 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Transfer of the Levator Scapulae, Rhomboid Major, and Rhomboid Minor for Paralysis of the Trapezius* , NEW YORK, N.Y.Investigation performed at The Shoulder Service, New York Orthopaedic Hospital, Columbia-Presbyterian Medical Center, New York City
Twenty-two patients who had paralysis of the trapezius muscle secondary to injury of the spinal accessory nerve had transfer of the levator scapulae and rhomboid major and minor muscles. In each patient, function of the trapezius had failed to improve with either physical therapy or an operative attempt at neurolysis or reconstruction of the spinal accessory nerve. The etiology of the injury was biopsy of a cervical node in thirteen patients, trauma in seven, and radical dissection in the neck in two. All patients had pain, visible deformity, and dysfunction of the shoulder girdle. Physical examination revealed asymmetry of the neckline, drooping of the shoulder girdle with lateral displacement of the scapula, and weakness of active elevation. Fourteen patients had had an incorrect clinical diagnosis, and twelve patients had had an inaccurate or incomplete electromyographic examination. A long thoracic nerve palsy developed in three patients. At an average of seven and a half years (range, two to fourteen years), the result of the operative procedure, as determined with the American Shoulder and Elbow Surgeons Shoulder Evaluation Form, was excellent for thirteen patients, satisfactory for six, and unsatisfactory for three. All but three patients had adequate relief of pain and demonstrable functional improvement.
Paralysis of the trapezius secondary to injury of the spinal accessory nerve (a cranial-XI injury) is a disabling problem that leads to pain, deformity, and loss of function9,12,14,21,34,35,38,43-45. The spinal accessory nerve provides the sole motor innervation to the trapezius muscle18,23,34,36, while branches of the third and fourth cervical nerves provide proprioceptive function14,23. Paralysis of the trapezius results in drooping of the entire shoulder girdle, winging and lateral displacement of the scapula, as well as weak forward elevation of the upper extremity6,14,23. The functional disability is compounded by discomfort, which may be secondary to traction on the brachial plexus, periscapular muscle spasm, frozen shoulder, subacromial impingement, or acromioclavicular synovitis6,11,14,33,34,45. Furthermore, thoracic outlet syndrome may develop14, and there may be orofacial involvement secondary to a referral mechanism to other cranial nerves17. The course of the spinal accessory nerve in the posterior cervical triangle is superficial, lying in subcutaneous tissue, and is susceptible to injury from either external3,4,13,23,30 or operative5,14,22,31,34,38,41,43-45 trauma. Excision of the nerve may be necessary during radical dissection in the neck2,19,21, although improved operative techniques permit the nerve to be spared in most instances7,36,37. Paralysis of the spinal accessory nerve of unknown etiology in the absence of trauma has also been reported16,25. For the most part, the results of non-operative treatment for this condition have been unsatisfactory6,12,14,34. While reports have suggested that neurolysis or reconstruction of the spinal accessory nerve has a more favorable outcome, particularly if it is performed within twelve months after the injury, the results have been variable2,6,8,13,14,19,33-35,38,43,45. Several reconstructive procedures to substitute for the paralyzed trapezius have been described. They have included stabilization of the scapula to the spinal processes of the thoracic vertebrae with fascia lata20 or transfer of the levator scapulae6,13,26,39,40,42, or both. For the most part, the long-term results of these procedures have been unsatisfactory because of progressive stretching of the reconstruction. The Eden-Lange procedure provides substitutes for all three components of the trapezius through transfer of the levator scapulae and rhomboid major and minor muscles15,27,28. This technique was first described by Eden15 in 1924, and Lange27,28 reported successful results in 1951 and 1959. Successful results were also reported in three patients6,15,27-29,32. In 1985, we reported the early results for seven patients who had been managed with this procedure6. We report here longer-term results (at a maximum of fourteen years) for twenty-two patients who were managed with this procedure.
Twenty-two patients who had paralysis of the trapezius muscle secondary to injury of the spinal accessory nerve were treated operatively with transfer of the levator scapulae and rhomboid major and minor muscles by the same one of us (L. U. B.), between 1979 and 1993. The preliminary results for seven of these patients were presented in a previous report6. The twenty-two patients ranged in age from eight to seventy-four years, with an average age of thirty-two years. There were six male and sixteen female patients. Thirteen patients had involvement of the dominant extremity. The etiology of the injury of the spinal accessory nerve was biopsy of a cervical mass in thirteen patients, trauma in seven, and radical dissection in the neck because of a tumor in two. Five of the patients who had a traumatic injury had been thrown through the windshield during a motor-vehicle accident, one patient had been struck by an automobile while riding a bicycle, and one patient had sustained a direct traumatic injury to the posterior cervical area. All patients had pain and weakness of the shoulder girdle, especially with overhead activity. The symptoms were varied and, at times, confusing because they originated from several sources. Routinely, patients described the feeling of having a brick on the shoulder. All patients had pain associated with periscapular muscle spasm as well as a dull ache or heaviness about the shoulder and scapula. Many patients also noted intermittent radiation of pain down the upper extremity secondary to traction on the brachial plexus from the unsupported weight of the shoulder girdle and upper extremity. Thirteen patients had pain from subacromial impingement due to malrotation of the scapula. Four patients had glenohumeral pain secondary to a frozen shoulder. In addition, four patients had acromioclavicular pain and tenderness and two patients had sternoclavicular pain and tenderness. All of the patients had an asymmetrical neckline and drooping of the shoulder girdle as well as lateral displacement and winging of the scapula. The patients were unable to shrug the shoulder girdle superiorly on the affected side. The women stated that they were not able to keep their bra strap up on the affected side or to wear a dress with a wide neckline. None of the patients were able to elevate the upper extremity consistently and forcefully above the horizontal plane in any direction, and all had weak forward elevation. They were all limited in their ability to perform overhead activities, lift heavy objects, write, cut meat, and drive an automobile. In addition, three patients had winging of the inferior part of the scapula secondary to serratus anterior palsy, the extent of which was not fully appreciated at the time of the initial consultation. Fourteen patients had been initially diagnosed inaccurately. The misdiagnosis was a herniated cervical disc in three; serratus anterior palsy in two; neuromuscular disorder in two; and thoracic outlet syndrome alone, thoracic outlet syndrome with glenohumeral instability, brachial plexus injury, stroke, acromioclavicular separation, subacromial impingement, and pain about the shoulder in one each. All patients had had an electromyographic study performed, and six patients had had more than one. The results of twelve of the thirty-five electromyographic studies were incorrect, incomplete, or inconclusive. The trapezius was never tested in seven patients. Seven patients had had a total of thirteen previous operative procedures related to the shoulder. Five patients had had a nerve repair, neurolysis, or nerve-grafting (a total of six procedures); two had had an operative procedure to decompress the thoracic outlet; three had had an anterior acromioplasty; one had had an acromioclavicular resection; and one had had glenohumeral arthroscopy. Some form of conservative treatment had failed for all patients, after an average of twenty-two months. The conservative treatment included physical therapy, transcutaneous nerve stimulation, external support, chiropractic consultation, management in a pain clinic, and the use of non-steroidal anti-inflammatory medications as well as narcotic analgesics. The average duration from the initial injury to the muscle transfer was thirty-four months (range, eleven to seventy-eight months). All patients had transfer of the levator scapulae, rhomboid major, and rhomboid minor. The two patients who were managed first in the series had a two-stage procedure that involved transfer of the levator scapulae with a fascial sling from the medial scapular border to the spinal processes, followed by transfer of the rhomboid muscles one year later. In both of these patients, stretching of the fascial sling became apparent by six months postoperatively, resulting in a recurrence of symptoms as well as winging and lateral displacement of the scapula. The American Shoulder and Elbow Surgeons Shoulder Evaluation Form was used to assess pain, function, the range of motion, and the over-all result. Pain was rated on a scale from 0 points (complete disability secondary to pain) to 5 points (no pain). Function was evaluated on a scale from 0 points (an inability to perform functions) to 4 points (normal function). Only forward elevation (0 to 180 degrees) was measured for the assessment of the range of motion as this was the only arc that was substantially deficient in all patients. (Most of the patients had normal external and internal rotation preoperatively.) The over-all result was rated as excellent, satisfactory, or unsatisfactory. An excellent rating was given when the patient was fully satisfied, had no substantial pain, and had full use of the extremity, normal strength, and active elevation within 10 degrees of that on the unaffected side. A satisfactory rating was given when the patient was adequately satisfied, had occasional pain, and had functional use of the extremity for activities of daily living, no more than 30 per cent less strength than the unaffected side, and 90 to 135 degrees of total elevation. An unsatisfactory rating was given when a patient did not meet these criteria.
Operative Technique and Postoperative Protocol Next, the supraspinatus and infraspinatus are elevated from the supraspinatus and infraspinatus fossae for a distance of five to six centimeters or at least a half of the width of the scapula. This modification of the previously described technique6 permits transfer of the rhomboid minor to the supraspinatus fossa cephalad to the scapular spine rather than to the infraspinatus fossa caudad to the scapular spine and places the muscle at a mechanical advantage to substitute for the function of the middle portion of the trapezius (Fig. 1). The rhomboid major is transferred as before to the infraspinatus fossa. With use of a Hall drill, two drill-holes are made cephalad to the scapular spine and four are made caudad to it, one and a half to two centimeters apart and four to five centimeters lateral to the medial border of the scapula. The transferred muscles are secured with number-2 heavy, non-absorbable sutures with a large curved needle that passes through the holes. The sutures are tied with the scapula in the reduced position and the upper extremity abducted approximately 90 degrees in the coronal plane. The infraspinatus is imbricated over the repair.
The spine of the scapula is palpated and a four-centimeter incision is made starting three centimeters medial to the posterior tip of the acromion and extending medially. The trapezius, deltoid, and supra-spinatus are dissected carefully so that three drill-holes can be made through the superior aspect of the scapular spine, taking care not to injure the suprascapular nerve on the deep surface of the suprascapular muscle. A tunnel connecting the medial and lateral wounds is made through the atrophied trapezius, in line with its superior fibers, for passage of the levator scapulae. The transferred levator scapulae is secured with heavy non-absorbable sutures. Care is taken not to transfer the levator scapulae too far laterally on the spine as this leads to a web-like deformity in the neck. The optimum position is five to seven centimeters from the posterolateral corner of the scapular spine. The incisions are sutured in layers, and an abduction foam wedge is applied at the side. An abduction brace was used early in the series but was extremely uncomfortable and was poorly tolerated by the patients, especially when they were supine and the posterior incision rubbed against the metal brace. A foam wedge is better tolerated and is effective in maintaining approximately 60 to 70 degrees of abduction at the glenohumeral joint for four weeks. The postoperative rehabilitation program was similar to that described in our original study6, with the addition of early passive range-of-motion exercises above the level of the wedge. The upper extremity is passively elevated in the scapular plane to 130 to 140 degrees and rotated externally in the range of 30 to 40 degrees. This modification was an effort to prevent stiffness, which can result from four weeks of immobilization. Passive motion is begun on the first postoperative day, before discharge from the hospital, which is usually on the second or third day, and is continued during the first four weeks. At four weeks, the wedge is removed and gentle strengthening exercises are added to the range-of-motion exercises. A special progressive strengthening program was devised that involves use of rubber tubing, free weights, and medicine-ball throws to achieve dynamic scapular stability. These exercises are aimed at strengthening the transferred levator scapulae and rhomboid muscles. Shoulder shrugs are performed with the patient holding light weights (three to five pounds [one to two kilograms]) initially and progressing to ten-pound (four-kilogram) weights as tolerated. Medial retraction of the scapula is also performed against resistance, with the use of either rubber tubing or three, four, or five-pound (one, one and a half, or two-kilogram) weights. During the exercises, the scapula is actively moved to the midline. Overhead throws of a six-kilogram medicine ball also improve scapulothoracic rhythm.
The average duration of follow-up was seven and a half years (range, two to fourteen years). Two patients, one who had squamous-cell carcinoma of the larynx and one who was seventy-four years old at the time of the operation, died during the follow-up period but had been followed for thirty-six months and forty-eight months, respectively. Over-all, the average pain score was 0.9 point preoperatively and 4.3 points postoperatively: an average improvement of 3.4 points. Postoperatively, nineteen patients had no pain at rest and slight pain or discomfort after strenuous activity. For these nineteen patients, the average pain score was 1 point preoperatively and 4.7 points postoperatively: an average improvement of 3.7 points. Three patients had inadequate relief of pain. Their average pain score was 0.3 point preoperatively and 2 points postoperatively: an average improvement of 1.7 points. One of these three patients had pain relief until sixteen months postoperatively, at which point increasing symptoms of subacromial impingement began to develop, with pain and decreased function. At the most recent follow-up examination, thirty-six months after the transfer procedure and one year after repair of a supraspinatus tear of the rotator cuff, the pain in the shoulder had decreased but had not resolved completely, and over-all function was still unsatisfactory. The second patient had persistent pain secondary to a greater auricular neuroma after multiple nerve-grafting procedures that had been performed before the muscle transfer. The muscle transfer eliminated the pain associated with muscle spasm and the pain that radiated down the upper extremity but did not affect the pain associated with the neuroma. At the time of the most recent follow-up examination, the patient was considering another neurolysis and removal of the scar tissue for relief of this pain. The third patient who did not have substantial pain relief had had twenty-eight previous procedures, including carotid endarterectomy, axillary artery bypass, release of the thoracic outlet, and two attempts at neurolysis of the spinal accessory nerve. The source of the continued pain was unclear. Thirteen patients obtained normal function. Six patients had satisfactory function with some limitation of the ability to carry out more strenuous activities. Three patients had difficulty with several activities listed on the American Shoulder and Elbow Surgeons Shoulder Evaluation Form, including throwing, overhead lifting, and performing work-related or routine sports activities. They were able to perform activities of daily living successfully, but the result was still considered unsatisfactory. The functional limitations of one of these patients, who was discussed previously, were due to occasional pain from a greater auricular neuroma. The second patient had been involved in a motor-vehicle accident and had initially been diagnosed as having thoracic outlet syndrome and glenohumeral instability. She was managed operatively with resection of the first rib as well as a repair to treat anterior instability of the shoulder three months later; however, she continued to have symptoms. Nineteen months after the injury, the patient had muscle transfer for paralysis of the trapezius and had good relief of pain but residual glenohumeral instability. Although she subsequently had a number of procedures to treat the glenohumeral instability, the scapula remained stable. The unsatisfactory result in the third patient was secondary to weak forward elevation. Nineteen patients were able to elevate the upper extremity above the horizontal with enough strength for functional activity. Three patients were unable to elevate the upper extremity above 90 degrees. Six patients had episodes of intermittent muscle spasm and pain that responded to strengthening exercises and modalities such as ultrasound and galvanic stimulation. Thirteen patients had an excellent over-all result, with no limitations. Symmetry of the shoulders was reestablished (Figs. 2-A and 2-B). Twelve of the thirteen patients were athletic, participating in sports activities such as tennis, skiing, aerobics, karate, swimming, volleyball, racquetball, bowling, and golf. One of these patients had sustained multiple injuries and participated in wheelchair sports. Six patients had a satisfactory result; they had some residual asymmetry of the neck with mild drooping of the shoulder and slight winging, but they had marked improvement compared with the preoperative status. Three patients had an unsatisfactory result, which was secondary to pain in one and secondary to weakness in two, as mentioned previously. Three patients had had a long thoracic nerve palsy in addition to the paralysis of the trapezius, which necessitated transfer of the pectoralis major tendon to the scapula. Two of these patients had improvement after this additional procedure.
Paralysis of the trapezius disrupts the dynamic equilibrium of muscle forces about the shoulder. Codman10 recognized the importance of this intricate muscle balance, and Inman et al.24 stressed the contribution of the trapezius to the balance of the scapular rotatory-force couples allowing smooth scapulothoracic rhythm (Figs. 3-A and 3-B). In our experience, transfer of the levator scapulae and rhomboid major and minor muscles (the Eden-Lange procedure) provides an adequate substitute for the complex function of the trapezius muscle through replacement of all three of its anatomical components. The levator scapulae and rhomboid muscles insert on the medial border of the scapula and normally contract as the trapezius acts during rotation, elevation, depression, and medial stabilization of the scapula. Because of their medial insertion, they are unable to prevent winging and lateral displacement of the scapula and drooping of the shoulder girdle when the trapezius is paralyzed. Lateral transfer of these muscles to the lateral aspect of the spine and body of the scapula changes the direction of their pull, creating a biomechanical advantage that allows them to approximate the function of the trapezius and to support the scapula (Fig. 1). These three muscles act together as a unit to help to replace the dynamic equilibrium of muscle forces from about the scapula that was lost with paralysis of the trapezius. Recently, we modified the original Eden-Lange procedure by transferring the rhomboid minor cephalad to the scapular spine within the supraspinatus fossa (Fig. 1). This more closely approximates the function of the middle portion of the trapezius, enhances stabilization of the superior angle of the scapula, and closes the gap between the rhomboid minor and the levator scapulae.
These transferred muscles are not as large as the trapezius; therefore, intensive and persistent strengthening exercises are needed to achieve optimum function. Twelve patients were participating in strenuous athletic activity at the time of the most recent follow-up examination, but they had to perform strengthening exercises to maintain optimum function. Successful treatment of paralysis of the trapezius is contingent on an accurate diagnosis and the development of a treatment plan that will best suit the needs of the patient. The diagnosis of paralysis of the trapezius may not be apparent. Fourteen of our patients were misdiagnosed before referral to us, and many had been examined by an orthopaedic surgeon, neurologist, neurosurgeon, or psychiatrist. The difficulty in making this diagnosis may be due in part to the fact that the clinical symptoms can be vague and the results of the physical examination can be confusing. The iatrogenic nature of this injury in some adults may have led to a delay in the diagnosis. The symptoms can be variable and often are related to the secondary etiology, such as subacromial impingement, frozen shoulder, radiculitis, and acromioclavicular or sternoclavicular synovitis, caused by the unstable shoulder girdle. In addition, an accurate physical examination may be compromised by impaired mobility due to pain or, as was sometimes the case in our series, by the patient not being completely undressed from the waist up during the examination. It is important to examine the patient from behind and to observe the medial border of the scapula during mobilization of the shoulder, with a visual comparison with the contralateral scapula. The electrodiagnostic examination must be precise and oriented toward evaluation of the trapezius and the proximal scapular muscles. Without the specific information provided by appropriate electromyography, it is not possible to assess which muscles are involved or the extent of their involvement. Part of the confusion with regard to the diagnosis may be related to the evaluation of the cervical roots or distal peripheral nerves. The electromyographer should be aware of the need for a complete evaluation not only of the levator scapulae but also of the rhomboid muscles. These muscles may be difficult to evaluate because they lie close to the chest wall and may be adherent to the undersurface of the atrophied trapezius. They must be carefully tested close to their insertion on the medial scapular border. Of interest were the three patients who had an additional long thoracic nerve palsy. The paralysis of the trapezius in all three patients was iatrogenic following an operative procedure. We believe that the etiology of the long thoracic nerve palsy was secondary to traction on the long thoracic nerve by the unsupported shoulder girdle. Our approach was initially to correct the paralysis of the trapezius and subsequently to evaluate the effect of the weakness of the serratus anterior. Only two muscles—the serratus anterior and the trapezius—can produce upward rotation of the scapula23. The serratus anterior produces upward rotation by pulling the inferior angle forward more rapidly than the medial border, while the proximal part of the trapezius assists by lifting the lateral angle of the scapula. In each patient, the scapular instability persisted and necessitated operative treatment to transfer the sternal portion of the pectoralis major to substitute for the function of the serratus anterior. Non-operative treatment may have a place in the treatment of paralysis of the trapezius, although it is usually reserved for older, sedentary patients. For active patients, exercise programs aimed at strengthening adjacent muscle groups proved inadequate and did not compensate sufficiently for the loss of the function of the trapezius. Transfer of the levator scapulae and the rhomboid major and minor as a reconstructive procedure should be considered for more active patients. However, if the diagnosis is made within one year after the injury of the spinal accessory nerve, neurolysis, nerve repair, or nerve-grafting should be considered before a muscle transfer. The results reported in the literature1-4 have been variable, although it seems that exploration of the injured nerve within six months after the injury yields the best results and that the latest that this type of operation should be performed is one year after the injury. All but one of the patients in our series had the muscle transfer more than one year after the initial injury, and the remaining patient had it eleven months after the initial injury. Also, operative reconstruction of the nerve had failed in five patients before the muscle transfer, with one patient having a persistent painful neuroma after grafting with the greater auricular nerve. Despite a successful muscle transfer, the result in this patient was compromised by this complication. We conclude that transfer of the levator scapulae and the rhomboid muscles is a valuable orthopaedic reconstructive procedure for the treatment of paralysis of the trapezius. It should not be performed as a primary procedure when neurolysis or repair of the spinal accessory nerve is indicated, but it is a reasonable salvage procedure for a patient who has pain, deformity, and diminished function of the shoulder girdle caused by irreparable injury of the spinal accessory nerve.
*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.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |