the upper extremity by the side, subscapularis transfer was considered as an alternative for gaining anterosuperior coverage. However, the subscapularis was transferred only when it was thought that this would allow a secure and complete closure of the defect of the rotator cuff. If the subscapularis was attenuated or of poor quality or if it was thought that the repair was likely to be less than optimum even with the transfer, the subacromial space was decompressed and the rotator cuff was debrided instead37. At the time of the operation, seventeen tears of the rotator cuff were noted to involve the supraspinatus and a substantial portion of the infraspinatus tendon; in six of these shoulders, the tear also involved the teres minor. The remaining three shoulders, in which only the supraspinatus tendon was disrupted, had marked contracture of the muscle and apparent loss of the tendon substance as well. It is noteworthy that only two of the six patients who had the repair within seven months after an alleged traumatic event had operative findings that were consistent with a subacute rupture of the tendon. The other four patients had pathological findings consistent with long-standing disease of the rotator cuff, including early osteoarthrosis of the glenohumeral joint and hypertrophy of the biceps tendon. The technique of subscapularis transfer (Fig. 1) was similar to that described by Cofield9 and others4,11,26,30. Beginning at the lesser tuberosity, the proximal half of the subscapularis tendon was dissected free from the underlying capsule. At its musculotendinous juncture, the muscle was incised along the course of its fibers, obliquely downward and medially, with attention paid to the proximity of the musculocutaneous and axillary nerves. Any subcoracoid or subconjoined adhesions of the tendon were released. An attempt was made to protect the transverse humeral ligament. If the ligament was violated either inadvertently or purposefully in the course of the mobilization of the subscapularis, it was reconstructed with locally available soft tissue or with the inferior border of the transferred subscapularis tendon to prevent subluxation of the biceps tendon. In the event that the anterior aspect of the capsule of the glenohumeral joint was transgressed during the mobilization of the subscapularis, it was repaired directly with non-absorbable suture.
The proximal half of the subscapularis was mobilized sufficiently to facilitate superior and lateral advancement. The inferior border of the transferred portion of the tendon was secured to a trough made in the cancellous bone along the medial aspect of the greater tuberosity, with use of number-1 non-absorbable sutures passed through holes created in the tuberosity. The lateral free edge and superior border of the transferred tendon were then sutured to the trimmed edges of the tears in the infraspinatus and supraspinatus tendons to close the superior portion of the defect. When the repair had been completed, the shoulder was moved through a range of motion. The degree of tension on the repair served as a guide to establish parameters for range of motion during the early phases of rehabilitation. A sling and swathe was routinely applied at the end of the procedure.
Postoperative Management All patients began physical therapy as outpatients within a week after the operation. This initially involved passive elevation of the involved limb in the plane of the scapula. Active-assisted elevation with the patient supine was begun six weeks after the operation. The patients then gradually progressed to active range-of-motion exercises, initially while lying supine and then while lying on an inclined board, and finally to active elevation of the limb against gravity. Starting at eight to ten weeks after the operation, hand-weights, ranging from eight ounces to three pounds (227 to 1361 grams), were worn during the elevation performed with the patient supine, in order to facilitate the transition to elevation of the upper extremity against gravity. This progression was determined by the patient's level of comfort and by evidence of control of the muscles of the shoulder. Pain with exercise or evidence of abnormal scapulohumeral mechanics was considered a sign that the patient was not ready for the next step in the progressive therapy protocol. Gentle active-assisted external rotation was not initiated until eight weeks after the operation, and it was never forceful. More formal muscle-strengthening exercises were not initiated until four months postoperatively. The patients attended physical therapy for a mean of six and one-half months (range, three and one-half to eleven months) postoperatively, and most patients continued to exercise at home for several months thereafter.
Postoperative Evaluation
The patients were evaluated at a mean of thirty months (range, twenty-three to seventy months) after the operation.
Pain
Function
Range of Motion
Satisfaction of the Patients One of these patients had had only slight pain (8 points) preoperatively, with no change postoperatively. He had been unable to elevate the upper extremity actively above shoulder height and had been able to use it only for very light activities. He had the operation solely for the purpose of improving active elevation and functional use of the upper limb. The operation was unsuccessful in this regard, resulting in a total active elevation of 105 degrees. Thus, of the two patients who had the reconstruction for functional improvement in the absence of pain, only one was satisfied with the result. The other two patients who were dissatisfied with the result had had a high level of pain preoperatively, and each had substantial relief (from 4 to 10 points in one and from 0 to 5 points in the other) after the operation. However, both patients had been able to elevate the upper extremity fully overhead preoperatively, and both had marked deterioration of active elevation (70 degrees in one and 110 degrees in the other) postoperatively. Each said that, in retrospect, he would have preferred to have lived with the pain rather than to have lost the ability to elevate the upper extremity fully overhead.
Return to Work
Complications
In this series of twenty patients who were managed with transfer of the proximal half of the subscapularis tendon to achieve closure of a massive tear of the rotator cuff, relief of pain and subjective satisfaction were comparable with those reported after repair of the rotator cuff with direct tendon-to-tendon or tendon-to-bone suture4,8,10,14,17,19,20,36. The high degree of pain relief in the current series is also consistent with previous findings that pain relief after repair of the rotator cuff with acromioplasty is independent of the size of the tear4,20,38. Although others have also reported on the use of subscapularis transposition for reconstruction of large defects of the rotator cuff2,4,9,11,26,27,30,31, Cofield is the only author, to our knowledge, who has critically analyzed the outcome of this procedure9. However, in his report, thirteen of twenty-six patients had subscapularis transposition in combination with glenohumeral prosthetic arthroplasty; therefore, the results in that study cannot be easily compared with those in the current report. Since closure of the defect of the rotator cuff with use of subscapularis transposition was accompanied, in each of our patients, by decompression of the subacromial space, it was not possible to ascertain accurately the impact of closure of the defect on the relief of pain. The rationale for performing this procedure is based on our belief that better results after the operative treatment of tears of the rotator cuff, even massive ones, can be achieved if the rotator cuff can be repaired1,10,20,27,29. The subscapularis transfer was selected for closure of these massive defects because the results of alternative methods, such as those that involve the use of fascial grafts, tendon allografts, or synthetic materials, have not equaled those achieved with direct primary repair. Enthusiasm for supraspinatus advancement techniques has to be tempered by the results of recent anatomical studies, which have demonstrated the limited mobility of the suprascapular neurovascular pedicle39. We have been reluctant to transfer the teres minor because of the potential for additional weakening of external rotation and abduction33. Dissection of the subscapularis muscle is facilitated by the fact that it receives its innervation from both the superior and inferior subscapular nerves. The muscle can be divided in the course of its fibers for a short distance without substantial risk for loss of function; in the current series, it was never necessary to extend the dissection for more than three or four centimeters medial to the musculotendinous junction before there was adequate mobility of the proximal half of the subscapularis tendon for transfer. The distal half of the muscle was always left intact because it is considered to be an important passive stabilizer of the shoulder. Recent studies have suggested that the edges of a torn rotator cuff are well vascularized and should be preserved and incorporated into an operative repair22. However, the edges of a chronic tear can be either bulbous and prominent or thin and friable. Under these circumstances, we believe that it is important to debride the torn edges of the cuff conservatively, to facilitate a secure reattachment of tendon to bone. Thus, in the current series, the quality of the tissue rather than vascularity sometimes justified limited débridement of the torn rotator cuff. Lateral advancement of the subscapularis did not result in a loss of external rotation of the shoulder, as compared with the external rotation reported in other series of repairs of the rotator cuff4,14,20. Of concern, however, was the loss of active elevation and function of the shoulder in two of our patients. The seriousness of this functional deficit cannot be underestimated, as these two patients thought that their condition had been made worse by the operation despite their improved level of comfort. It has been suggested that one of the functions of the subscapularis muscle is to assist in controlling the humeral head in the glenoid by providing a downward direction of force at its insertion21. Some patients who have a massive tear of the rotator cuff may be very dependent on the subscapularis and its downward vector for depression of the humeral head7. It is possible that the patients who lost overhead elevation of the shoulder after transfer of the subscapularis did so by virtue of the loss of some of its effect on stabilization of the humeral head. This in turn could have compromised the efficiency of the deltoid in its role in elevating the upper extremity overhead6,30. Moreover, recent biomechanical studies have demonstrated the capacity of the subscapularis to contribute to elevation in the plane of the scapula33. Thus, in some patients, violation of the subscapularis may have a direct effect on elevation. The only other patient in the current series who had an unsatisfactory result had not had pain before the reconstruction of the rotator cuff. This patient had wanted to have the operation to improve the elevation of the upper extremity; however, the operation was not successful in this regard. The primary indication for reconstruction of the rotator cuff should be pain. The two patients in this series who had slight or no discomfort but marked functional impairment preoperatively were advised at the outset that the functional gains after the repair were less predictable than was relief of pain. Recent studies have documented the difficulties inherent in the quantitative assessment of the results of operative treatment of disease of the rotator cuff23. This is particularly relevant with regard to patients who have a massive tear of the cuff, for whom relief of pain and improved function would seem to be more realistic goals than the restoration of normal function and strength11. Again, the primary indication for repair of the rotator cuff is relief of pain; in the current series, the levels of pain relief and patient satisfaction were high, although almost half of the patients continued to have difficulty with prolonged or repetitive overhead activities. In summary, subscapularis transfer is useful for the closure of massive defects of the rotator cuff that are not amenable to direct repair with more traditional tendon-to-tendon or tendon-to-bone techniques. However, the subscapularis should not be transferred indiscriminately; the tendon should be healthy, and the surgeon should be able to achieve a secure and complete closure of the cuff with the transferred portion of the tendon. Because of the risk of deterioration of active elevation and function of the shoulder after the operation, subscapularis transfer should be used with caution in patients who have full overhead elevation. These latter patients with so-called functional rotator-cuff tears may be better served with alternative techniques such as partial repair7 or decompression and débridement alone.
NOTE: The authors thank Dr. Steven Wenner for his assistance in reviewing this manuscript.
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||