Traumatic and Neoplastic Diseases of the Brachial Plexus

The brachial plexus is the collective name for the lower motor neuron (LMN) components of the lower cervical and upper thoracic spinal cord segments (C6 toT2), which innervate the shoulders and forelimbs [1]. Injuries to the brachial plexus, either by trauma or neoplasia, result in dysfunction of muscle groups innervated by the involved nerves. In this chapter we present an overview of the anatomy and function of the brachial plexus and a review of the incidence, mechanisms of injury, diagnostic methods, prognosis, and treatment options for traumatic and neoplastic diseases of the brachial plexus.

Anatomy and Function

Fig. 44-1 depicts the location of the spinal cord segments, peripheral nerves, and forelimb muscles associated with the brachial plexus' general somatic efferent (GSE) function. Each LMN consists of a cell body located in the ventral gray matter of the spinal cord, the ventral root, the spinal nerve, the peripheral nerve, and the myoneural junction [2]. Motor nerves of the forelimb are formed by ventral branches of spinal cord segments C6, C7, C8, and T2 [3]. The fifth cervical segment occasionally contributes to the plexus, but when it does, the second thoracic segment usually contributes nothing. When both C5 and T2 contribute to the brachial plexus, the ventral branches of each are small. The ulnar, median, radial, axillary, musculocutaneous, subscapular, suprascapular, thoracodorsal, and lateral thoracic branches are the motor nerves formed by the brachial plexus, and they innervate intrinsic and extrinsic muscles of the shoulder, brachium, and antebrachium [1].

Figure 44.1. Anatomy of the brachial plexus: cervical vertebrae (Roman numerals), spinal cord segments, distribution of the spinal cord segments to the peripheral nerves of the brachial plexus, and muscles innervated by the peripheral nerves of the brachial plexus.

Injuries to the Brachial Plexus

Causes

Automobile accidents are the most frequent cause of injuries to the brachial plexus. Additional causes include projectiles, falling or jumping from moving vehicles, and foreign bodies that enter the axillary region [4].

Mechanism

Avulsion of the brachial plexus follows extreme abduction of the shoulder or traction of the forelimb caudally, placing traction and longitudinal forces on the nerves [3]. Spinal nerve roots, the ventral nerve root branches, and the brachial plexus are subject to injury. Spinal nerve root avulsion probably occurs more frequently because these structures contain less connective tissue and, so, are less elastic than the extravertebral components of the LMN. Peripheral nerve dysfunction resulting from an avulsion injury is usually severe and permanent, although one study showed a 28% (8/29) rate of return to acceptable limb function 4 months after injury. Loss of radial nerve function is most critical because adequate limb function cannot be achieved without at least partial triceps muscle function [5].

Classification

Neuropraxia is a temporary loss of nerve function without structural damage. Axonotmesis is loss of function from separation of the axons when the endoneurium remains intact. Complete severance of the nerve trunk defines neurotmesis (Fig. 44-2). Return of function following neuropraxia may take 1 to 3 weeks. With axonotmesis or neurotmesis, the nerve undergoes Wallerian (axonal and myelin) degeneration distal to the site of injury. Regeneration of the axon fibers can occur, albeit slowly, if unimpeded by organized blood clots in the neural sheath (which occurs with axonotmesis) or by separation of the neural sheath (neurotmesis) [6].

Figure 44.2. Classification of nerve injuries and associated structural changes.

Clinical Signs

Animals with brachial plexus injuries frequently have a history of loss of function in one forelimb noted immediately following trauma. Bilateral brachial plexus injuries are rare. Readily apparent clinical signs may include inability to fully extend the elbow or bear weight on the limb, and knuckling of the forepaw. Ipsilateral miosis, protruding third eyelid, enophthalmos, and ptosis (Horner's syndrome, Fig. 44-3) may be evident, and the cutaneous trunci (panniculus) reflex may be absent on the affected side. Animals presented 10 to 14 days after brachial plexus injury exhibit neurogenic atrophy of the affected muscles (Fig. 44-4) [3]. Injury to any component of the LMN, whether temporary or permanent, results in dysfunction of the entire unit. The neuropathic syndrome characterizes the clinical signs accompanying LMN dysfunction: decreased or absent muscle tone (hypotonia, atonia), reduced or absent segmental reflex responses (hyporeflexia, areflexia), and neurogenic muscle atrophy [7].

Figure 44.3. Anisochoria/partial Horner’s syndrome in a dog with a left brachial plexus avulsion. The miotic left pupil results from a reduced or absent sympathetic nerve supply.

Figure 44.4. Right forelimb of a dog 2 weeks after a partial brachial plexus avulsion. Complete loss of ulnar, median, radial, and axillary nerves was noted; minimal musculocutaneous and suprascapular nerve function was preserved. Notice the extreme neurogenic atrophy.

Complete avulsion of the brachial plexus spinal nerve roots (C6 toT2) results in a flaccid, fully extended limb that is incapable of bearing weight. Avulsion of only the caudal plexus roots (C8 to T2) causes the animal to carry the limb with the elbow and shoulder flexed as a function of the remaining musculocutaneous, axillary, and suprascapular nerves.8 Injury to the T1 nerve root may also damage sympathetic innervation to the eye, causing ipsilateral Horner's syndrome. Avulsion of the C8 nerve root results in ipsilateral loss of the cutaneous trunci reflex [3].

Radial, axillary, and thoracodorsal nerve dysfunction caused by direct nerve damage or injury to the C7, C8, and T1 roots results in more ventral carriage of the elbow from loss of a majority of the shoulder flexors. The limb is unable to support weight. Cutaneous analgesia includes the caudal scapular region, a small area of the lateral brachium, the dorsum of the paw, and portions of the cranial and lateral antebrachium [3,8].

Peripheral nerve injuries not caused by traumatic avulsions of the brachial plexus can involve specific nerves, and the level of injury dictates the type of dysfunction. Radial nerve injury distal to the triceps muscle branches causes knuckling of the forepaw and loss of carpal and digital extensor function. Most animals compensate by learning to flip the carpus forward when walking or running. Distal radial nerve damage causes analgesia of the dorsal surface of the paw and of portions of the cranial and lateral antebrachium. More proximal radial nerve damage creates similar cutaneous sensory loss but more severe motor dysfunction. The elbow is carried more ventrally because the shoulder flexor function of the long head of the triceps muscle is lost. Intact elbow and shoulder extensor function causes the limb to be carried upward (flexed). Attempts to use the limb result in knuckling of the forepaw and collapse of the limb. An animal with radial nerve injuries that cause loss of triceps function is unable to bear weight on the limb [3].

Selective involvement of specific nerves of the brachial plexus other than the radial nerve does not produce substantial disability [3]. Sensory loss is evident in areas associated with the injured nerve (Fig. 44-5). Loss of ulnar nerve function, which provides sensory innervation to the lateral aspect of the foot, predisposes animals to abnormal wear and abrasions of the lateral aspect of the footpads as the foot is allowed to slide laterally when bearing weight [9].

Figure 44.5. Autogenous zones on the lateral and medial aspects of the canine forelimb.

Tumors of the Brachial Plexus

Nerve sheath tumors are infrequent in dogs, rare in cats, and most often involve peripheral nerves derived from the brachial plexus. A survey of all entries from 21 veterinary colleges to the Veterinary Medical Data Program records for a 10-year period revealed the incidence was less than 1 per 32,000 dogs (0.003%) and less than 1 per 160,000 cats (0.0006%). These neoplasms have been described as schwannomas, neurilemmomas, neurinomas, and neurofibrosarcomas. Collagen content, encapsulation characteristics, and structures from which the tumors arise have been used to differentiate the tumor types; however, even with special staining and electron microscopy, differentiation is frequently a difficult and speculative process in veterinary pathology. Nerve sheath tumor is perhaps the best accepted term for these neoplasms in dogs and cats [10].

Lymphosarcomas occasionally involve peripheral nerves or spinal nerve roots [7,11]. Other primary neoplasms such as meningiomas, which are extramedullary and intradural, can produce signs referable to the brachial plexus. Metastatic neoplasms, especially those involving the vertebrae, can also cause signs referable to the brachial plexus [7].

Nerve sheath tumors of the brachial plexus are usually slow growing, rarely metastasize, and are classified as extramedullary and intradural when growth of the mass extends into the spinal canal [10]. The most caudal cervical (C7, C8) and the first thoracic spinal roots were more frequently involved in two clinical reports, but any of brachial plexus spinal nerve roots, ventral branches, or peripheral nerves can be primary sites of origin [10,12]. Clinical signs are slowly progressive and usually manifest as pain followed by muscle atrophy, limb dysfunction, and finally, spinal cord compression with continued growth of the tumor. Occasionally, the first sign noted is the animal's licking or chewing of the foot or carpus of the affected limb. Involvement of the T1 spinal nerve root may result in ipsilateral Homer's syndrome; with C8 spinal nerve root neoplasms, the cutaneous trunci reflex may be decreased or absent on the involved side [3,10].

Diagnosis is based on typical history and clinical signs, palpation of a mass in the axillary region, electrodiagnostic tests, myelographic changes consistent with an extramedullary or intradural structure, and surgical biopsy findings. Early diagnosis is crucial to successful long-term management of animals with nerve sheath tumors, but it is frequently delayed by incomplete diagnostic evaluations and symptomatic treatment for suspected musculoskeletal or intervertebral disk disease [7,10]. One study reported an interval of more than 6 months from onset of signs to diagnosis in a survey of 18 dogs with nerve sheath tumors [10].

Diagnostic Methods

History and Signalment

Brachial plexus injuries are usually associated with a history of trauma, although injuries from foreign bodies that penetrate the axillary region may be the exception. Owners of animals with brachial plexus tumors often relate a history of obscure lameness, resistance to palpation of the limb, possibly licking and chewing of the foot or carpus, and finally a reluctance to bear weight on the limb. The course of this history may be a few days to several weeks [10,13].

A survey of brachial plexus nerve sheath tumors reported from 21 veterinary colleges to the Veterinary Medical Data Program during a 10-year period did not reveal a breed predisposition for brachial plexus nerve sheath tumors. Large-breed dogs (i.e., Labrador retrievers) appear to be over-represented in the literature, but this likely represents breed popularity rather than increased incidence rate.

Physical and Neurologic Examinations

Thorough examination is important when brachial plexus injury or neoplasia is suspected. Animals presented following motor vehicle trauma should be assessed for thoracic and abdominal injuries as well as fractures or nervous system injuries remote to or associated with the brachial plexus. In addition to a thorough physical examination, animals with suspected brachial plexus tumors are carefully assessed for the presence of other tumors.

Neurologic deficits found with brachial plexus injuries or tumors reflect the location and extent of brachial plexus and spinal cord involvement. Early signs of a nerve sheath tumor may not include neurologic deficits, but advanced signs may include hyporeflexia, muscle flaccidity, cervical spinal cord compression, and involvement of the contralateral limb. Brachial plexus trauma produces lower motor neuron dysfunction of the injured spinal nerve roots, ventral branches, or peripheral nerves. Complete brachial plexus involvement abolishes triceps, biceps, and forelimb flexor reflexes [3].

Sensory innervation mapping is used to locate areas of cutaneous anesthesia on the limb. This examination requires a patient examiner, a cooperative patient, and a quiet, uninterrupted environment. Beginning on the dorsum of the paw, small, curved Halsted forceps are used to pinch the skin in a two-step procedure (Fig. 44-6). An assistant gently restrains the animal while distracting its attention from the examiner. First, a small amount of skin is grasped lightly and lifted slightly with the forceps. Next, the hemostats are used to pinch the skin. Intact sensation is determined by vocalization of the animal, looking at the site of stimulation, or attempts to discourage the examiner from pinching the skin. Only consistent and repeatable results are considered valid. The examination is continued, proceeding up the antebrachium on the medial, lateral, cranial, and caudal sides. A cutaneous zone is the total cutaneous area supplied by a peripheral nerve. Some peripheral nerves have cutaneous zones that overlap. Autogenous zones are cutaneous regions that receive sensory innervation from only one peripheral nerve. Analgesia in an autogenous zone implies a loss of sensory function for the nerve represented by that zone [3,4]. Fig. 44-5 illustrates the autogenous and cutaneous zones of the forelimb in the dog; the distribution is similar in cats. Certain dorsal spinal nerve branches in the brachial plexus region also have cutaneous and autogenous zones. The distribution of the cutaneous analgesia is used to determine which spinal and peripheral nerves are impaired.

Figure 44.6. A and B. Two-step method for cutaneous sensory evaluation of the forelimb, a modification of the technique described by Bailey.

Electrodiagnostic Tests

Several electrodiagnostic tests can be used to confirm and further characterize suspected nerve dysfunction. The electromyogram (EMG) is a graphic and auditory analysis of muscle function. Certain waveforms and sounds recorded from muscles may reflect denervation, reinnervation, or normal function.

Spontaneous activity consisting of fibrillation potentials and positive sharp waves are patterns associated with denervation and can be recorded 5 to 14 days after denervation. Low-amplitude monophasic or polyphasic motor unit action potentials (MUAP) or giant MUAP may be demonstrated from muscle undergoing reinnervation. Early stages of reinnervation can result in small MUAP when the number of regenerated axons is few, and polyphasic waves indicate a disparity in the degree of remyelination among the regenerating fibers. Giant MUAP occur when denervated muscles are reinnervated by adjacent functional nerves, resulting in a low muscle nerve branch to muscle fiber ratio.

Early evidence of reinnervation is not predictive of full recovery of function. Serial EMG evaluations that produce consistent physical and electrophysiologic evidence of active muscle reinnervation are more reliable indicators of a reinnervation process sufficient to produce some return of function. Serial microscopic examination of muscle biopsy specimens from the most proximal affected muscles (e.g., supraspinatus, deltoid) may be used to further confirm the regenerative process. Normal muscle exhibits insertional activity as the EMG electrode is passed through it; in anesthetized animals, however, electrical silence is evident after the electrode is stationary. A thorough EMG examination 2 to 3 weeks after a brachial plexus injury documents the extent of muscle denervation and confirms the distribution of nerve root injury. The EMG may be helpful in early diagnosis of brachial plexus tumors, especially when limb dysfunction is minimal or absent and a pattern of scattered spontaneous activity indicative of denervation (fibrillation potentials, positive sharp waves) is found in the distribution of one nerve.

Nerve conduction studies are of limited value in the assessment of brachial plexus dysfunction. Soon after avulsion, before Wallerian degeneration has proceeded far distally, a motor nerve can conduct an evoked impulse at a velocity within the normal range. After Wallerian degeneration, impulses cannot be conducted. This information adds little to the findings of physical and electromyographic examination. Sensory nerve conduction studies may be useful in determining the location of the nerve injury. When sensory nerve fibers are avulsed proximal to the dorsal root ganglia, however, the sensory fibers of the peripheral nerve do not degenerate and are capable of conducting an evoked impulse, even with complete analgesia in the autogenous zone [14].

Spinal evoked potentials (cord dorsum potentials) may be more valuable for determining the integrity of the sensory nerve function. Stimulation of the nerve in the distal portion of the limb and recording an evoked response from an electrode over vertebra C6 or C7 indicate that a portion of the sensory tracts is functional for that nerve. The value of this test for predicting return of function is unknown.

Other motor evoked responses that might be useful but have not been sufficiently evaluated in veterinary medicine are the H wave and F wave. The H wave is produced by a submaximal stimulation and is equivalent to a monosynaptic reflex. The H wave can be recorded only if the dorsal and ventral roots are intact. The F wave is produced by a supra-maximal stimulus, represents antidromal and orthodromal transmission of the electrical impulse, and requires an intact peripheral nerve, ventral root, and ventral horn cell. Presence of an F wave indicates a functional motor nerve [15].

Radiography and Advanced Imaging

Radiography and advanced imaging techniques (myelography, ultrasonography, computed tomography, magnetic resonance imaging) are used to evaluate the vertebrae and spinal cord of the caudal cervical region [16-20]. Brachial plexus nerve sheath tumors that invade the spinal canal often create bony lysis at the intervertebral foramen. The widened foramen may be demonstrated on lateral views of the spine. Myelography usually depicts spinal cord compression and an extramedullary intradural pattern [19]. With brachial plexus avulsions of the spinal roots, the contrast material may be seen outside the subarachnoid space, at the site of the avulsion. The expanded availability of ultrasonography, computed tomography, and magnetic resonance imaging in veterinary medicine has proven useful in determining avulsion sites and the extent of nerve sheath tumors in the caudal cervical region (Fig. 44-7) [16-18].

Figure 44.7. Computed tomographic images of a patient with brachial plexus avulsion. Note the extreme atrophy of the right appendicular muscles, but relatively mild atrophy of the scapular muscles. Electrodiagnostics in this patient indicated avulsion of the caudal cervical nerve roots (C7 through T2) and, therefore, some sparing of the suprascapular and subscapular nerves. A. Level of the C6 vertebra; B. Level of the C7 vertebra.

Therapeutic Options and Guidelines

Injuries to the Brachial Plexus

Most brachial plexus injuries unfortunately result in permanent dysfunction of the affected nerves and associated muscles of the forelimb. Most plexus avulsions occur at or within the spinal cord, and surgical repair would require extraordinary effort, expertise, and expensive equipment (e.g., operating microscope, microsurgical instruments) [9]. More distal injuries, involving peripheral nerves, warrant surgical exploration and, when indicated, primary surgical repair. Caution and prudence should guide the decision for primary nerve root repair. Low-frequency laser technology for nerve anastomosis may warrant consideration in the future.

Definitive therapy for brachial plexus injury is delayed until the full extent of injury is known. Fig. 44-8 is an algorithm that outlines a schedule for serial evaluation of brachial plexus injuries and the various treatment options available. The initial concern is protection of the limb from further injury and prevention of flexor tendon contracture. If elbow flexion is not sufficient to prevent the carpus from dragging on the ground, the carpus is bandaged. Limb carriage and sensory mapping are documented in the medical record for future reference. In 7 to 14 days the limb is re-evaluated: sensory mapping is repeated, muscle atrophy is noted, and EMG evaluation is performed if available. Evidence of self-mutilation, skin abrasions on the dorsal surface of the carpus, and flexor tendon contracture are addressed at this time. Amputation is considered if there is total analgesia distal to the elbow and evidence of self-mutilation and if improvement is not noted within three weeks. To prevent flexor tendon contracture, a coaptation splint can be applied or the owners can perform hyperextension exercises on the limb for 10 minutes four times a day. The latter approach requires cooperation from the animal and strict compliance from the owner. Weekly reevaluations are scheduled for three weeks. If improvement is noted, conservative therapy is continued. Axon regeneration does not begin until 7 to 14 days after injury, and neuropraxia may take three to six weeks to fully resolve [6]. Muscle biopsy and EMG are needed to confirm evidence of reinnervation. If no improvement is noted during the three weekly visits, therapeutic options are these: conservative (no further treatment) and surgical (amputation, tendon transposition, nerve-muscle transposition).

Figure 44.8. Algorithm of serial evaluation and treatment options for brachial plexus injuries. EMG, electromyography; MNCV, motor nerve conduction velocity; SpEP, spinal evoked potential. (Adapted from Knecht CD, Raffe MR. Diseases of the brachial plexus. In: Textbook of Small Animal Orthopaedics. Newton CD, Nunamaker DM (eds). Philadelphia: JB Lippincott, 1985.

When no improvement is noted, if the limb is not constantly subjected to trauma (dorsal carpal skin abrasions, self-mutilation) and limb appearance is acceptable to the client, no additional care is needed or indicated.

Tendon transpositions have been reported by several authors as therapeutic options when musculocutaneous function is intact [9,21-23]; however, most reports are of experimental and not clinical data. In most spontaneous brachial plexus injuries, any musculocutaneous nerve function remaining is partial, and neither the brachialis nor the biceps brachii muscles have sufficient functional muscle fibers to both flex and extend the elbow [5]. The transposition procedure is intended to supply function to the triceps muscle, allowing extension of the elbow. Even if sufficient triceps function can be supplied, sensory deficits remain and self-mutilation can begin any time after the injury. If the tendon transposition is successful, carpal arthrodesis should be considered as an adjunct to triceps function. Arthrodesis should align the carpus in slight hyperextension (8° to 12°) and 5° to 8° outward rotation [9]. With each step of the tendon transposition or carpal arthrodesis therapy, owners are again counseled about the ever-present threat of self-mutilation.

Nerve-muscle transpositions have shown limited success in experimental radial nerve transection. The efficacy of nerve allografts has varied, with failure usually related to immune- mediated reactions [9]. Intact neurovascular muscle pedicle grafts have been considered; this requires microvascular technique and may have future clinical applications.

Tumors of the Brachial Plexus

Early diagnosis is the single most important factor in successful treatment of brachial plexus tumors. Any well founded suspicion of a brachial plexus tumor is an indication for surgical exploration, although the more common items in the differential diagnosis must first be ruled out. Brachial plexus neuritis, trauma, and vascular thrombus are all characterized by acute onset of substantial lower motor neuron dysfunction of one or both pelvic limbs [5]. Radicular and spinal cord compression may have a slower, more insidious onset, although clinical signs and radiographic studies should differentiate these conditions from brachial plexus tumors. Radicular compression from an intervertebral disk extrusion involves only one cervical segment, even when progressive, and should be demonstrable on plain film radiography or myelography. Caudal cervical spinal cord compression from an intervertebral disk extrusion or spinal cord neoplasm usually produces clinical signs, which are more evident in the hind limbs than the forelimbs, although neck pain can be the only sign. Again, radiographic studies should demonstrate the lesion.

When brachial plexus tumors are diagnosed or when other more common causes of forelimb LMN dysfunction are ruled out, exploration of the brachial plexus to determine the extent and distribution of the suspected neoplasm is the best method of definitively diagnosing the problem. All too often, the decision to explore is delayed unnecessarily [5].

The surgeon tailors the surgical approach to the location of the suspected mass. Careful palpation, before and after induction of anesthesia, is helpful, as is a thorough knowledge of brachial plexus anatomy, EMG changes in the limb, and results of the neurologic examination. If the tumor is palpable and located distal to the mid-humerus, a linear incision is made from the proximal third of the humerus to the middle of the antebrachium. The tumor is identified and resected, along with a margin of normal-looking nerve proximal and distal to the mass [5].

Exploration of the region from the brachial plexus to the spinal canal is accomplished through an incision over the scapula followed by muscle dissection on the cranial, caudal, and dorsal aspects of the scapula and reflecting the scapula toward the surgeon (Fig. 44-9). Neural structures from the spinal canal to the mid-humerus can be explored with this exposure. The axillary lymph node can be biopsied with this approach. Prior knowledge of the precise location of the mass, through palpation or advanced imaging techniques, allows some modification of this approach. If the mass is in the cranial-most brachial plexus area, transection of the omotransversarius muscle and retraction plus blunt dissection allow exposure of this region. For the caudal brachial plexus, the rhomboideus muscle is transected and retraction plus blunt dissection are adequate for exposure of this area [5].

Figure 44.9. Exploration of the brachial plexus. 1, Initial skin and subcutaneous tissue incision; 2, incision through tendinous attachments to the spine of the scapula; 3, incision through omotransversarius (0) muscle attachment to the scapula; 4, incision through rhomboideus (r) muscle attachment to the scapula (t, trapezius; d, deltoideus); 5, incision through serratus ventralis (sv) muscle (s, supraspinatus; i, infraspinatus); abduction of the scapula follows, to expose the brachial plexus nerves (aln, axillary lymph node; sc, scalenus; U, ulnar nerve; M, median nerve; R, radial nerve; tm, teres major; Ax, axillary nerve; MC, musculocutaneous nerve; Sbs, subscapular nerve; SpS, suprascapular nerve).

An alternative surgical approach is the craniolateral approach. The reported advantage is improved exposure of the proximal brachial plexus; whereas the craniomedial approach described above offers a better exposure of the peripheral nerves [24].

If the mass extends into the spinal canal, the tumor is resected to the point of the intervertebral foramen and the spinal canal is explored later (preferably within a week). Nerve sheath tumors that invade the spinal canal may also extend to the opposite side, necessitating extensive exposure of the canal for resection.

Limb amputation is a practical consideration if the mass extensively involves the brachial plexus or if resection will cause permanent triceps muscle dysfunction. If the mass extends into the spinal canal and has extensive brachial plexus involvement, euthanasia is a rational decision for the owners to make, since a cure is unlikely, long-term prognosis is poor, and the postoperative discomfort, physical rehabilitation, and dysfunction may be considerable. Palliative therapy is reserved for animals that cannot or should not undergo surgery for any reason (including recurrence of the tumor) and whose pain or discomfort can be controlled well enough to allow quality life in the time remaining.

Estimating Prognosis

Estimating prognosis is difficult early in the course of brachial plexus injuries and when proof exists of complete removal of a nerve sheath tumor on microscopic examination of tissue margins. Most avulsions cause permanent damage. Most nerve sheath tumors recur. It is the responsibility of the veterinarian to give clients accurate information so that they may make an informed decision.

Postoperative Rehabilitation

Care of the animal with brachial plexus injury that has undergone surgery for tendon transposition or nerve repair includes immobilization of the limb for at least 2 weeks. When arthrodesis of the carpus is performed, 6 to 8 weeks' immobilization is needed to allow fusion. Amputees seldom require major care after the immediate postoperative period. Animals recovering from surgery for brachial plexus neoplasms require physical therapy and much the same care as animals with traumatic brachial plexus injuries: protection of the carpus and hyperextension exercises to prevent flexor tendon contraction. Because the involved nerves are usually severed and return of function is not a consideration, the owners are made aware of the need for life-long commitment to this care. Any changes in the animal's ability to use the limb, signs of pain, or other factors that might herald recurrence of the mass should be reported immediately.