Nerve Reconstruction in Brachial Plexus Birth Injuries

Synopsis

The vast majority of brachial plexus birth injuries (BPBIs) are caused by traction to the brachial plexus during labor. The incidence of BPBI is about 1 to 2 per 1000 births. Typically, the C5 and C6 spinal nerves are affected. The prognosis is generally considered to be good, but the percentage of children with residual deficits may be as high as 20% to 30%. The neuropathophysiological basis in BPBI is that the damaged nerves are usually stretched and damaged and form a neuroma in continuity . Typically, the nerves are not completely ruptured. Even in the most severe BPBIs, at least some axons will pass through the neuroma-in-continuity and reach the tubes distal to the lesion site. These axons may be particularly prone to abnormal branching and misrouting, which may explain the typical features of co-contraction and frustration of functional regeneration. An additional factor that may reduce functional regeneration is that improper central motor programming may occur.

Surgery should be restricted to severe cases in which spontaneous restoration of function will not occur, that is, in neurotmesis or root avulsions. Selection of infants for surgery, and prediction of whether function will be best after spontaneous nerve outgrowth or after nerve reconstruction, is problematic. Results achieved by surgery are claimed to be superior to outcomes in conservatively treated subjects with equally severe lesions. A prospective randomized trial to answer the question of whether surgery improves functional outcome has in fact never been performed. During early surgical exploration the intraoperative appraisal of the neuroma-in-continuity and decision whether to cut or not can be difficult.

In the following sections, we present an overview of our current knowledge of BPBI based on our understanding of the neuropathophysiology.

Introduction

A BPBI is caused by traction to the brachial plexus during labor. ^, In the majority of cases delivery of the upper shoulder is blocked by the mother’s symphysis (shoulder dystocia). If additional traction is applied to the child’s head, the angle between neck and shoulder is forcefully widened, overstretching the ipsilateral brachial plexus.

The incidence of BPBI varies from 0.42 to 2.9 per 1000 births in prospective studies. There is a trend toward a decrease in incidence during the last 15 years. Risk factors that have been identified for the occurrence of BPBI reflect the disproportion between the child and the birth canal. The main fetal risk factor is macrosomia, which is a risk factor not only for the occurrence of BPBI but also for its neurological severity. Maternal factors include gestational diabetes and multiparity. Shoulder dystocia and assisted delivery by forceps or vacuum cup are well-known risk factors for the development of BPBI. Recently a relationship has been described between the risk and severity of a BPBI and the amount of downward traction. A less common delivery pattern concerns infants, usually with low birth weight, born in a breech position. Infants born in breech carry a high risk for the presence of root avulsions.

There is an ongoing debate whether BPBI can be prevented and whether the obstetrician can be held responsible. This debate is fed by numerous high-compensation malpractice suits. Case reports describing spontaneous deliveries without traction applied to the child during labor, with the occurrence of brachial plexus injury, have been published to acquit the obstetrician.

The upper brachial plexus is most commonly affected, resulting in paresis of the supraspinatus, infraspinatus, deltoid, and biceps muscles, as first described by Erb and Duchenne. Typically, in the C5, C6 lesion type, the affected arm rests on the surface in adduction, internal rotation, and extension. The wrist and fingers are continuously flexed when C7 is damaged as well. Hand function is additionally impaired in approximately 15% of patients. ^{, ,} An isolated injury to the lower plexus (Déjèrine–Klumpke’s type) is rare.

The traction injury may vary from neurapraxia or axonotmesis to neurotmesis and avulsion of rootlets from the spinal cord. The severity of neural damage will become clear by evaluation of recovery in the course of time, because nerve lesions of different severity initially present with the same clinical features. Neurapraxia and axonotmesis eventually result in complete recovery. Neurotmesis and root avulsion, on the other hand, result in permanent loss of arm function, and in time development of skeletal malformations, cosmetic deformities, behavioral problems, and socioeconomic limitations.

Neuropathophysiology

In BPBI infants, the damaged nerves are usually not completely ruptured, in the sense that there is a gap between two stumps. This is most likely caused by the gradual exertion of traction forces over a small distance that acts during a relatively long period and by the changes of direction of these forces during the delivery. The two crucial factors that determine good functional recovery are the number of damaged axons that successfully elongate past the lesion site, and their routing. Axonal outgrowth and restoration of connections with original motor or sensory end organs can only take place when the basal laminal tubes surrounding the axons—which are in this context the crucial anatomical structures—remain intact. The distance from the lesion site, which is in BPBI almost always at root and trunk level, to the end organ determines the length of the time interval after which recovery will take place. Proximal muscles recover, therefore, at an earlier stage than more distally located ones. Spontaneous recovery of predominantly axonotmetic BPBI is usually seen within the first 3 to 4 months of life.

When the traction lesion is severe the basal laminal tubes are ruptured, but the perineurium and epineurium remains more or less intact. Outgrowing axons will than not end up directly in any tube. Typically in BPBI, the stretched and damaged nerve forms a neuroma-in-continuity: a tangled mass of connective scar tissue and outgrowing, branching axons. The local environment encountered by the axonal growth cone may impede outgrowth and may ultimately block the restoration of axonal continuity. Even in the most severe C5/C6 BPBIs, at least some axons will pass through the neuroma-in-continuity and reach the tubes distal to the lesion site. The number of axons that will not pass the lesion site depends on the severity of the lesion, which is determined by the magnitude and angle of the exerted traction forces. There is a minimum number of axons that should reconnect with the end organs in order to regain function. In addition, for the restoration of useful function there is a minimum of axons that should be properly routed to their original end organ. We presume that those axons in the BPBI neuroma-in-continuity are particularly prone to abnormal branching and misrouting. Since the direction of outgrowth after severe lesions is essentially random, outgrowing axons growing through a neuroma-in-continuity are likely to end up in the wrong tube. In addition, axons in a neuroma-in-continuity course through multiple focal globular areas with markedly diminished myelination, this may be the pathobiologic basis for conduction blocks in patients with BPBI. Each BPBI case is unique on the axonal level, in the sense that the number of ruptured axons and basal laminal tubes differ for each intraplexal involved nerve element. This subsequently leads to the wide variety in level of functional recovery which can be found in individual cases. Branching and misrouting can also explain co-contraction, a typical feature of BPBI at a later age, in which shoulder abduction and elbow flexion, or elbow flexion and extension, become irreversibly linked. Misrouting may even result in the phenomenon of the “breathing arm”; clinically, when the proximal arm is at rest involuntary movements simultaneous with the breathing rhythm can be observed. This can be explained by misrouting of ruptured C4 or C5 axons that were originally connected to the diaphragm through the phrenic nerve and erroneously grow in to a superior trunk neuroma-in-continuity to shoulder muscles or the biceps muscle.

The most severe lesion type, which is specifically related to traction to the spinal nerves that form the brachial plexus, is a root avulsion. The result is a complete discontinuity of the neural connections of the central nervous system to the peripheral nervous system. Outgrowth of axons, and thus neuroma formation or misrouting, will not take place in the case of a root avulsion.

An additional factor, beyond the inadequate number of outgrowing axons and misrouting that may reduce functional regeneration, is that improper central motor programming may occur. There are various reasons why the formation of motor programs may fail in BPBI. First, BPBI causes deafferentation as well as weakness; many functions in the central nervous system depend on afferent input in a specific time window or they will not be formed correctly. Systematic abnormalities in sensory function were found in adults with conservatively treated BPBI as well as in children with an upper plexus lesion (either surgically or conservatively treated). BPBI affected imagined but not actual elbow flexion in functional MRI scanning, suggesting an impairment of motor planning, which would indicate abnormal motor programming in BPBI. A correlation between diminished sensibility and fine task motor performance was present. Second, aberrant outgrowth of motor axons may present the central nervous system with conflicting information. A motor command for shoulder abduction may, for instance, cause elbow flexion in addition to abduction through misrouted motor axons. The resulting feedback may well hamper the formation of a selective abduction program, as there is probably no way for the central nervous system to identify the “misbehaving” motor units. ^, Third, sensory axons might also be prone to misrouting, compounding the problem. A final hurdle for the central nervous system may be the severity of paresis. In such cases the only way to effect certain movements may be through “trick movements” (such as scapular rotation instead of glenohumeral rotation), which then represent a functional adaptation.

Natural History

The prognosis of BPBI is generally considered to be very good, with complete or almost complete spontaneous recovery in over 90% of patients. However, this opinion is based on a limited number of series, ^, which are cited indiscriminately, and without considering methodological aspects of these studies. In a systematic literature review, we discussed the methodological flaws in the available natural history studies. We found that no study presented a prospective, population-based cohort that was scored with a proper scoring system with adequate follow-up of 3 years. In other words, there is no scientifically sound evidence to support the common perception of complete spontaneous recovery from BPBI. The often-cited excellent prognosis may be too optimistic. Analysis of the most methodologically sound studies led us to estimate the percentage of children with residual deficits at 20% to 30%. This analysis was subsequently confirmed by two other studies, which prospectively investigated a population-based cohort. In the first, the British Paediatric Surveillance Unit notification system performed a nationwide registration of BPBI injuries. Follow-up was restricted to 6 months, and in this period only half of the infants showed full recovery. A prospective, population-based study from a Swedish region revealed that 18% of the children had residual deficits at 18 months.

Electromyography and Prognosis

Ancillary testing, in particular electromyography (EMG), is not considered reliable enough for prognostication of BPBI. ^, A needle EMG might seem a useful tool in this respect, but at present its role is debated. A main reason for this is that EMG findings may be discordant with clinical findings at 3 months of age, the age at which the biceps test is performed. In a paralytic biceps brachii muscle, the expected findings are an absence of MUPs and the presence of positive sharp waves and/or fibrillation potentials (called denervation activity ). But in a typical BPBI case, MUPs are present and denervation is absent in a paralytic biceps muscle at 3 months of age. This confusing finding has been noted by others, ^, and may have contributed to the opinion that the EMG is not useful in BPBI. ^{, ,} We previously outlined several possible explanations for “inactive MUPs,” that is, MUPs in a paralytic muscle. These suggest that the presence of inactive MUPs may depend on time after injury, as they reflect incomplete outgrowth of damaged axons and the hampered formation of motor programs in the central nervous system.

Spontaneous recovery of useful extremity function has been observed in subsets of patients without elbow flexion at 3 months of age. In one study, 20 of 28 infants who had no biceps function at 3 months had developed biceps contraction at 6 months. Together with our findings ^, that MUPs can almost always be found in the biceps muscle at 3 months, this strongly suggests that the age of 3 months does not represent a stable state in BPBI. In fact, the outgrowing axons may well have only just arrived in the various muscles, and the central nervous system may not yet have learned to cope with the situation. In nerve lesions in adults, one may expect all motor programs to be ready and waiting for the restoration of peripheral connections. In BPBI, axonal outgrowth may be only the starting point for restoration of function, as formation of central nervous system motor programs may commence only after enough axons have arrived to start exerting force. At the same time, forming such central motor programs may be more difficult and thus take longer than in healthy children, as the central nervous system must somehow take aberrant outgrowth and the confusing feedback it causes into account. Faced with a degree of inescapable co-contraction, it may not be easy to program effective elbow flexion, abduction, or rotation. In this hypothetical view, the age of 3 months may well be the very worst period imaginable to correlate the EMG with clinical findings: it is late enough to show evidence of axonal outgrowth, but too early for the brain to control contraction efficiently. This leaves the role of the EMG for prognosis at 3 months undetermined at present; we showed that severe cases of BPBI can be identified reliably at 1 month of age based on clinical findings and needle EMG of the biceps. ^,

For less extreme cases, that is, the majority of BPBI cases, the challenge lies in predicting whether function will be best after spontaneous outgrowth through a neuroma-in-continuity, resulting in reinnervation through tangled paths, or after nerve grafting, in which the grafts serve as a straight path that can be targeted. Results achieved by surgery are claimed to be superior to the outcome in conservatively treated subjects with equally severe lesions. ^{, ,} However, this comparison relies on historical controls ; no randomized study has been performed. ^, The best way to answer this question may be a controlled trial comparing nerve surgery to spontaneous recovery. In view of the current standard of treatment practice, it seems extremely difficult to perform such a prospective randomized trial.

Conservative Treatment in the First Few Months of Life

In the past there has been a tendency to immobilize the arm directly after birth to prevent secondary damage to the injured nerve elements of the brachial plexus. It is, however, highly unlikely that secondary damage to the brachial plexus can occur during the passive movements of the arm in a physiological range of motion during exercises or caretaking. We recommend frequent mobilization of the joints from the very beginning to prevent joint contracture formation. Additionally, there is no scientific proof that immobilization might be of any benefit in accelerating or improving the nerve regeneration process. Joint contracture formation, however, might be detrimental to final functional outcome when contractures limit the effective contraction of reinnervated muscles. It may also lead to improper modeling of the joints, of which the glenohumeral joint is most frequently affected. Contracture formation may start as early as 2 to 3 weeks after birth. The type of joint contractures that we most frequently see are those resulting in an internal rotation, flexion, and pronation fixed position of the upper limb. Exercises that focus on prevention of this type of contracture formation and optimization of joint mobility consist of passive external rotation in adduction and supination with a ninety-degree flexed elbow, just to the point where a certain tension can be felt. The arm should than be held in this position for a few seconds and then released. This passive movement should be frequently repeated during each session. We advise the parents to mobilize the affected arm as frequently as possible during the day, but at least every time the diaper is changed. In addition, we recommend the parents move both arms in a symmetrical fashion. In this way, the parents can use the unaffected arm range of motion as a reference value and therapy target for reach of the affected arm. Joint mobility should initially be evaluated every week by a specialized child physiotherapist. Besides active mobilization of the joint, keeping the shoulder joint in an externally rotated position using a specially designed splint has been proposed as early treatment for shoulder contractures and to encourage active external rotation.

Nerve Surgery