Functional outcome of spinal accessory nerve transfer to the suprascapular nerve to restore shoulder function: Results in upper and complete traumatic brachial plexus palsy in adults
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.250708
Source of Support: None, Conflict of Interest: None
Keywords: Brachial plexus palsy, brachial plexus surgery, nerve transfers
Shoulder stability, abduction and external rotation are vital for the performance of usual daily tasks, and important for optimizing more distal functions. In consequence, the restoration of these functions is one of the priorities in the surgical reconstruction of the injured brachial plexus.
To restore shoulder function, the spinal accessory nerve has been transferred to the suprascapular nerve (SASNT) for many years with highly variable results, ranging from 0 to 100% success for shoulder abduction and from 0 to 86.8% for external rotation.,,,,,,,,,,,,,,,,,,,, Arguments for using the spinal accessory nerve as a donor for this transfer rely on four points: (1) It is a nerve almost entirely motor; (2) it has a high number of motor fibers (roughly 1700); (3) its functional characteristics are relatively similar to those of the suprascapular nerve; (3) it is in close proximity to the suprascapular nerve; and, (4) its integrity is preserved in almost 95% of patients with traumatic brachial plexus palsy.
It is mentioned in the literature that SASNT yields better results with upper-type versus total plexus palsies. As our results in complete palsies appeared to be quite acceptable, at least for shoulder abduction, we decided to critically compare the functional outcomes in the shoulder following SASNT performed for both partial (upper) and complete traumatic lesions of the brachial plexus in adults.
After approval from the ethics committee of the University of São Paulo Medical School, the medical records of adult patients with traumatic brachial plexus injury operated upon within the Peripheral Nerve Surgery Unit over a period of nine years (2004-2012) were retrospectively reviewed, and all patients undergoing SASNT identified. The inclusion criteria for the study were: (1) the patients' age ≥18 years old; (2) lack of shoulder abduction and external rotation; (3) injury-to-surgery interval of less than 12 months; (4) normal trapezius muscle function; (5) lesion affecting C5, C6 in partial and complete plexus injuries; (6) no passive limitations in shoulder joint motion; (7) no secondary reconstruction in the shoulder region; and, (8) follow up of at least 18 months. Patients also submitted to a second transfer, linking a branch of the radial nerve to the triceps, to the axillary nerve (Somsak's procedure), to increase the shoulder abduction, were discarded as well.
The clinical assessments included measurement of the range of motion (ROM) of the shoulder via goniometry with the patient standing. To evaluate the range of shoulder abduction, the patient was asked to move the extended arm laterally against gravity and the angle formed between the arm axis and thorax was measured. When 30° of shoulder abduction is achieved, shoulder subluxation is usually corrected and it is possible to move the glenohumeral joint. Hence, this range of shoulder abduction and beyond were considered as a good functional result.,, To measure the range of external rotation (the angle between the forearm and abdomen), the patient was instructed to start movement with the shoulder fully internally rotated and the forearm actively or passively flexed 90 degrees and placed transversally over the abdomen. An external rotation of at least 55° is necessary to allow the arm to be away from the body for adequate elbow flexion to perform most of the upper-limb activities of daily living., This range of external rotation and beyond were considered a good functional result. Data concerning the patients' age and gender, body mass index (BMI), the mechanism of injury, passive range of motion of the affected shoulder, and the injury-to-surgery interval were also collected from medical files.
A 10-cm transverse incision was made, 1 cm above and parallel to the clavicle, extending laterally from the posterior border of the sternocleidomastoid to the acromioclavicular joint. Division of the platysma allowed supraclavicular nerve identification and preservation. Sometimes, when the insertion of the trapezius in the clavicle extends too medially, it can be detached, retracted posteriorly, and reinserted at the end of the surgery. The spinal accessory nerve could be identified just deep to the superolateral margin of the trapezius muscle near the clavicle with the transverse cervical artery and vein as landmarks. A nerve stimulator could be helpful in this dissection and should be used to confirm that the nerve is functional. The proximal branches to the upper part of the trapezius were preserved to avoid massive denervation of the trapezius muscle, whereas the terminal branch of the nerve was dissected and divided as far distally as possible, near to its entrance in the muscle, after the origin of 1 or 2 branches for the lowest part of the muscle. The muscle was reflected proximally. Next, the suprascapular nerve was identified arising from the superolateral aspect of the upper trunk, that is often involved in the upper trunk neuroma. It was transected in a healthy area at a point as close as possible to its division from the upper trunk. Eventually, this nerve can be displaced further distally to a retroclavicular or infraclavicular level owing to the downward shifting of the avulsed C5 and C6 roots. The dissection and mobilization of both the nerves permitted a tension-free coaptation, under the operating microscope in the supraclavicular fossa, with epineural sutures using 10-0 monofilament nylon and fibrin glue.
For descriptive purposes, categorical variables were presented as absolute and relative frequencies. Quantitative variables were assessed for normality through the skewness and kurtosis values as well as distributional graphical methods. Age and injury-surgery interval had a normal distribution and were presented as means, but the ranges of motion (ROM) values were non-parametric and were presented as medians and quartiles. The ranges of motion related to abduction and external rotation were compared via the Mann-Whitney U test. The percentages of patients with a favorable outcome (defined as the abduction ROM ≥30 degrees and the external rotation ROM ≥55 degrees) were compared by the chi-square test. All tests were two-tailed and P values <0.05 were considered statistically significant. All analyses were conducted using the software Statistical Package for the Social Sciences (SPSS Statistics, Version 24.0. Armonk, NY: IBM Corp.).
The sample characteristics of the study are summarized in [Table 1].
Spinal accessory to suprascapular nerve transfer was performed in 76 patients who met the study's inclusion criteria: 23 (30.2%) cases of upper-plexus injuries (C5, C6) and 53 (69.7%) cases of complete brachial plexus palsy (C5-T1). The results are summarized in [Table 2].
The average age of patients with upper plexus palsy was 24.6 ± 8.4 years. A good shoulder abduction was achieved in 15 (65.2%) patients and a good external rotation in 5 (21.7%) patients in this group. In those patients with a good recovery, the average range of motion (ROM) was 53° (range 30-90°) for shoulder abduction and 71.2° (range 55-90°) for external rotation. Two patients who achieved a good result in their abduction movement and 3 who did not, had a body mass index (BMI) above 30. In relation to the external shoulder rotation, the result was bad in all the 5 patients.
Patients with complete palsy had an average age of 26.2 ± 8.9 years. Thirty-six patients (67.9%) in this group had a good recovery of their shoulder abduction with an average ROM of 30.7° (range 30-89°), but only 3 patients (5.6%) recovered a good shoulder external rotation with an average ROM of 68.3° (range 60-80°). The BMI was equal or greater than 30 in 2 patients with complete palsy and both had a good result for shoulder abduction and a bad recovery of their external rotation.
There was no difference in the outcome of abduction between the two groups. The median ROM for abduction was 30 (20-48) degrees for the upper plexus injury group and 35 (8 – 48) degrees for complete plexus palsy group (P = 0,737). The percentage of patients who achieved a ROM ≥30° was 60 (9%) and 67 (9%) (P = 0,551), respectively, in the two groups. The external rotation outcome was better in the upper plexus palsy group, who had a median ROM of 0 (range: 0-52) degrees (vs 0 [range: 0-0] degrees, P = 0,004) and 21 (7%) patients achieved a ROM ≥55° (vs 6, 0%, P = 0,099).
More than half of our patients (64.7%) were operated upon beyond six months post injury. The average interval between injury and surgery among patients with a good functional result was 6.8 ± 2.8 months, versus 8.0 ± 3.4 months among those with a poor outcome. When we compared partial and complete lesions, we found an interval from injury/surgery of 6.7 ± 2.8 versus 7.2 ± 3.1 months among those with a good outcome, and 8.1 ± 3.4 versus 7.7 ± 3.2 among those with a poor outcome. None of these differences were statistically significant.
When we divided our patients into those less than versus equal to or greater than 40 years old, we observed that 68.9% of patients under 40 years of age recovered a good abduction, while 16.4% achieved a good external rotation. Among patients 40 years and older, the corresponding percentages were 41.0% and 5.8%.
The spinal accessory nerve is frequently used as a donor of axons during nerve transfers. Despite the wide variation of results, its transfer to the suprascapular nerve is considered the “gold standard” for restoring shoulder abduction and external rotation.
The supraspinatus, infraspinatus, and deltoid muscles are involved in glenohumeral motion, whereas the trapezius, levator scapulae, rhomboid, and serratus anterior muscles control the scapular thoracic motion. Although subject to criticism, the reported angles of arm abduction by most authors, including us, is a combination of both the movements. Malessy et al., in a more critical analysis, considered only the active glenohumeral shoulder function and observed a poor suprascapular neurotization with the SASNT.
The reinnervation of the suprascapular nerve by a SASNT offers a good chance of restoring shoulder stability in many patients, producing adequate shoulder abduction and, in some patients, even restoring some degree of external rotation.,, Considering shoulder abduction, the percentage of patients achieving a good result in the literature averages 66.4%, which compares favorably with the 67.1% we achieved across the two treatment groups in our series. There was no significant difference between the results for shoulder abduction in the two treatment groups. The restoration of external rotation with SASNT is usually inferior to that of shoulder abduction.,,,, The potential reasons for this include: (1) the difference in the number of myelinated fibers from the donor (accessory nerve – 1700 fibers) to the receptor (suprascapular nerve – 3800 fibers) nerve; (2) the potential for the presence of two-levels of lesions; as we usually do not dissect the nerve distally, any injury affecting the suprascapular nerve after its branch to the supraspinatus, consequent to a scapular fracture,, for example, had the potential to be overlooked; and, (3) more intense reinnervation of the first muscle reinnervated, the supraspinatus., The recovery of external rotation in our series was considerably lower than that reported in the literature (10.5% versus 37.4%). There are three possible explanations for this: (1) The large number of patients with complete lesions (70%) in our series; (2) many series considered any recovered ROM (less than 55°) as a good result; and,(3) some undetectable technical problem might have occurred during the procedure.
Preservation of the long thoracic nerve in partial lesions assures the presence of scapular thoracic motion and scapular stabilization during active shoulder abduction and should be considered essential for abduction recovery after a nerve transfer.,, According to some authors, that is the reason why SASNT yields better results with upper-type versus total plexus palsies. However, as noted previously, the range of abduction recovered with complete lesions, and consequently with palsy of the serratus anterior muscle, was not affected. These results were very similar to the results in upper-plexus palsies – 65.2% and in complete lesions (67.9%).
Many authors,,,, have emphasized that the simultaneous neurotization of suprascapular and axillary nerves optimizes outcomes related to shoulder abduction. As we wanted to assess the results of isolated SASNT, only the cases of upper plexus palsy operated before the widespread acceptance of the Somsak's procedure were included in the study.
Body mass index (BMI) is defined as the individual's body mass divided by the square of their height, and is strongly correlated with the weight of the extremities. Therefore, it can be used to study the relationship between the upper extremity weight and the results of brachial plexus surgery. Studying 18 patients submitted to spinal accessory-to-suprascapular nerve transfer, Socolovsky et al., concluded that a high BMI is a negative predictor of outcome. Although only 7 patients (9.2%) in our series had a BMI equal or greater than 30, our results partially support their conclusion. Some patients (n = 4) had a functional recovery of shoulder abduction, but all had a bad shoulder external rotation recovery.
For brachial plexus surgeries, poor results are usually associated with a delay between the initial injury and surgery. The results are also inversely proportional to the age of the patient. In our study, comparing outcomes between patients operated upon within, versus beyond six months post injury, revealed no statistically significant difference in the outcome, while patients younger than 40 years old had a trend towards better results.
According to our results, spinal accessory-to-suprascapular nerve transfer is quite consistent as a procedure to achieve functional recovery of shoulder abduction after a partial or complete plexus injury, with both situations having very similar results. However, the outcome of external rotation was quite disappointing, with the percentage of good results being unacceptably low across our treatment groups. It behooves surgeons to keep searching for new techniques to improve this crucial outcome.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2]