Reconstructive approach needs to be individualized
Thorough discussion with patients is important for realistic expectation and discussions of the risk–benefit ratio
Early exploration when possible
Intraoperative monitoring is useful for decision making
Multiple approaches are available (neurolysis, nerve grafts, nerve transfers, free functioning muscles, tendon transfers, arthrodesis) and can be used in various combinations
Physiotherapy pre- and post-operatively
Both primary and secondary reconstructive procedures should be considered
Long-term follow-up is essential
Treatment of associated pain is important
Brachial plexus injuries are devastating injuries that lead to partial or total functional impairment of the upper extremity, psychological distress, and socioeconomic difficulties. Multiple factors influence the exact nature of the injury, including the mechanism of injury and associated vascular, bony or musculoskeletal injuries. These injuries are complex, diverse, and require an individualized approach; no single reconstructive procedure or set of reconstructive procedures is applicable to all patients. The time at which the evaluation for reconstructive procedures is performed as well as the nature of the brachial plexus injury will influence the available options. Treatment of these injuries has been evolving over the past several decades and outcomes have been improved by newer surgical techniques. The introduction of novel nerve transfers has improved shoulder abduction and elbow flexion, and in select cases, a combination of techniques can even give a reasonable possibility for rudimentary grasp function.
This chapter focuses on reconstructive procedures for the upper extremity most commonly used by our group. The main surgical options include primary reconstructive surgery, such as neurolysis, direct repair, nerve graft, nerve transfer (neurotization), and possibly free functioning (gracilis) muscle transfer, and secondary reconstructive procedures, such as tendon transfer, free functioning (gracilis) muscle transfer, and bony procedures. Each of these procedures is used, either alone or in combination, for particular clinical scenarios.
The indications and contraindications for the reconstruction of upper extremity function after a brachial plexus injury are summarized in Table 19.1 and can be divided into 3 concepts: who to operate, when to operate, and the priority of functions depending on the nature of the brachial plexus injury.
Who to operate;
When to operate
|No hope for spontaneous or further recovery||Sharp nerve transection||Acutely (as early as possible)|
|Blunt or ragged nerve transection||Acutely or subacutely (within 3–4 weeks)|
|Stretch injury without clinical or electrographic signs of recovery||Subacutely (between 3 to 6 months; earlier if root avulsions are suspected or demonstrated)|
|Potential for functional improvement with secondary procedures||Late presentation (>12 months) |
Incomplete recovery after spontaneous recovery or repair
Who not to operate
|Late presentation (>12 months) for primary nerve reconstruction|
|Ongoing spontaneous recovery|
|Unwillingness to undergo procedure given the risk–benefit ratio|
|Functions to prioritize|
Indications: patient selection and timing (who to operate, when to operate)
A surgical procedure to reconstruct upper extremity function is indicated in patients with brachial plexus injury when there is no hope for spontaneous recovery or for further recovery. This general principle dictates both the selection of patient and the timing of the surgery. In patients with sharp open injury, immediate exploration and repair is indicated, because there is no hope for spontaneous recovery. Similarly, patients with open blunt or ragged nerve transections have no potential for spontaneous recovery and require surgery. However, these patients are generally treated subacutely (approximately 3 to 4 weeks), to allow better definition of the zone of injury before exploration. It is important to note that brachial plexus injuries caused by gunshot wounds are considered “closed injury”, as the nerve damage is generally due to the shock wave and not to nerve transection; therefore, these patients can have spontaneous recovery and should have a period of observation.
In patients with brachial plexus stretch injuries, the same principle applies, i.e. surgery is indicated for patients with no hope of spontaneous or further recovery. However, the exact timing of surgery is more controversial. The general principle is that surgery is indicated in patients without clinical or electrophysiological evidence of recovery after a period of observation, because these patients have little potential for spontaneous or further recovery. The exact time allowed for this period of observation until there is “no evidence of recovery” is controversial and is influenced by multiple factors, including the mechanism of injury, the physical examination, the imaging findings, and the surgeon’s preference. We believe that when there is a high suspicion of root avulsion(s), early exploration and reconstruction is indicated. At our institution, we are not performing “ultra-early” surgery in these circumstances as advocated by some. Instead, we operate these patients 2 to 3 months after their injury when there is no evidence of recovery. In other patients without high suspicion of root avulsions, we routinely perform exploration between 3 to 6 months after injury when no adequate signs of recovery are present. We may be more willing to wait 5 to 6 months in patients with higher chance of spontaneous recovery, such as patients with incomplete brachial plexus palsy or complete brachial plexus palsy without evidence of root avulsion. It is possible that an early intervention may not allow sufficient time for spontaneous recovery; however, the routine use of intraoperative monitoring (such as nerve action potentials) will help determine if there is early subclinical evidence of recovery at the time of exploration. We choose this approach as long delays before exploration (after 6 months and even more after 12 months) may unnecessarily decrease the outcomes of a reconstructive procedure due to the motor end plate failure with time. For patients presenting late (i.e. more than 12 months from the injury), primary (nerve) reconstructive procedures are generally contraindicated due to significantly poorer results. However, a surgical intervention consisting of secondary reconstructive procedures, such as tendon transfer, free functioning gracilis muscle transfer, and bony procedures, may be indicated.
Contraindications (who not to operate)
The surgical approach needs to be individualized based on the patient’s specific injury, his or her needs and expectations, and the goals and priorities set by the treating team and patient. The patient should be informed about the possible surgical options, risk–benefit ratio, postoperative rehabilitation program, success rate, and long recovery time. A contraindication to surgery is unrealistic goals or unwillingness of the patient to undergo the procedure given its risk–benefit ratio. Surgery is also contraindicated in patients with spontaneous ongoing recovery, as outcomes will generally be better without surgery in these patients. Stiffness and contractures, age, medical comorbidities, associated traumatic brain injury, and associated spinal cord injury should be considered in the decision process and may need to be addressed, but they are not, generally, contraindications.
Priorities for functional reconstruction
A complete functional recovery is the ultimate goal in patients with brachial plexus injury. Unfortunately, in most patients, this goal cannot be achieved due to the severity of the injuries and the limited availability of functioning donor nerves. Our priorities for functional reconstruction is generally as follow: 1) elbow flexion, 2) shoulder stability (in particular reduction of subluxation and external rotation), 3) hand sensibility, 4) wrist and finger flexion, 5) wrist and finger extension, and 6) intrinsic hand muscle function. The indicated surgery and available options vary depending on the extent of the brachial plexus injury. In general, the brachial plexus injuries can be divided into complete (pan) plexus injuries, C5–C6 injuries, C5–C6–C7 injuries, and C8–T1 injuries. We are not discussing infraclavicular brachial plexus injury patterns, which are caused by different mechanisms and are less common in our practice.
In all these situations, if post-ganglionic neuromas conducting nerve action potentials are found at exploration, neurolysis is done; on the other hand, when nerves are found at exploration to be ruptured or to have a postganglionic neuroma not conducting a nerve action potential, interpositional grafts are typically performed between the proximal nerve and specific targets. We routinely use intraoperative electrodiagnostic assessment to confirm the proximal nerve integrity (i.e. preservation of somato-sensory and motor-evoked potentials). Generally, C5 is used to target shoulder stability/abduction (suprascapular nerve, axillary nerve, or posterior division of upper trunk), C6 to target elbow flexion (anterior division of upper trunk, or musculocutaneous nerve), and C7 to target elbow extension (posterior division of middle trunk, or radial nerve) ( Figure 19.1 ). When proximal nerves are found to be avulsed or have significant preganglionic injuries, different options are available depending on the brachial plexus injury pattern as described below (see Table 19.2 ).
|Type of brachial plexus injury||Nerve axon donor||Common targets|
|Any injury with neuroma-in-continuity conducting NAP||Neurolysis|
|Any injury with nerve rupture or post-ganglionic neuromas not conducting NAP and good proximal nerves (preserved SSEP and MEP)||Direct repair or repair with graft if possible|
|C5 nerve||Shoulder stability, abduction, and external rotation : |
Posterior division of UT
|C6 nerve||Elbow flexion : |
Anterior division of UT
|C7 nerve||Elbow extension : |
Posterior division of MT
|Complete C5–T1 panplexus injury||Nerve transfers: |
|Suprascapular nerve |
Free functioning muscle
|Motor ICN||Musculocuatenous nerve |
Motor branch to the biceps
|Sensory ICN||Lateral cord contribution to the median nerve|
|Other potential donors: |
Contralateral C7 nerve
|C5–C6–C7 and C5–C6 |
|Ulnar nerve fascicle(s)||Motor branch to the biceps|
|Median nerve fascicle(s)||Motor branch to the brachialis |
Motor branch to the biceps
|Triceps motor branches from radial nerve||Axillary nerve proper |
Anterior division of axillary nerve
|Other potential donors: |
Medial pectoral nerves
Ipsilateral C7 nerve
|Tendon transfers |
Selective arthrodesis of joints
Free functioning muscle transfer
Complete (pan) brachial plexus injury
In patients with complete brachial plexus injury, useful recovery may still be obtained despite obvious limitations in the availability of functional donor nerves. If no proximal nerve is available, a common strategy consists of transferring the spinal accessory nerve to the suprascapular nerve (for shoulder stability), motor intercostal nerves to the musculocutaneous nerve or to the motor branch of the biceps (for elbow flexion), and sensory intercostals nerves to the lateral cord contribution of median nerve (for hand sensibility). When proximal nerves are available, a reconstructive strategy using multiple nerve transfers and nerve grafts may allow the possibility of regaining additional useful functions, including even rudimentary hand function. Additional donors have also been described, such as the phrenic nerve and contralateral C7 nerve. Our group stopped using the phrenic nerve, in view of our population of patients (often overweight and at risk to develop cardio-respiratory disease in their lifetime) and of the occurrence of respiratory function compromise in one patient. After review of our results from hemi-contralateral C7 nerve transfer for hand function, we now seldom choose this approach to regain hand function because of the poor functional grasp that was obtained. However, we currently often neurotize a free functioning gracilis muscle for elbow flexion in the acute stage to increase the chances of achieving adequate elbow flexion. If the triceps can be reinnervated (such as by nerve grafts from C7 or spinal accessory nerve), we consider using a free functioning muscle for both elbow flexion and finger flexion by prolonging the distal gracilis tendon to the flexor digitorum profundus and flexor pollicis longus tendons.
C5–C6 brachial plexus injury
When preganglionic injuries of the C5 and C6 nerves are present, several nerve transfer options exist to restore shoulder stability and elbow flexion. Potential donor nerves include the spinal accessory nerve, the triceps motor branches of the radial nerve, the medial pectoral nerves, the ulnar or median nerve, the intercostals nerves, the phrenic nerve, the ipsilateral and the contralateral C7 root. A common treatment approach includes the transfer of ulnar nerve fascicle(s) to the biceps motor branch of the musculocutaneous nerve to restore elbow flexion (Oberlin’s transfer), and a spinal accessory nerve transfer to the suprascapular nerve. The Oberlin’s transfer, with its good results, is now the procedure of choice for biceps reinnervation in preference to intercostal or spinal accessory nerve transfers. In our group, we attempt dual reinnervation when possible of both the shoulder (supraspinatus and deltoid) and elbow (biceps and brachialis muscles). The deltoid can be reinnervated using a triceps motor branch of the radial nerve to the axillary nerve and the brachialis muscle can be reinnervated with a fascicle from the median nerve.
In patients with postganglionic ruptures or postganglionic neuromas not conducting nerve action potentials, both nerve grafting and nerve transfers can be done and there is controversy as to which approach is preferable. We usually prefer to use nerve grafting from C5 for shoulder function if the repair is done early and the proximal nerve has a good macroscopic appearance as well as good somato-sensory and motor-evoked potentials intra-operatively; we typically use double fascicular transfer for elbow flexion. If the exploration and repair is done late, or there are uncertainties about the integrity of the proximal nerves, then we prefer to perform double nerve transfers for shoulder function and elbow flexion.
C5–C6–C7 brachial plexus injury
When C7 is involved and the triceps muscle is weak, the triceps motor branch to axillary nerve transfer cannot be done. In that situation, the same strategy as described for the C5–C6 brachial plexus injury can be used, but intercostals nerves may be used to reinnervate the axillary nerve if necessary. If the wrist and finger extension are affected, we generally reinnervate the triceps muscle with nerve grafts or nerve transfers and the distal function is addressed by tendon transfers either simultaneously or at a subsequent surgery.
C8–T1 brachial plexus injury
We treat lower trunk brachial plexus injuries by a combination of tendon transfers, selective arthrodesis of joints, and/or free functioning muscle transfer. The approaches to restore hand function are described in greater detail in Tu and Chung’s chapter on hand reconstruction. Tendon transfers, such as extensor carpi radialis longus to flexor digitorum profundus and brachioradialis to flexor pollicis longus, can be performed to restore finger flexion and pinch. Selective arthrodesis of the interphalangeal and carpometacarpal joint of the thumb can also improve both grasp and pinch. As the results of nerve repair are poor in lower trunk injuries because of the distance and time for reinnervation, we generally do not attempt nerve surgery in these patients. Others have suggested distal nerve transfers using the supinator motor branch to the posterior interosseous nerve, the brachialis branch of musculocutaneous nerve to the anterior interosseous median nerve fascicles, or the radial nerve branch to the extensor carpi radialis brevis muscle to the anterior interosseous nerve. At present, when good tendon transfer and arthrodesis options exist, we prefer these approaches.
Before attempting any reconstructive procedure for patients with brachial plexus injuries, a multidisciplinary evaluation is necessary. This multidisciplinary approach should be used from the time of the first evaluation, intraoperatively, as well as in the immediate and long-term post-operative period. The multidisciplinary team consists of a neurologist, a neurosurgeon, an orthopedist, and/or a plastic surgeon, a pain specialist, a physiotherapist, and an occupational therapist.
The importance of the initial evaluation, including a detailed history and physical examination (neurologic, musculoskeletal and vascular), as well as review of the nerve conduction, electromyographic, and imaging studies was discussed in previous chapters and will not be reviewed here. However, a few particular pre-operative considerations are specific to our reconstructive approaches and need to be discussed.
When intercostal nerve transfers are considered, dynamic (inspiratory and expiratory) chest radiography is useful to assess phrenic nerve function and prior rib fractures. If the patient has a phrenic nerve dysfunction and/or a history of rib fractures, pneumothorax, hemothorax, pulmonary contusions, or respiratory compromise, pulmonary function tests (PFT) should be considered to better characterize the pulmonary function. These evaluations, with an assessment by a pulmonologist if necessary, will help determine if intercostal nerves may be utilized as donor nerves for transfers. Generally, even if a phrenic nerve dysfunction is present, the patient will tolerate up to 4 or 5 intercostal nerve transfers, unless his pre-operative PFT are very low (i.e. less than 40% of expected, but no definite criteria exist). The patient’s body habitus and smoking habits should also be taken into consideration before the use of intercostal nerves, especially in the context of a phrenic nerve dysfunction. The presence of rib fractures does not preclude the use of intercostal nerves. In our experience, there is a 10% risk that the intercostal nerve at the level of the rib fracture will be non-functional (i.e. will not stimulate) and will not be usable as a donor. It is important to counsel patients appropriately and make them aware of the possibility that the nerves may not be functional; this is especially important if a patient had multiple rib fractures at the time of trauma. Intra-operatively, direct stimulation is the usual definitive method to assess function of the intercostal nerves.
When the spinal accessory nerve is considered for transfer, its function should be evaluated pre-operatively. The function can be assessed clinically by evaluating trapezius muscle function (shoulder shrug), comparing its strength to the opposite side. However, since this method is not sensitive, electromyographic study of the trapezius demonstrating normal motor unit potentials and absence of abnormal insertional activity is useful in documenting an intact spinal accessory nerve. If the spinal accessory nerve has been injured at the time of the trauma, it may still be explored and utilized if there is good stimulation intraoperatively, knowing that the nerve may not be completely normal.
Not all patients are appropriate candidates for free functional (gracilis) muscle transfer (FFMT). A few particular pre-operative considerations are specific to this particular reconstructive approach and need to be discussed. First, a viable arterial and venous supply must be present. Second, an expendable motor nerve must be present near the transferred muscle’s neurovascular pedicle to reinnervate the transferred muscle. The third and fourth intercostal motor nerves and the spinal accessory nerves are usually good candidates. The patients considered for FFMT must understand the inherent risks, including flap loss, and the expected outcomes. They must be prepared to undergo a major operation with hospitalization and extensive rehabilitation and understand the goals of the surgery.
Supple joints are preferable before function is restored. If full passive range of motion is lacking, therapy or capsular release may be indicated. When restoring elbow flexion, especially if a FFMT is considered, shoulder stability should be discussed. Although shoulder stability is not a pre-requisite, if the shoulder is subluxed, a FFMT anchored to the clavicle and acromion will initially use some of its force to reduce the subluxed shoulder; this will in turn diminished elbow flexion strength of the transferred free muscle.
The upper extremity recipient bed for the transferred muscle must provide adequate skin coverage and ideally be free of adherent scar, to allow for muscle and tendon gliding. However, this may not always be possible in cases of brachial plexus injury. If the skin is significantly scarred, it may be difficult to cover the muscle transfer. Transferring a skin paddle with the muscle may help, but this is reliable only proximally for the gracilis myocutaneous flap and coverage of the distal end of the muscle and tendon may remain an issue. Using the muscle transfer for the “dual purpose” of providing elbow flexion and coverage of exposed structures or hardware is not recommended as it will lead to inevitable scarring and result in poor outcomes.
Another important consideration is the assessment of the vasculature to the injured upper extremity, especially when a FFMT is considered. In these patients, especially when the first rib has been fractured, the axillary and subclavian vessels may have sustained an injury. In the authors’ experience, a normal peripheral pulse does not preclude vascular injury, especially in the major branches. Therefore, we generally evaluate the vasculature using MR angiography (MRA). Special attention is given to the thoracoacromial trunk, our preferred source of arterial supply for the FFMT. Other options to formally evaluate the vasculature include CT angiography or conventional angiography. Any of these vascular studies will also assess post-traumatic stenosis and the patency of a previous vascular repair or reconstruction. In cases in which the thoracoacromial trunk is not intact, alternative options may be explored for the vascular anastomoses, including vein graft, arteriovenous loop preparation, and end-to-side anastomosis into the axillary artery.
The procedural approach for upper extremity function reconstruction after a brachial plexus injury is highly unique. It needs to be individualized, based on the patient’s specific injury, his or her needs and expectations, and the goals and priorities set by the treating team and patient. The evaluation of the risk–benefit ratio is done for each patient and the procedure is tailored to the patient’s comfort level to a procedure with acceptable risks and set of goals.
Step-by-step description of all reconstructive options is beyond the scope of this chapter. However, we will briefly described the most commonly used approaches including brachial plexus exploration and neurolysis, nerve grafts, common nerve transfers (spinal accessory nerve, intercostal nerves, Oberlin procedure, and triceps branch to axillary nerve transfer), and the use of a free functioning muscle (gracilis muscle) transfer for elbow flexion. Other techniques to obtain hand function are described in Chapter 20 .
Patient preparation and positioning
Performed in the supine position under general anesthesia, the surgery is done without the use of long-acting muscle relaxant. Adequate vascular access should be obtained and an indwelling urinary catheter is inserted. The patient’s upper body is slightly elevated by the use of a flat pillow placed behind the ipsilateral scapula. To facilitate the exposure of the supraclavicular plexus, we turn the head toward the unaffected side and the neck is slightly extended. Proper padding and positioning to protect bony prominence is essential, as brachial plexus reconstructive procedures are long. Prepping and draping must allow wide exposure of the neck, shoulder, chest, and axilla; the affected arm should be able to be moved freely. Both lower extremities are also prepped and draped for sural nerve harvesting, should nerve grafts be needed. The use of an operating microscope or magnifying loupe is suggested.
If a free functioning (gracilis) muscle transfer is planned, the body core temperature should be maintained throughout the surgery as it will facilitate peripheral tissue perfusion. This can be accomplished by increasing the ambient operating room temperature, covering the exposed skin, and using warming devices (such as fluid warmers and warming blankets). All these techniques should be initiated as soon as the patient enters the operating room and continued into the post-operative period.
Supraclavicular brachial plexus exploration
The supraclavicular brachial plexus is generally exposed through a transverse, off-midline, 5 to 6 cm skin incision, parallel and approximately 2 fingerbreadths above the clavicle. The platysma muscle is incised along the skin incision and the external jugular vein is identified and preserved. Skin flaps comprising the platysma muscle are carefully raised both superiorly and inferiorly. The sternocleidomastoid muscle is exposed and retracted medially. The dissection is continued into the pre-scalene area where the superficial layer of the deep cervical fascia is opened. The cervical fat pad is then opened and retracted. The tissues are opened to the superior border of the clavicle. The omohyoid muscle is identified, divided, and each portion is retracted medially and laterally. The transverse cervical vessels are ligated, and divided.
The upper trunk is usually the first brachial plexus element to be visualized. Medially, it is important to identify the phrenic nerve which runs on the anterior surface of the anterior scalene muscle, parallel to its muscle fibers. The use of an intraoperative nerve stimulator can confirm its identification, except in cases in which the phrenic nerve function is impaired (e.g. very proximal or pre-ganglionic C5 nerve injury). Once identified, the phrenic nerve is a helpful landmark as it can be traced proximally up to the C5 phrenic contribution, confirming the identification of the C5 nerve. Dissection medial to the phrenic may lead to iatrogenic injury of vital neck structures, but may be necessary in cases of proximal C5 injury. The phrenic nerve should be mobilized with care; mobilization will allow safe resection of the anterior scalene muscle necessary to expose the proximal cervical nerves. The brachial plexus exploration is then continued. The upper trunk is exposed proximally to its C5 and C6 nerve origin. The C5 nerve is usually smaller than the C6 nerve. The C6 nerve is inferior, more medial, and less vertical in direction than the C5 nerve. The upper trunk is then exposed distally to identify its anterior and posterior division as well as the origin of the suprascapular nerve. Attention is then directed to the C7 nerve and middle trunk, which is more medial, more posterior, and horizontal in direction than the C6 nerve. Even more inferior and posterior are the C8 and T1 nerves which form the lower trunk. These lower nerves are closely associated with the subclavian artery. Occasionally, mobilization or osteotomy of the clavicle is needed for better exposure of the lower two roots. In most cases however, the C8 and T1 nerves are not specifically exposed as repair is rarely, if ever, indicated in our opinion, given the poor outcome of these lower nerve injuries.
Infraclavicular brachial plexus exploration
When indicated, the infraclavicular brachial plexus exploration is done via a deltopectoral interval incision, from the clavicle to the anterior axillary line. The deltopectoral interval is opened, preserving the cephalic vein, which is retracted laterally. The pectoralis minor muscle is exposed and its tendinous attachment at the coracoid process is divided, tagged, and retracted; it will be reattached at the time of closure. The distal part of the brachial plexus is located immediately deep to the pectoralis minor. The lateral cord will be the first structure encountered. The other elements of the infraclavicular plexus as well as the associated vessels are then dissected and exposed. Once the supra- and infra-clavicular brachial plexus are exposed, the retroclavicular area can be addressed. A tunnel is made from the neck under the clavicle to the deltopectoral area by a gloved finger. The most lateral part of the pectoralis major muscle clavicular attachment is detached. The subclavius muscle is divided. Patients with a healed clavicle fracture require special attention; in these cases, blunt dissection underneath the clavicle can be difficult because of the fibrotic nature of the soft tissue. The risk of major vascular and pleural damage must be considered, especially if the lower brachial plexus elements are exposed medially. Once the middle part of the clavicle is freed from the surrounding soft tissue, it can be mobilized upward or downward by manual retraction. If clavicle osteomy is needed, pre-drilling of the screws for a plate fixation should be made before division of the clavicle. Once all brachial plexus elements have been exposed, intra-operative electrodiagnostic assessment ( Figure 19.2 ) can be performed. We recommend and routinely use intraoperative electrodiagnostic assessment. Nerve action potentials are utilized to assess post-ganglionic neuromas in continuity. We also use somato-sensory and motor-evoked potentials to assess for pre-ganglionic injury. The surgical strategies are then tailored to the intra-operative findings.