Understanding the management of brachial plexus birth palsy and the roles of nerve transfers

INTRODUCTION

Nerve damage is common morbidity across the global population and is a particularly common complication in birth. Nerve transfers as a treatment for such disorders have seen an increase in use as well as an increase in the number of techniques in recent years. However, there is a lack of concise literature and standardization regarding best practice protocols, as proven by evidence/research. Much of the evidence seen in the literature is scattered, and while valid, the magnitude of variety in opinion on the best practices indicates a need for a single concise source serving as a review of the most up-to-date evidence on the topic. Neonatal brachial plexus injury (NBPP) or Brachial plexus birth palsy (BPBP) presents as weakness or paralysis of the upper extremity at birth. Early recognition and management in the newborn period can significantly impact outcomes. NBPP is not uncommon, occurring in approximately 1-4 children per 1,000 live births^[1]. Despite advancements in obstetric care, this incidence seems to be increasing, due to increasing birth weights^[2].

While many of the existing management options provide good clinical evidence, there is a great lack of clear and available information regarding long-term outcomes and the overall morbidity from complications. This is compounded by the lack of good outcome measures to examine patients and the nerve transfer patient population in the long term. Our hope is that our chapter provides more extensive options for examining long-term clinical outcomes in the case of nerve transfers.

The initial management of NBPP focuses on conservative management. Physical therapy is instituted immediately once a diagnosis is made in an attempt to recover flexion. In cases where spontaneous recovery does not occur after conservative management, surgical intervention may be warranted at 3-8 months of age to prevent further nerve damage and permeant sequelae such as strength loss, decreased size of the extremity, and range of motion deficits.

Risk factors primarily include shoulder dystocia, macrosomia, and maternal diabetes or maternal weight gain. Affected roots commonly involve C5 or C6 (upper plexus palsy), but occasionally involve C7 (extended upper plexus palsy), or even C8 and T1 (total/global palsy). Lower plexus palsy (C8 and T1) has been reported in the literature, but is usually associated with total brachial plexus injuries with avulsion of the lower roots and subsequent recovery in the upper roots [Table 1].

There is an abundance of literature and evidence regarding the use of nerve transfers in the setting of both plexus palsy. However, there is a lack of conciseness in terms of cumulative review of the evidence in order to provide best practice guidelines, and as such, there is a lack of standardization or protocols for nerve transfers in the setting of birth plexus palsy. Our aim is to provide a review of the most up-to-date evidence regarding nerve transfers in the setting of birth plexus palsy, hopefully allowing for a more concise view of the evidence as a whole.

INDICATIONS

Plexus injuries can be self-limiting (80%-90%) and resolve over the course of the first few weeks of life without significant long-term morbidity^[3]. Physical therapy is instituted immediately after birth and continues as part of the multidisciplinary care plan. Recovering elbow flexion by two to three months is a positive prognostic sign overall, including as a sign of potential recovery at the shoulder. When spontaneous or complete recovery does not occur, surgical intervention is required to prevent or limit permanent sequelae from denervation, such as decreased strength, decreased size and girth of the extremity, range of motion limitations, joint dysplasia, joint contractures, or persistent neurologic deficits^[4]. Shoulder anatomy and function can be rather challenging to reconstruct, especially shoulder abduction, for which there are no good secondary options. Abnormal muscle innervation can lead to abnormal muscle development, changes to biomechanics, and joint development, with worsening progression over time, resulting in stiffness of the glenohumeral joint, with compensatory winging of the scapula. Elbow flexion contractures are also extremely common^[5].

ASSESSMENT

Based on serial physical examinations within the first few months of life, which can be challenging in a newborn, the decision is made on whether surgical intervention is indicated. In addition to general medical clearance, imaging and nerve studies can be helpful, particularly when root avulsion is suspected. In root ruptures (postganglionic), neuromas develop. In mild forms, some conduction through the neuroma may permit recovery to near-normal function; in severe forms, extensive paralysis and stiffness can develop. In root avulsions (preganglionic), no improvement is expected, and there is little value in delaying operative reconstruction.

Computed Tomography (CT) Myelography or, more likely, magnetic resonance (MR) myelography has a high (> 90%) true positive rate for diagnosing avulsion injuries^[6-8]. In 2017, in a small cohort, Bauer et al. described a new technique using three-dimensional (3D) volumetric proton density MRI to evaluate nerve roots in neonates without the need for radiation, contrast agents, or sedation/anesthetic risk^[9]. The sequence allowed visualization of preganglionic pseudomeningoceles with and without nerve root avulsion, postganglionic neuromas, and nerve root thickening, aiming to grade the severity of injury earlier than serial clinical examinations. Ultrasound alone, even high-resolution sonography, is not reliably comprehensive enough to be used alone, but can be used as an adjunct to MRI^[10]. Electrodiagnostic studies (EDS) can be used when there is doubt surrounding avulsion injuries, but because they can sometimes be misleading, EDS are not used for prognosis^[11]. Diaphragmatic ultrasound examination should also be performed to assess whether cervical nerve injury has affected diaphragmatic innervation and could impact safety for surgery.

OPTIONS FOR SURGICAL INTERVENTION

Up to a third of patients with birth palsy may require surgical intervention^[12]. The approach to surgical repair varies depending on the surgeon and the nature of the injury. The timing of surgery is controversial but is typically done between 3 to 8 months of age. Options include neuroma resection and primary nerve grafting and/or nerve transfers, which can be intra- or extra-plexal, as adjunctive or standalone procedures^[13].

When limited proximal nerve roots are available for nerve grafting, or when too many nerve roots are involved, exceeding the supply of nerve graft, extra-plexal nerve transfers are then indicated^[14,15] nerve transfer in BPBP follows all the traditional principles- selection depends on cataloging what nerves are deficient, what nerves are available/expendable, their anatomic proximity, and agonist function (caliber, motor vs. sensory function)^[16-18]. Compared to excision and grafting alone, by performing transfers as close as possible to the motor endplates, nerve transfers bring more specific axonal inflow (pure motor donors), and offer quicker reinnervation, allowing for faster recovery. While early intervention is preferred (between 3 and 6 months for most authors), older children presenting late (> 1 year since birth) with partial nerve recovery may still potentially benefit from nerve transfers, which can be offered to late-presenting infants^[15,19].

UPPER PLEXUS PALSY (C5-6) AND EXTENDED UPPER PLEXUS PALSY (C5-7)

Upper plexus palsies most likely occur as ruptures. When occurring as sequelae of breech presentations, there should be a high index of suspicion for nerve avulsion and the need for earlier intervention. In managing upper plexus palsy, establishing the stability of the shoulder is of paramount importance^[20].

The shoulder is a complex joint and has the most range of motion of any joint in the body. It is believed to have played an important role in evolution, allowing for vertical climbing and, with the ability to carry things, creating an incentive for bipedal locomotion. In order to position the hand in a wide range of space, the shoulder requires stability and mobility, conferred by the glenohumeral joint, the scapulothoracic-gliding plane, and the muscles that surround it. The rotator cuff muscles and deltoid are innervated by branches of C5-6: the suprascapular nerve (supraspinatus, infraspinatus), upper and lower subscapular nerves (subscapularis), and axillary nerve (teres minor, deltoid)^[21]. The supraspinatus and deltoid are abductors, the infraspinatus and teres minor external rotators, and the subscapularis is an internal rotator.

As such, the targets for nerve transfer are the suprascapular nerve and/or the axillary nerve. Kozin has previously discussed his indications, with a preference to target both nerves for flail shoulders and isolated nerve transfer to the suprascapular nerve when adequate deltoid function exists without external rotation^[22,23]. The most common options to reconstruct these nerves are the spinal accessory to suprascapular nerve transfer and the radial to axillary nerve transfer. Specifically, when addressing the suprascapular nerve, some have recommended decompressing the suprascapular nerve at the suprascapular notch to obviate the double crush phenomenon^[24,25]. As for the axillary nerve, the long or lateral head of the triceps branch of the radial nerve is transferred to the anterior division of the axillary nerve (Somsak transfer, Figure 1)^[26].

Figure 1. Somsak transfer: Radial nerve to axillary nerve transfer. (Photos courtesy of Dr. Gharboui, senior author).

Previous studies comparing grafting vs. transfer for shoulder reconstruction showed no significant difference, but patients were not always matched based on injury severity^[27]. In a 2020 multi-center, retrospective cohort study, Manske et al. matched Narakas types [Table 2], preoperative Active Movement Scale (AMS) scores, and shoulder subluxation to compare outcomes after grafting vs. transfer^[28-30]. They found regardless of surgical technique, shoulder external rotation recovery was disappointing, but more children undergoing spinal accessory nerve (SAN) to suprascapular nerve (SSN) transfers achieved functional shoulder external rotation (AMS > 5) and fewer secondary shoulder procedures, such as tendon transfers or osteotomies. They concluded SAN to SSN transfer should be performed in all infants undergoing brachial plexus reconstruction. These results were corroborated in a 2021 meta-analysis^[31]. Likewise, Segal et al. noted regardless of starting point, SAN-SSN transfer improved shoulder function^[32].

Subsequent to shoulder stability, elbow motion is critical to upper extremity function and warrants high-priority consideration during nerve reconstruction^[16]. The elbow is involved in forearm flexion, extension, supination, and pronation. Not only is the basic function of “hand-to-mouth” essential (“cookie test”), the absence of anti-gravity elbow flexion at key milestones (3, 5-6, or 9 months) is a primary indicator of poor functional recovery and the need for surgical intervention^[33]. For elbow flexion, the targets for reconstruction are the brachialis muscle (the primary elbow flexor) and the biceps muscle (the primary forearm supinator and a secondary elbow flexor). The musculocutaneous nerve (C5-7) innervates both muscles, and the brachialis receives an additional contribution from the radial nerve (C5-6).

Donor nerve selection is dependent upon the extent of the palsy: For infants with upper (C5-6)/middle trunk (C7) deficits and an intact lower trunk (C8-T1), the surgeon can use local ulnar and/or median nerve donors to the musculocutaneous nerve.

Specifically, the Oberlin procedure is one of the most commonly performed transfers for recovery of elbow flexion. The ulnar nerve fascicles to the Flexor Carpi Ulnaris (FCU) can be transferred to the biceps branch (Oberlin transfer, Figure 2), and the median nerve fascicles to the Flexor Carpi Radialis (FCR) can be transferred to the brachialis branch^[34,35]. Other commonly used options include the intercostal nerves [Figure 3], thoracodorsal nerve, medial pectoral nerve, and long thoracic nerve. The phrenic nerve, routinely used in adults, is rarely used in obstetric cases. Finally, ipsilateral and contralateral C7 roots allow the addition of a large number of axons without deficits for the donor site^[36]. A review of the literature finds small sample sizes, varying outcomes, and inconsistent comparison to traditional nerve grafting^[37]. The consensus is that the procedure is safe and, sometimes, effective in restoring elbow flexion^[38-41]. Compared to traditional nerve grafting alone, O’Grady et al. have reported triple nerve transfer with nerve grafting to have better functional shoulder external rotation and forearm supination, faster recovery, and lower cost^[42] [Figure 4].

Figure 2. Oberlin Procedure for reconstruction of musculocutaneous nerve for elbow flexion: (A) Ulnar nerve fascicle to FCU transfer to brachialis branch. (B) Double Oberlin procedure, with ulnar nerve to brachialis, and median nerve fascicles to FCR transferred to biceps branch. (C) Elbow flexion after Oberlin procedure. (Photos courtesy of Dr. Gharboui, senior author).

Figure 3. Intercostal nerves identified as donors in preparation for nerve transfer. (Photos courtesy of Dr. Gharboui, senior author).

Figure 4. Distal nerve transfers. (Top) Posterior approach transferring a distal branch of the spinal accessory nerve to the suprascapular nerve and also releasing the scapular ligament. (Center) Transferring a motor branch of the radial nerve to the lateral triceps to reinnervate the axillary nerve close to its neuromuscular junction in the deltoid muscle. (Below) Using a motor branch of the ulnar nerve to the flexor carpi ulnaris muscle to reinnervate the biceps muscle. Shared with permission by authors^[41].

TOTAL/GLOBAL PLEXUS PALSY

In global plexus palsy, not only is the shoulder and elbow compromised, but also the hand, to varying degrees (complete vs. near complete). Global plexus palsies often occur as a rupture of the upper roots and avulsion of the lower plexus. As such, with an avulsive component, surgery is to be expected within the first few months of infancy^[43]. If possible, intra-plexal transfers are performed, with the C5 root as a donor to the two lower avulsed roots, the C6 root as a donor to the lateral and posterior cords, and, if C7 is available, motor fibers can be used to neurotics C8 and T1 roots via end-to-end repair. However, in true global palsy, intra-plexal transfers are not possible.

In the unfortunate case of complete palsy, due to the likelihood of pre-ganglionic avulsions, in addition to attempting nerve grafts, extra-plexal nerve transfers are needed to restore shoulder stability and elbow flexion. Common donors in this setting include the spinal accessory nerve (transfer to suprascapular nerve, Figure 5), and intercostal nerves (3rd to 6th, transfer to the musculocutaneous, radial nerve). The unaffected, contralateral C7 root can also be used (partial/posterior division or complete)^[44-47]. The hypoglossal nerve has been used in the past, with poor functional results^[48,49]. Of note, in a 2020 series of qualitative interviews, Lanier et al. noted that most surgeons in North America avoided extraplexal donors other than the intercostal nerve (ICN) or SAN due to concerns for safety and morbidity^[50].

Figure 5. Spinal accessory nerve to suprascapular nerve transfer (SAN to SSN). (Photos courtesy of Dr. Gharboui, senior author).

Finally, though they have been described, sensory nerve transfers are generally not part of the primary brachial plexus reconstruction.

CONCLUSION

Neonatal birth plexus palsy is a varied and complicated diagnosis requiring multidisciplinary care. From physical examination and treatment to psychosocial issues and family impact on the patient and caregivers, NBPP remains a complex and challenging problem. In addition to primary nerve grafting, nerve transfers should be considered essential tools in the surgical reconstruction of NBPP. Nerve transfers allow for faster, more targeted recovery, and can be used in older children, and more research is needed to better characterize, understand, and compare outcomes.