Scapular Stabilization in Patients With Spinal Cord Injury

Abstract

Background:

Severe scapular instability can be a considerable problem for people with high-level cervical spinal cord injury. Scapular instability reduces the effectiveness of the already weakened shoulder flexors and abductors, thereby limiting hand-to-mouth and hand-to-head activities. The winged scapula may cause inferior pole skin breakdown, as well as neck and shoulder pain.

Objective:

To report the efficacy of a fusionless scapular stabilization procedure as a means to enhance function in a consecutive group of patients with high-level cervical spinal cord injury.

Methods:

Four people with spinal cord injury at C4–C5 (2 male, 2 female; mean age  =  17.3 years, range  =  14–20 years) underwent scapular stabilization via scapulothoracic fusion (N  =  2) or by tethering the scapula to the rib cage with Mersilene tape as a fusionless stabilization (N  =  2). One patient died of unrelated causes 18 months after surgery, and the remaining 3 were followed for 26, 39, and 41 months, respectively. Data collection included radiographic analysis, active range of motion measures, and functional assessment.

Results:

Active shoulder flexion and abduction remained unchanged in 2 patients, but functional scores improved with regard to feeding and grooming capability. All patients reported satisfaction with postoperative appearance, and 3 patients reported considerable reduction in shoulder pain after surgery. Radiographs demonstrated maintenance of stable scapular alignment in all patients at final follow up. Wound breakdown, requiring removal of instrumentation, occurred in 2 patients.

Conclusion:

Scapular stabilization with or without fusion is a viable option to improve appearance, pain, and upper extremity function in people with high-level tetraplegia and scapular instability.

INTRODUCTION

Dysfunction of the scapulothoracic articulation has been well documented, most commonly in the face of scapular winging (1). Scapular winging may result from a multitude of etiologies. Frequently, it is the result of a traumatic injury to the serratus anterior muscle or long thoracic nerve (2–4). Atraumatic etiologies include poliomyelitis and muscular dystrophy (5–7). Facioscapulohumeral dystrophy, a degenerative muscle disease characterized by slowly progressive weakness of the muscles of the face, upper arm, and shoulder girdle, has been commonly described as leading to scapular instability secondary to muscle weakness (6,8–10). However, in our experience, this clinical finding has been under appreciated in patients with spinal cord injury (SCI).

Various surgical treatment options have been described for scapular instability secondary to serratus anterior palsy and facioscapulohumeral dystrophy. Conner et al (11) described split pectoralis major transfer for serratus anterior palsy with a 90% success rate. Bunch et al (9) and more recently Krishnan et al (7) reported different techniques for scapulothoracic arthrodesis for muscular dystrophy and serratus anterior palsy with significant improvements in shoulder range of motion (ROM) and patient satisfaction. Atasoy et al (12) demonstrated positive results with scapulothoracic stabilization using autogenous fascia lata grafts for patients with brachial plexus injuries.

Shoulder complaints have been reported in up to 51% of patients with SCI, regardless of the level of injury (13–15). Although not commonly described, symptomatic scapular winging and instability can be significant problems for people with mid- to upper-level cervical SCI (C4–C6) from both a functional and cosmetic standpoint. This presents as difficulty and weakness with shoulder abduction and flexion and hand-to-mouth activities, as well as skin breakdown from a prominent inferior pole of the scapula. Unfortunately, cervical SCI with scapular instability is often refractory to conservative modalities of scapular stretching and strengthening therapy secondary to global weakness (13). Orthotic use, soft or molded, can provide some comfort and limited functional improvement. However, in our patients, the bulkiness of the orthosis, the need to don/doff the device, and pressure needed to support the scapula limit long-term usage. We sought to retrospectively evaluate the efficacy of surgical scapular stabilization as a means to enhance quality of life and shoulder ROM in this select patient population, because it has not previously been described in people with SCI.

METHODS

From 1999 to 2001, 4 patients with C4–C5 SCI and severe scapular instability with functional deficits were identified and recommended for stabilization. There were 2 male and 2 female patients, with an average age of 17.3 years (range  =  14–20 years) at the time of surgery. Two patients underwent scapulothoracic fusion with titanium cables and plates. In the remaining 2 patients, a fusionless scapular stabilization procedure was performed by tethering the scapula with plates to the rib cage with Mersilene tape. One patient with a fusion (patient 1) died of unrelated causes 18 months after surgery. The remaining 3 patients were followed an average of 35.3 months (range  =  26–41 months).

Indications for the procedure included limited forward flexion and abduction of the adversely affected shoulder, early fatigue with attempted active shoulder ROM, a bony prominence causing pain and skin breakdown at the inferior pole of the scapula, and a subjectively undesirable cosmetic appearance. One patient had no active ROM of the shoulder but did have significant scapular winging that resulted in considerable pain and skin breakdown at the inferior pole of the scapula. This patient poorly tolerated figure-of-eight bracing, which attempted to passively stabilize a substantial scapular prominence. All patients demonstrated a positive response to the “scapular stabilization test” (3), in which the scapula is manually reduced to the chest wall by a therapist while the patient performs various ROM activities. When this maneuver was performed in our patients, an improvement in ROM was noted (Figure 1).

Figure 1.

Example of “scapular stabilization test.” (a) Maximal active forward flexion achievable in a patient with C5 tetraplegia. (b,c) A noticeable increase in active shoulder range of motion is demonstrated after manual external scapular stabilization is applied by a therapist.

Patients were examined and evaluated preoperatively and postoperatively at regular intervals by a senior orthopaedic surgeon and therapists experienced with upper extremity issues in patients with SCI. The patients' medical charts were retrospectively reviewed after approval from our Institutional Review Board. Data collection included pre- and postoperative radiographic analysis, active ROM measures, functional assessment with the Canadian Occupational Performance Measure (COPM) and Functional Independence Measure (FIM), and overall satisfaction with regard to pain and cosmetic appearance at final follow up. Both the COPM and FIM are validated objective measures of a patient's functional capacity (16,17).

Operative Technique

Two patients (patients 1 and 2) underwent formal scapulothoracic fusion. This was achieved by affixing a 3.5-mm pelvic reconstruction plate to the medial border of the scapula. Titanium wires were passed around ribs 4 through 7 and then subsequently through the holes in the reconstruction plate on the scapula. The reconstruction plate was utilized to prevent the wires from possibly cutting through the scapula after tensioning. The ribs and under surface of the scapula were partially decorticated, and the wires were tensioned with the scapula in approximately 15° of external rotation and 10 cm abducted from the midline (Figure 2) (6,7,12). This position was chosen to maximize possible postoperative shoulder motion. Fixation of the scapula at too large of an angle (>30° external rotation) will inhibit patients' ability to fully adduct their shoulders, whereas fixation at a more acute angle (<10° external rotation) will limit potentially greater shoulder abduction (6,12).

Figure 2.

(a) Patient 2 demonstrating bilateral medial scapular winging. (b) Preoperative radiograph of patient 2 with bilateral scapular winging (outlined). (c) Postoperative radiograph of patient 2 after bilateral scapular fusion.

In patients 3 and 4, fusionless scapulothoracic stabilization was performed. A one-third tubular AO plate was placed on the lateral border of the scapula, and a malleable retractor was held deep to the ribs to protect the pleura and chest cavity contents. Mersilene tape was then wrapped around each rib and passed through the scapula and one-third tubular plate. The plate was similarly utilized in this procedure to prevent cutout of the Mersilene tape through the scapula. The scapula was reduced to approximately 15° external rotation, and the Mersilene tapes were tied down so as to allow some scapular motion with movement of the upper extremity. A pelvic reconstruction plate was applied with screw fixation to the eighth rib medial and lateral to the scapula at its inferior border. No screws in this plate attached to the scapula (Figure 3). This was utilized as a way to stabilize the scapula but prevent a fusion. Upon gentle ROM of the upper extremity, minor movement of the scapula was visible at the completion of the procedure. Patients were placed in a sling postoperatively, and ROM exercises were initiated after 2 weeks.

Figure 3.

(a) Patient 3 demonstrating substantial medial winging of the right scapula. (b) Preoperative radiograph of patient 3 with medial winging of the right scapula (outlined). (c) Postoperative radiograph of patient 3 after fusionless stabilization of the right scapula.

RESULTS

Active shoulder flexion and abduction improved in 1 patient (patient 3), but both the COPM and FIM scores improved with regard to feeding and grooming capability (Tables 1–3) for both patients with scapulothoracic stabilization (patients 3 and 4). Three with (patients 1, 3, and 4) reported a considerable reduction in shoulder pain (Table 4). All patients reported satisfaction with postoperative appearance. Radiographs demonstrated maintenance of stable scapular alignment in all patients at final follow up.

Table 1.

Active Range of Motion of the Shoulder

Table 2.

Functional Independence Measure^a

Table 3.

Skin Breakdown

Table 4.

Shoulder Pain (Visual Analog Scale)

Patient 1 received a unilateral scapular fusion for her significant scapular winging. She had no active ROM of her upper extremities but had considerable pain of the involved shoulder, as well as significant stage II ulcerations about the inferior pole of the scapula, which failed conservative treatment. A unilateral scapular fusion was performed in an effort to relieve her pain and complications stemming from continued skin breakdown about her scapula. Postoperatively, her pain was relieved and the pressure ulcerations about her scapula resolved. Unfortunately, the patient died of unrelated causes (systemic sepsis) at 18 months postoperatively.

Patient 2 underwent bilateral scapular fusions. Preoperatively, she had active ROM of 0° and 30° of shoulder flexion and abduction, respectively. She demonstrated difficulty with hand-to-mouth activities, as well as stage II ulcerations about the inferior poles of both scapulae. A bilateral scapulothoracic fusion was performed in an effort to stabilize her scapula and possibly improve her shoulder active ROM, because she demonstrated moderate improvements with external scapular stabilization applied by a therapist. Postoperatively, she demonstrated no improvement in active ROM or independence with activities of daily living as measured by the COPM and FIM. The instrumentation was removed 6 weeks postoperatively at the patient's request secondary to an undesirable “stiffness.” Intraoperative exploration revealed a stable fibrous scapulothoracic union with approximately 2 cm of allowable motion, similar to that of patient 4. Her preoperative pressure ulcerations about her scapula resolved; however, she continued to have chronic neck and shoulder pain of 4/10 on a visual analog scale (18) at her last visit.

Patient 3 underwent unilateral fusionless scapular stabilization. He had shoulder active ROM of 50° and 85° of flexion and abduction, respectively, as well as significant unilateral shoulder pain and skin breakdown at the inferior scapular pole. This patient was noted to have a considerable improvement in shoulder active ROM, as well as a decrease in shoulder fatigue with activities when external scapular stabilization was applied by our therapist (Figure 1). Because our previous patient with scapulothoracic fusion (patient 2) was dissatisfied with postoperative shoulder stiffness, it was postulated to offer this patient a novel scapulothoracic stabilization procedure without formal fusion.

Postoperatively, patient 3 demonstrated increased active ROM, stability, and decreased fatigue with shoulder flexion and abduction. Preoperative shoulder flexion and abduction measured 50° and 85°, respectively, and improved to 110° and 100° after scapular stabilization. This manifested as an increased ability with regard to eating and grooming, as indicated with the COPM and FIM. At 39 months, the patient reported being “pain free,” and preoperative difficulties with skin breakdown at the inferior pole of the scapula had been eliminated. The patient stated he was “very satisfied” and ranked his satisfaction with the procedure a 9/10 on the Likert scale (18).

Patient 4 underwent bilateral fusionless scapular stabilization as a staged procedure. This patient had shoulder active ROM of 90° and 90° of flexion and abduction, respectively. However, he had difficulty with independent hand-to-mouth activities, considerable bilateral shoulder pain, and continued skin breakdown about the inferior pole of both scapulae. Similar to patient 3, this patient was offered a scapulothoracic fusionless stabilization to promote improved shoulder active ROM, decrease fatigue, and prevent shoulder stiffness. Patient 4 also demonstrated considerable improvement in functional ability with external scapular stabilization applied by a therapist preoperatively. Postoperatively, the patient's ROM remained unchanged; however, an increase in grooming capabilities was seen with the COPM and FIM. Similar to patient 3, this patient also exhibited decreased fatigue in his bilateral upper extremities, allowing an improvement in activities of daily living for extended periods. The hardware had to be removed at 10 months secondary to loosening. Upon exploration, a fibrous scapulothoracic union was achieved as planned, allowing approximately 2 cm of excursion of both scapulae with shoulder ROM. At final follow up, the patient reported satisfaction with the procedure and cosmetic appearance. He reported no pain, and skin remained intact without evidence of pressure ulceration.

Complications

Three of 4 patients experienced postoperative complications with wound breakdown. One patient (patient 4) experienced recurrent wound breakdown and subsequent hardware loosening, requiring removal of all hardware at 10 weeks. Wound breakdown about the scapula had completely resolved in all patients by the time of final follow up.

One patient (patient 2) with a scapulothoracic fusion was displeased with the postoperative “stiffness.” The patient was taken back to the operating room for removal of hardware and takedown of the fusion after 6 weeks. A fibrous union was noted upon removal of the plates and wires, which permitted approximately 2 cm of motion of each scapula.

DISCUSSION

Fusionless scapular stabilization should be a consideration in high-level SCI, where winging leads to skin breakdown, shoulder pain, and reduced shoulder motion and where the already limited breadth of functional activities in these patients may be negatively affected by absent scapular mobility. The latter reason is an important delineation for considering the fusionless technique vs scapular fusion in patients with high-level SCI.

Kinematic analysis of arm elevation in patients with C5–C6 SCI revealed a reduction in total motion with concomitant reductions in both glenohumeral elevation and scapulothoracic lateral rotation (19). Along with muscle paralysis about the shoulder and scapula, components of scapular winging were reported (19). The global paralysis observed in our subjects with C4–C5 spinal lesions resulted in significant scapular winging that persisted both at rest and during arm elevation (Figure 4). When 2 of the primary scapula stabilizers (serratus anterior, rhomboids) are weak or incompetent, the glenohumeral musculature is placed at a mechanical disadvantage. Thus, effective glenohumeral motion is diminished due to the lack of scapulothoracic control (9). Scapulothoracic position is significantly improved by stabilizing the scapula to the rib cage (8–10,12). This provides a fixed point for the glenohumeral joint to rotate in an effort to improve overall shoulder ROM (8–10,12).

Figure 4.

Medial winging of bilateral scapulae demonstrated by considerable protrusion and rotation of the medial border of both scapulae.

As previously stated, there are a multitude of causes for scapular instability and resultant winging. Traumatic injury to the long thoracic, spinal accessory, or dorsal scapular nerves may lead to scapular instability (2–4). Atraumatic etiologies include neuralgic amyotrophy, polio, and muscular dystrophies, notably, facioscapulohumeral muscular dystrophy (5–10). Scapular instability as a result of mid- to upper-level cervical SCI has not been extensively described.

In C4–C5 level SCI, loss of stimulated lower motor neuron function has been reported in the C5 and C6 innervated muscles along with an extended zone of lower motor neuron loss up to several segments caudal to the primary neurologic motor level (20,21). Independent from the cervical level of lesion, the zone of spinal injury may extend several segments beyond the primary motor neurologic level (22). Clinically, subjects in our investigation had little to no volitional motor function caudal to the C5 segment and variable motor preservation within the C5 segment. Of the primary scapula movers (trapezius, levator scapulae, serratus anterior, rhomboids), patients in our investigation had no clinical evidence of serratus anterior activation and retained functional trapezius muscle activation. Other muscles that affect scapular motion and that receive a greater proportion of innervation from spinal cord segments caudal to C5, such as the clavicular fibers of the pectoralis major and pectoralis minor, were paralyzed. Shoulder muscles with C5 innervation (middle > anterior deltoid, rotator cuff muscles, teres major, and elbow flexors) were variable in presentation while significantly weak. The scapular winging and postural asymmetries present in our patients, while largely attributed to serratus anterior paralysis, were likely exacerbated by muscle imbalances about the scapula, shoulder, and spine.

Patients with high-level SCI and scapular instability differ from those with isolated scapular winging in that the patients with C4–C5 SCI have a more global upper extremity weakness pattern and are entirely reliant on their upper extremities and compensatory scapular/spinal movement for function. Attempts at conservative treatment failed in all 4 of our patients. Due to comprehensive weakness of the shoulder girdle in patients with C4–C5 SCI, strengthening of the scapular stabilizers is often not possible. Warner and Navarro (3) describe a “scapular stabilization test,” in which the scapula is manually reduced to the chest wall by a therapist while the patient performs various ROM activities. When this maneuver was performed in our patients, an improvement in ROM and a decrease in fatigue with feeding and grooming activities were noted. A variety of figure-of-eight braces and taping configurations were attempted to maintain scapular reduction against the chest wall. These conservative measures were all poorly tolerated by our patients, who reported that the braces were painful and cumbersome. It is in this setting of failure with conservative treatment that formal scapular stabilization may be indicated (1,7).

Surgical stabilization of the scapula has been reported with both soft-tissue procedures and arthrodeses. Pectoralis major transfer has been demonstrated to provide satisfactory functional improvement in patients with isolated serratus anterior palsy (3,11). However, this procedure is not indicated if muscles other than the serratus anterior are compromised. Additionally, nerve transfers have been described for reduction of scapular instability and improvement in ROM for scapular winging (23,24). These procedures have proved quite beneficial, with few complications for scapular winging secondary to isolated nerve injury/dysfunction, and do not require instrumentation. However, these soft-tissue procedures are unfortunately not applicable to many patients with cervical SCI, who, as described above, have little or no neurologic function at or below C5. Sufficiently functional muscle units and nerves are thus not available for transfer or grafting. In this case, a fusion or stabilization procedure is usually warranted and may be the only surgical option (7,8).

Scapular arthrodesis and stabilization has been more commonly described for facioscapulohumeral muscular dystrophy, with no reports in patients with SCI. Letournel et al (10) reported an improvement in flexion and abduction of 33° and 25°, respectively, after scapulothoracic arthrodesis with a transfixing rib. In this procedure, the superior most rib is cut and passed through a hole created in the medial border of the scapula. The rib is affixed to the adjacent scapula with a plate and screws. Complications, including pneumothorax and/or pleural effusions, occurred in one third of the patients. Bunch and Siegel (9) described a fusion technique using multiple stainless steel wires and iliac crest autograft. These patients demonstrated improved shoulder abduction but required immobilization in a shoulder spica cast for 2 months. Diab et al (8) fused the scapula to the rib cage using 16-gauge wires passed through a semitubular plate affixed to the posteromedial border of the scapula. Patients were also immobilized for a period of 2 months. Shoulder ROM improved in 7 of 11 patients.

Ketenjian (6) and, more recently, Atasoy and Majd (12) described scapulothoracic stabilization without fusion in which fascia lata graft and/or Mersilene tape is passed around the ribs and through drill holes in the medial border of the scapula. Shoulder ROM reportedly improved 37% to 70%, with markedly less postoperative shoulder immobilization required. Follow up averaged 34 months (6) and 6.3 years (12), respectively. Scapular stabilization without fusion presents the important added benefit of a markedly shortened postoperative immobilization when compared with scapulothoracic fusion. Without the need to achieve an osseous fusion, a patient can begin ROM exercises in as little as 3 to 7 days (6), whereas scapulothoracic fusion requires shoulder immobilization for up to 3 months (7).

To our knowledge, there are no reported studies describing possible treatments for scapular instability secondary to SCI. The patient with SCI and resultant scapular instability and winging is at a significant disadvantage. Cervical SCI differs from facioscapulohumeral muscular dystrophy in that the patients are entirely reliant on their upper extremities for all function. Independence can be significantly hindered or lost if a patient is unable to perform hand-to-mouth or hand-to-head activities secondary to scapular instability. Further, a prominent inferior pole of the scapula can serve as a significant pressure point while in a wheelchair, commonly resulting in skin breakdown and decubitus ulcerations. A multitude of different custom wheelchair padding were attempted for our patients to relieve or offload the pressure point of the prominent scapula, but this was met with limited success. Formal surgical scapular stabilization was thus recommended to improve function, pain, skin care, and appearance.

Overall, our patients did well postoperatively. Subjectively, all patients reported that they were “very satisfied” with the postoperative cosmetic appearance. Three of 4 patients reported complete relief of preoperative shoulder pain (5/10 to 0/10 using the visual analog scale). The patient with bilateral fusion (patient 2) reported dissatisfaction with the “rigid” feel of her shoulder and complained of continued neck and shoulder pain. Her instrumentation was removed for this reason, but she achieved little benefit with regard to pain relief and shoulder rigidity.

Considerable improvement in functional capabilities was achieved in both patients with fusionless stabilization compared with patient 2 with a bilateral scapulothoracic fusion. Patient 2 did not demonstrate a marked gain in shoulder ROM or functional capability, again reporting dissatisfaction with postoperative shoulder rigidity. This is in contrast to patient 3, with unilateral fusionless stabilization, who demonstrated a significant gain in shoulder ROM, resulting in 110° flexion and 100° abduction. Patient 4 with bilateral fusionless stabilization did not exhibit a significant improvement in shoulder ROM. However, both patients with fusionless scapular stabilization (patients 3 and 4) displayed considerable gains with regard to functional ability with activities of daily living, progressing from required assistance to independence with feeding and grooming as assessed by the COPM and FIM. This was thought to be related to improved motion but was also related to patient reports of a substantial decrease in fatigue they appreciated when performing activities of daily living after scapular stabilization.

Difficulties with skin breakdown at the inferior pole of the scapula preoperatively had been eliminated in all patients at final follow up. However, 3 of 4 patients (patients 1, 2, and 4) did have difficulties with early postoperative wound healing. Minor postoperative surgical wound breakdown was effectively managed with conservative local wound care measures for patients 1 and 2. Patient 4 required removal of prominent hardware at 10 months, after which there were no recurrent wound complications. Postoperative wound care is challenging in this patient population, because the surgical incisions are in a relatively dependent area when the patient is seated in a wheelchair. Although only 1 of 4 patients required removal of prominent instrumentation, the patient was counseled that the risk for wound complications is higher in this population. Consideration may be given to a planned removal of instrumentation as part of a 2-stage procedure if there is little soft-tissue coverage available over the planned operative site.

This study presents two surgical procedures that attempted to address the difficulties of scapular instability in patients with cervical SCI. Formal scapular fusion did provide some limited benefit to 2 of our patients. However, subsequent shoulder rigidity was not well tolerated, and there was a limited improvement in functional capabilities. Both patients undergoing the fusionless stabilization procedure were able to achieve considerable improvement in functional capability, pain relief, and skin care. Thus, the fusionless stabilization procedure appeared to result in better overall outcomes than scapulothoracic fusion in this small case series. Further study on these procedures in this patient population is certainly warranted.

CONCLUSION

Scapular instability and winging can result in a multitude of problems for people with SCI, including loss of motion affecting independence, pain, and skin breakdown, and an undesirable cosmetic appearance. The fusionless stabilization procedure aims to restore scapular stability while still allowing a small amount of scapular motion in an effort to recreate more normal shoulder kinematics. This scapular stabilization procedure is beneficial over formal scapulothoracic fusion in that there is no need to achieve an osseous fusion; therefore, postoperative immobilization time is drastically diminished and formal therapy can be initiated in as few as 3 to 7 days. We believe fusionless scapular stabilization is a viable option for improving appearance, pain, and upper extremity function in patients with high-level tetraplegia and scapular instability.