brachial plexus injury in adults

CME

The Surgical Treatment of Brachial PlexusInjuries in AdultsJulia K. Terzis, M.D., Ph.D., and Konstantinos C. Papakonstantinou, M.D., Ph.D.Norfolk, Va.

Learning Objectives: After studying this article, the participant should be able to: 1. Describe the mechanisms andpathophysiology of posttraumatic brachial plexus palsies in the adult population. 2. Evaluate clinically a patient withposttraumatic brachial plexus palsy and incorporate in the diagnosis results from electrophysiologic and radiologic studies.3. Formulate a preoperative treatment plan, describe various reconstructive strategies, and identify the usual postoperativefollow-up of the patient. 4. Anticipate the possible functional outcome of brachial plexus reconstruction and identify thefactors affecting the outcome.

Posttraumatic brachial plexus palsy is a severe injuryprimarily affecting young individuals at the prime of theirlife. The devastating neurological dysfunction inflicted inthose patients is usually lifelong and creates significantsocioeconomic issues. During the past 30 years, the sur-gical repair of these injuries has become increasingly fea-sible. At many centers around the world, leading surgeonshave introduced new microsurgical techniques and re-ported a variety of different philosophies for the recon-struction of the plexus. Microneurolysis, nerve grafting,recruitment of intraplexus and extraplexus donors, andlocal and free-muscle transfers are used to achieve optimaloutcomes. However, there is yet no consensus on thepriorities and final goals of reconstruction among thevarious centers. (Plast. Reconstr. Surg. 106: 1097, 2000.)

HISTORY

Injuries to the brachial plexus with subse-quent paralysis of the upper extremity are asold as warfare. Homer, in his Iliad, and Thucy-dides, in his History of the Peloponnesian War,eloquently described the devastating nature ofdirect or indirect injury to the upper extremity,with injured chariot drivers constituting thepatient population at that time. War, unfortu-nately, and other events resulting in injurieshave stimulated the studying of this clinicalentity, even in the modern era.

Anatomic findings of avulsed nerves fromthe spinal cord were described by Flaubert in1827,1 and Thorburn in 19002 reported hisintraoperative findings and direct repair of thebrachial plexus elements in a woman trauma-

tized in an industrial accident. The early oper-ative reports by many authors, such as Clark etal.3 and Taylor,3,4 as was the case with obstetricparalysis, were succeeded by a period of pessi-mism, mainly as a result of the unfavorableresults and the high morbidity and mortalityrates encountered.

Advances in surgical and anesthetic tech-niques during the second half of the centuryplus a large number of injured soldiers duringWWII permitted pioneer surgeons with a greatinterest in the surgical treatment of these inju-ries to revisit an old problem. In 1947, Seddon5

published his proposed method of surgical cor-rection of traction injuries with application oflong interposition nerve grafting. Despite thepoor outcomes reported by this author,6 theintroduction of magnification in the treatmentof these lesions, and the persistence of othersurgeons, such as Millesi in Austria and Nara-kas in Switzerland, stimulated many aroundthe world. Moreover, the introduction and re-finement of new diagnostic modalities such aselectromyography, the lamina test, and CT/myelography also helped to promote the sur-gical treatment of brachial plexus.

EPIDEMIOLOGY

The demographics of brachial plexus palsieshave been studied by different authors. In most

From the Microsurgery Research Center, Division of Plastic Surgery, Department of Surgery, Eastern Virginia Medical School. Received forpublication November 30, 1999; revised April 18, 2000.

1097

studies, including our own patient population,high-velocity motor vehicle accidents accountfor the majority of the cases (59 percent); moststudies report that motorcycle accidents areresponsible approximately twice as often as au-tomobile accidents. In other studies,7 this num-ber has been reported as high as 84 percent. Itis interesting, however, to note that the overallincidence of brachial plexus injuries in multi-trauma patients secondary to motor vehicleaccidents ranges from 0.67 to 1.3 percent.8This number increases to 4.2 percent for vic-tims of motorcycle accidents. This differencecan easily be explained by the increased forcesapplied to the brachial plexus of the unpro-tected body during a motorcycle accident: themost common type of injury found in this pa-tient population is traction injury, which re-sults in a greater incidence of avulsions as com-pared with the crush injuries encounteredmore often in car accidents. Other commoncauses include (in different percentages ac-cording to different studies) industrial acci-dents, pedestrian vehicle accidents, snowmo-bile accidents, gunshot wounds, and otherpenetrating injuries.9–16 Other causes of bra-chial plexus injuries in adults include stretchfrom falling from a roof or tree, skiing acci-dents, or iatrogenic insult.7 Multiple case re-ports of brachial plexopathy have been pub-lished, and a variety of causes have beenencountered. Direct compression of the plexusfrom a hematoma with or without associatedbone or vascular injury17 can induce paralysisof the upper extremity, which if mistreated canlead to permanent deficit. Tumors of the bra-chial plexus with different pathologic features,such as schwannomas, neurilemmomas, myxo-matous cysts, have been reported18,19; they canbegin with temporary tingling, which is en-hanced by tapping of a sometimes palpablemass, and escalate to true muscle paralysis.Isolated clavicular fractures, if left untreated,can occasionally be responsible for brachialplexus injuries.20,21 The symptoms can be de-layed and could be caused by direct compres-sion from the formation of callus or bony spur,nonunion, or a subclavian pseudoaneu-rysm.22–25 The brachial plexus, mainly becauseof lack of mobility, can be vulnerable duringanesthesia and surgical procedures such asmastectomies, axillary dissections, and breastaugmentations through an axillary approach.Because it is firmly attached to the vertebraeand the prevertebral fascia, the brachial plexus

can be stretched inadvertently with retractorsor injured directly during percutaneous cannu-lations.26 An anesthetized patient is unable torespond to pain, numbness, or weakness, somalposition of a patient with an overstretchedneck can cause traction to the plexus, whichcould be a reason for malpractice claims.27,28

Young men are most commonly affected. Inour population of 312 patients with posttrau-matic brachial plexus injuries, 88 percent weremale and 12 percent female. The age at injuryranged from 4 to 63 years. Both sides wereaffected equally. These statistics are in accor-dance with other studies.8,29

MECHANISMS OF INJURY

On the basis of all the above-mentionedcauses of injury, three major mechanisms areimplicated30:

1. Crush type of injury, caused by directblunt trauma to the neck and upper ex-tremity, with or without associated inju-ries. The plexus is crushed between theclavicle and the first rib.

2. Traction of the plexus. This is usuallycombined with flexion of the neck towardthe contralateral side and/or hyperexten-sion of the arm. Caudal traction of thearm will usually affect the upper roots andtrunks. Cephalad traction will most likelyinvolve the lower plexus.

3. Compression of the plexus from hemato-mas or adjacent tissue elements that havebeen injured.

All the above proposed mechanisms can co-exist, which usually results in a less-favorableprognosis. In addition to the position of thearm in relation to the neck and the trunk, thevelocity and magnitude of force applied to theroots and trunks are of greater importance. Inmost cases, these factors determine the severityof the injury and affect the overall prognosis.31

ASSOCIATED INJURIES

Associated injuries are quite common in thispatient population. Because motor vehicle ac-cidents account for most brachial plexus inju-ries, it is logical to expect multiple organ inju-ries. In most reports, closed head injuries leada long list of associated trauma. Injury to thechest (hemopneumothorax or broken ribs) isalso very common. Vascular injuries to the sub-clavian or neck vessels, dislocations of theshoulder, and fractures of the clavicle, scapula,

1098 PLASTIC AND RECONSTRUCTIVE SURGERY, October 2000

and long bones are equally frequent. Injuriesto the subclavian artery usually involve theproximal or distal third. It should be men-tioned that this is often a delayed diagnosisbecause of the excellent collateral circulationaround the shoulder. Abdominal visceral inju-ries are the most common cause of emergencysurgical explorations, followed by decompres-sion of intracranial hematomas.8

PRESENTATION AND DIAGNOSIS

The patient with posttraumatic brachialplexus palsy usually presents in the emergencyroom with multiple trauma. Potentially life-threatening associated injuries are obviously ofprime importance and are treated first. Theusual scenario involves a patient being sedatedfor a few days then medically stabilized. Afterthe patient is weaned of sedation, the paralyzedarm is noticed, and the first evaluation is usu-ally done by a physician not experienced withthis type of injury. Because the long-term con-sequences of such injuries could have deleteri-ous effects on the patient’s lifestyle, evaluationby an expert is necessary so that a managementstrategy can be established early.

History

The formulation of a diagnosis, treatmentplan, and prognosis can be largely accom-plished by means of a careful and detailedhistory and physical examination. Recordingdata as accurately as possible from the first visitis very important. The mechanism of injury hasto be documented. Because most patients areyoung adults who are involved in motor vehicleaccidents, questions pertaining to wearing ahelmet or a seat belt or documented positiveloss of consciousness could be helpful. Lowervelocity injuries, such as falling from a tree or aroof, result in injury usually but not necessarilyof a lesser magnitude. If the injury is sustainedduring an industrial accident, the patient hasto be explicit about the circumstances sur-rounding the injury, describing, for instance,how the upper extremity was caught andpulled by the rotor of a machine or how aheavy object landed on the unprotected shoul-der. In many industrial accidents, injuries arecaused by peers attempting to free the patientby pulling on the trapped upper extremity. Aswas mentioned above, documentation of thecircumstances of the injury is very importantand should be accompanied by other demo-graphic data, such as the patient’s preexisting

neurologic status or his or her occupation andhandedness. Most patients experience maxi-mum loss of function at the time of injury;therefore, they should be asked about recoveryof function during the period between injuryand evaluation. When a Horner syndrome isobserved, it is important to inquire about priorabnormalities of the pupils or eyelids. Inquiryabout the onset, nature, and localization ofpain and the use and effectiveness of variousanalgesics should be made. Lastly, the overalleffect of the injury on the patient’s life and hisor her expectations regarding surgical treat-ment should be discussed.

Physical Examination

As with any other disease, the physical exam-ination should be as thorough as possible. Inmost centers, the results of the clinical exami-nation are reported in preprinted brachialplexus diagrams that include all muscle groupsof the upper extremity, sensory mapping, andpain threshold. The patient should be comfort-able and undressed from the waist up. Whilethe patient is undressing, it should be observedwhether the patient functions independentlyand how he or she has adapted to the loss offunction of the one extremity.

Examination usually begins with inspectionof the overall symmetry and observation of ob-vious scars related to either the initial traumaor subsequent surgery. The range of motion ofall joints and the neck is assessed. Some pa-tients with injury to the cervical spine mightpresent with cervical scoliosis.32 Injury to thesternocleidomastoid muscle should also benoted when the patient is asked to turn his orher head away from the injury. The supracla-vicular and infraclavicular areas should be in-spected and palpated for obvious scarring orfor the existence of bony spurs. Calluses frommalunions of the clavicle can be palpated, andtheir presence could suggest compression ofthe underlying plexus. Palpation should alsoinclude the ipsilateral chest for diagnosis offractured ribs. This is important when one con-templates harvesting of intercostal nerves asmotor donors for brachial plexus reconstruc-tion.

It is important to keep in mind that high-velocity trauma quite frequently causes nerveinjury at several levels. One should be suspi-cious of such multilevel injuries if the patientpresents with corresponding radiologic find-ings. In these cases, exploration and recon-

Vol. 106, No. 5 / BRACHIAL PLEXUS INJURIES 1099

struction of the brachial plexus at the clavicu-lar level only will not yield functional recoveryif there is concomitant nerve injury more dis-tally. For example, in documented associatedfractures of the humerus, the radial nerveshould also be explored through a posterolat-eral incision at the level of the mid-humerus.In cases of elbow fractures, the ulnar and me-dian nerves should also be explored and re-paired at that level.

The examination of passive range of motionof all joints should be done before examina-tion of active range of motion. Limitations inthe former are quite common and could mis-lead the examiner during the active muscletesting. Deficits are most often the result ofcontractures from inadequate physical therapy.Manual muscle testing33 begins by observingsigns of muscle atrophy, palpation of the toneof each muscle group, and exertion of resis-tance by the examiner’s hands. It is often nec-essary to observe supplementary motions thatthe patient uses to achieve a certain movement.The British Medical Research Council musclegrading scale is used by most physicians. Exam-ination of sensibility should include light touchand moving and static two-point discriminationand should be documented separately in thechart.

The physical examination is not completewithout evaluation of the vascularity of thearm. A traction force strong enough to avulsenerve roots from the spinal cord can easilydisrupt and damage the subclavian vessels.Moreover, associated fractures of the claviclecan easily cause subclavian vessel injury withsubsequent pseudoaneurysm formation. Sim-ple palpation of the radial and ulnar arteriesand evaluation using Doppler ultrasoundmight not be adequate to determine the ade-quacy of the vascular supply to the arm. Ifmajor surgery is contemplated, then angiogra-phy night be indicated. A loud bruit over thesupraclavicular or infraclavicular area also war-rants the need for angiography.

The presence of a Tinel34 sign, elicited bytapping the supraclavicular area, can be quiteuseful in differentiating between root avulsionand root rupture. A positive Tinel sign, whichis perceived by the patient as tingling in theanesthetic arm or hand, is a strong indicator ofrupture. This is important; if it is correlatedwith intraoperative findings, it means that thereconstructive surgeon has available intra-plexus donors for grafting purposes. It is pos-

sible to elicit a Tinel sign by applying a me-chanical stimulus at the level of the exit of thevarious cervical roots in a craniocaudal fashion.It is easier to perform this maneuver with thepatient lying down in a supine position andwith the head turned away from the lesion. Incase of ruptures, the patient typically will re-spond by pointing with the contralateral handto the site where the tingling sensation occurs.Thus, if the C4 root is tapped, the patient willlocalize the elicited sensation over the supra-clavicular area. If the C6 root is tapped, thepatient will point to the lateral antebrachialregion. At times, this dermatome may includethe thumb and the radial dorsum of thehand.35 The Tinel sign can be repeated onsubsequent clinical evaluations and its ad-vancement distally is presumptive evidence ofaxonal regeneration. Lastly, the eyes of thepatient should be examined to determine thepresence of Horner syndrome (meiosis, uppereyelid ptosis, and absence of facial sweating onthe affected side). A positive Horner’s signimplies disruption of the sympathetic supply tothe ipsilateral eye and face through the lowerroots (C8 and T1) and is a strong indicator ofavulsion of these roots.

Radiologic Evaluation

Patients who sustain brachial plexus injuriesare usually evaluated with routine x-rays of theaffected extremity at the time of presentation.A complete study of the cervical spine and theinvolved shoulder with special attention to theclavicle and scapula is absolutely necessary. De-pending on the mechanism of injury, fracturesof the transverse processes might indicate avul-sions of the corresponding roots, due to theattachments of the deep cervical fascia betweenthe cervical roots and the transverse process-es.36 Fractures of the ribs should be diagnosedradiographically because they could indicateinjury to the intercostal nerves, a potentialsource of motor fibers for subsequent neuroti-zation. Fractures of the clavicle and scapula ordissociation of these bony structures from thethorax could indicate a worse supraclavicularinjury to the plexus. Bony spicules and callusesformed from fractured clavicles can be visual-ized with a simple shoulder x-ray and couldindicate laceration or compression of the un-derlying subclavicular plexus. Deep inspiratoryand expiratory chest x-rays or, preferably, flu-oroscopy of the diaphragm can provide vitalinformation about the function of the phrenic


nerve and its involvement in the injury. This isimportant when the ipsilateral phrenic nerve isbeing considered as a motor donor. Moreover,if the intercostal nerves are to be harvested forneurotizations, one has to be sure that theipsilateral phrenic nerve is functioning.

CT/Myelography

Many reports have been published regardingthe value of combined myelography and cervi-cal CT.37–39 Since the first application of my-elography by Murphy et al. in 194740 to evalu-ate traction injuries to the plexus, thistechnique has been modified with the adventof better contrast materials, and its value hasbeen enhanced when it is correlated with a CTscan of the cervical spine. The positive predic-tive value of combined CT/myelography hasbeen found to be more than 95 percent. In areview article by Walker et al.,41 it is stated thatof 76 cervicothoracic levels imaged (C5 to T1),72 were adequately visualized. Nerve root avul-sion or preganglionic disruption was shown at21 levels. Associated pseudomeningoceles ordeformities of the subarachnoid space werevisualized in 12 of the 21 avulsion levels (57percent). Subsequent surgical exploration andintraoperative somatosensory-evoked poten-tials showed complete preganglionic nerveroot avulsion in 22 levels. Only one of thecomplete avulsions revealed by surgery was notincluded on the patient’s CT/myelogram. Ofthe 21 imaged levels, 20 were correctly revealedon CT/myelography. The overall sensitivity ofthe diagnostic modality was found to be 95percent and the specificity 98 percent. Themain advantage of the CT/myelography is thevisualization of pseudomeningoceles, whichare usually the result of meningeal tears andsubsequent scarring following root avulsion. Itis better to perform the myelography at least 1month after the injury to allow for the pseudo-meningocele to be sealed and to prevent thedye from flowing freely to the surroundingspaces. The decision to subject a patient to aninvasive procedure such as the myelogramshould be made in conjunction with the deci-sion to operate. To achieve better definition ofthe lesions, water-soluble contrast is used withcomputed tomography. The cervical CT scanmight reveal the absence of rootlets from thecorresponding spinal level; this indicates rootavulsion. To better use the results of such astudy, one should consider that the presence ofpseudomeningoceles is strongly indicative of

avulsion; however, avulsed roots can exist de-spite a normal myelogram. Moreover, intactroots can exist in a formed pseudomeningo-cele42 if the traction force is strong enough tocreate a tear in the dura mater but not strongenough to avulse the root from the spinal cord.Because neuroradiology is not pathogno-monic, many authors no longer use CT/myelography.43 Despite reports that claim thatCT/myelography is unable to delineate indi-vidual nerve rootlets,44,45 in our institution weconsider high-resolution CT/myelographywith thin, contiguous axial sections to be veryuseful in the preoperative evaluation of theinjured patient.

Magnetic Resonance Imaging

MRI has the advantage of good visualizationof the brachial plexus beyond the spinal fora-men. Many reports have advocated the useful-ness of this method in establishing a diagnosisin distal plexus lesions.46,47 High field strengthMRI with multiplanar views can easily distin-guish the nerves at the distal plexus from thesurrounding vessels and muscles. On the otherhand, good visualization and delineation of theintradural portion of the rootlets are difficult,mainly because of technical deficiencies. Tech-nically, a conventional MRI cannot providegood anatomic depiction of root sleeves andnerves because of the insufficient contrast be-tween the subarachnoid space and the neuralstructures, a problem caused mainly by cere-brospinal fluid pulsatility.48 Moreover, correla-tion of the intradural surgical findings withconventional MRI studies is not reliable in thepreoperative diagnosis of root avulsions in 48to 52 percent46,47,49 of patients. This is mainlybecause of partial root avulsions and intraduralfibrosis. Other studies, however, have esti-mated the overall sensitivity of magnetic reso-nance imaging to 81 percent.50 The diagnosticproblems related to conventional MRI can bebypassed with the use of the more advancedthree-dimensional magnetic resonance my-elography, as advocated by Gasparotti et al.,51

who claims that this new technique has 89percent sensitivity, 95 percent specificity, and92 percent diagnostic accuracy. In contrastCT/myelography provides a reliable preopera-tive diagnosis that correlates with the intraop-erative diagnosis in more than 95 percent ofpatients.52


Electrophysiologic Studies

Electrophysiologic studies should be carriedout by someone experienced in brachial plexusand peripheral nerve injuries. Axonal disconti-nuity results not only in predictable pathologicfeatures but also in time-related electricalchanges that parallel the pathophysiology ofdenervation. Wallerian degeneration results inthe emergence of spontaneous electrical dis-charges or fibrillations that will appear at least3 weeks after the injury. Therefore, a needleelectromyogram should be postponed for atleast that long. In addition to denervation po-tentials (fibrillations), a needle electromyo-gram can also elicit larger potentials (sharppositive waves). These are valuable when thereis a question whether muscle denervation iscomplete or if there is some attempt at rein-nervation. Needle electromyogram of theparaspinal muscles, which are innervated bythe dorsal rami of the spinal roots, should alsobe routinely performed; denervation of thesemuscles provides strong evidence of avulsion ofthe corresponding roots.53 On the contrary, ifthese muscles are electrically intact, then theinjury is most likely infraganglionic and theroot is most likely ruptured. Of course, volun-tary potentials in different limb muscles indi-cate some neuromuscular continuity. As Bon-ney and Gilliat demonstrated in 1958,54 inaddition to motor conduction studies, sensoryconduction velocities should also be recordedto differentiate between ruptured and avulsedspinal roots. If, in a flail anesthetic arm, normalsensory conduction velocities are elicited, thisis a bad prognostic sign that implies root avul-sion, which makes spontaneous nerve regener-ation impossible.

Many surgeons find the use of intraoperativesomatosensory-evoked potentials useful to ver-ify a suspected avulsion of a root or to deter-mine whether resection of a neuroma and in-terposition nerve grafting should beperformed.55–57 The advocates of this intraop-erative electrophysiologic technique believethat root avulsion is definitely excluded only ifdirect stimulation of the individual surgicallyexposed cervical nerve root elicits reprodu-cible cortical somatosensory-evoked poten-tials.58 Others believe that intraoperative trans-cranial electrical motor-evoked potentials canbe of use in assessing the connectivity of theroots to the spinal cord.59

However, the false positive and false negative

recordings are quite high; in addition, this typeof study is time consuming and susceptible toelectrical interferences in the operating the-ater.60,61 In our institution, we routinely per-form intraoperative electrical stimulation ofthe brachial plexus elements with a DC stimu-lator at 0.5, 1.0, and 2.0 mA.

SURGICAL PLAN, INTRAOPERATIVE DIAGNOSIS, AND

RECONSTRUCTION

The patient is cleared medically and the op-eration is performed under general anesthesia.Special considerations include the use of lightanesthetics, such as sufentanil, and avoidanceof any kind of paralytics, so that intraoperativeelectrical stimulation studies can be per-formed. The sterile field includes the affectedarm, both sides of the neck and bilateral shoul-ders (in case we need to use contralateral mo-tor donors), the anterior and posterior chest tothe midline, and bilateral lower extremities.The incision for the exploration of the brachialplexus is usually parallel to the posterior bor-der of the sternocleidomastoid, then turns lat-erally over the clavicle and continues along thedeltopectoral groove. Some surgeons performclavicular osteotomies to improve exposure.We do not sacrifice the clavicle or the pectora-lis minor to gain exposure to the infraclavicu-lar plexus. Open clavicular osteotomies havethe tendency to create malunions and, moreoften, nonunions. Therefore, they should beavoided. The omohyoid muscle and the trans-verse cervical vessels are identified and re-tracted. The phrenic nerve is identified in itsusual anatomic position in the anterior borderof the anterior scalene muscle, and it is tracedsuperiorly to its C4 and C5 origins. Thephrenic nerve is stimulated to verify its integ-rity, and the movements of the diaphragm arerecorded. Sometimes the phrenic nerve is in-volved in adhesions and is very tightly fixatedto spinal nerve C5, which is the first root to beexplored. Care should be taken to isolate it andpreserve it, except in cases in which it is rup-tured; in these cases, if the distal stump cannotbe identified, the proximal stump should beused as a donor for neurotization procedures.Then the surgeon should continue with iden-tification and stimulation of the rest of theroots. Neuromas are identified and can extendbetween the upper roots and the trunks andalso distally at the cord or peripheral nervelevel. Neuromas can be in continuity or at theend of a ruptured plexus segment; the latter


can be resected back to normal tissue untilnormal fascicles are visualized. The response ofthe patient to the resection of a neuroma canalso indicate connectivity of the correspondingroot to the spinal cord. If the vital signs of amildly anesthetized patient rise during resec-tion of the neuroma, then one can concludethat the root is in continuity with the spinalcord and not avulsed. This is an importantdetermination, because then the proximalstump of the neuroma could safely serve as anintraplexus motor donor for the reconstruc-tion of the distal plexus. Spinal nerves that feel“empty” to palpation and are pale in appear-ance and negative to electrical stimulation areusually avulsed. Simultaneously biopsies can beobtained for carbonic anhydrase histochemis-try62 or cholinesterase staining. Furthermore,frozen biopsies of a root can determine theamount of scar present and the absence orpresence of dorsal root ganglion cells, whichindicate a nonreparable lesion. The enzymecarbonic anhydrase can differentiate betweenmotor and sensory fascicles of peripheralnerves.63 Acetylcholinesterase histochemistryhas also been used in conjunction with periph-eral nerve surgery because this enzyme can betransported by axonal flow from the cell bodiesto the nerve terminals and has higher activitiesin a motor fascicle than in a sensory one.64,65

However, prolonged incubation time (morethan 24 hours) needed to visualize the acetyl-cholinesterase in motor fibers makes this stain-ing technique nonpractical because it meansthat the patient is subjected to two operativeprocedures within a 2-day period. Intraopera-tive nerve action potentials can also assist indistinguishing ruptured from avulsed rootsand are used by many authors.66 The seniorauthor prefers to use electrical stimulation ofthe identified nerve structures with a DC stim-ulator at 0.5, 1.0, and 2.0 mA of amplitude.This is especially useful in lesions in continuity,because a decision has to be made to proceedeither with neurolysis or resection and interpo-sition nerve grafting.

After exploration, identification of the level,type, and extent of the lesion takes place, andan intraoperative plan is established. Motorand sensory donors are matched to their cor-responding distal targets. Intraplexus motordonors, such as the proximal stumps of rup-tured roots, yield, in general, better resultsbecause they carry a higher number of axons.Extraplexus donors, such as intercostal nerves,

the accessory nerve, and the contralateral C7root, are often used in multiple-root avulsions.

Microneurolysis can be used when the ex-plored elements of the plexus are hard onpalpation, with dense epineurium that com-presses the axons within and with the longitu-dinal vasa nervosum compromised. To relievethe intraneural pressure, longitudinal epineu-riotomies can be performed under the opera-ting microscope with a 45-degree-angle dia-mond knife. These can extend to the perineur-ial and interfascicular level as indicated. Bulg-ing of the entrapped fascicles stronglyindicates that microneurolysis is effective.

Supraclavicular postganglionic lesions (neu-romas secondary to root ruptures) are usuallytreated with a combination of microneurolysisand interposition nerve grafting. Excision ofthe neuroma to healthy fascicles distally andproximally is of great importance. While excis-ing a partially ruptured lesion, one has to becareful not to downgrade existing function andconvert it to a complete one. Direct coaptationof nerve roots or cords following excision ofneuromas is almost impossible, except in earlycases of penetrating trauma involving the bra-chial plexus. In all other cases, interpositionnerve grafting is indicated. Sural nerves areusually harvested through three or four smallincisions on the posterolateral aspect of thecalf. Some authors also advocate the endo-scopic harvesting of sural nerves.67 Bilateral sa-phenous nerves can also be harvested and usedas interposition nerve grafts.

Infraclavicular lesions should be explored intheir entirety. Healthy proximal and distalstumps should be identified. A combination ofmicroneurolysis and interposition nerve graft-ing bypassing the neuroma can be used forreconstruction. Again, one should be carefulnot to downgrade existing function, especiallyin partially injured elements of the infraclavic-ular brachial plexus. It should also be men-tioned that associated injuries in the infracla-vicular area can result in severe scar formation,making the exploration of the infraclavicularplexus tedious.

Root avulsions always carry the worst prog-nosis and make the reconstruction of theplexus more challenging. A variety of extra-plexus donors should be recruited in thesecases to reconstruct the distal plexus elements.When the upper plexus roots (C5 and C6) areavulsed from the spinal cord, reconstruction ofthe shoulder and elbow function can be


achieved by means of the C7 root, which can besacrificed. In that case, the distal targets inner-vated by the C7 root can be neurotized withextraplexus motor donors. When the three up-per roots (C5, C6, and C7) are avulsed, moreextraplexus motor donors are needed for re-construction. Intercostal nerves, if harvestedproperly, yield acceptable results, especially forreconstruction of the axillary, triceps, or mus-culocutaneous nerve.68 Intercostal nerves arechallenging and time-consuming to harvest,and each one may yield approximately 1200 to1300 myelinated fibers. When neurotization isplanned using a larger nerve, such as the mus-culocutaneous nerve, which is made of approx-imately 6000 fibers, at least three intercostalnerves need to be used for this neurotization toyield an acceptable result.69 The musculocuta-neous nerve can also be reconstructed withmotor donors from the pectoral nerve,70 thedistal accessory, or the contralateral C7 ante-rior division fibers.71 The distal part of thespinal accessory nerve is routinely used in ourcenter to reconstruct the suprascapular nerve,either by direct end-to-end coaptation or byinterposition nerve graft. In this manner, rais-ing of the shoulder is still possible because onlythe distal portion of the trapezius muscle isdowngraded. Other extraplexus donors in-clude the contralateral C7 and the phrenicnerve.

It should be emphasized that even if extra-plexus donors have resulted in obtaining use-ful muscle function at the M3 level, there arelimitations, particularly in the number of nervefibers they carry. Studies in animal models haveshown that the number of axons in the donornerves was partially correlated with functionalrecovery of the targeted muscle; the more ax-ons in the donor nerve, the better the chancefor functional recovery.72,73 In a study per-formed in our center,74 the myelinated axonsof all common extraplexus nerve donors inseven fresh cadavers and the donor nerves usedin 50 actual clinical cases were counted. Theresults were correlated with the functional re-turn of the reconstructed muscles. It was foundthat the mean number of axons was 2145 in theaccessory nerve, 1093 in each intercostal nerve,1756 in the phrenic nerve, and 893 in thecervical plexus donors; the number of fibersranged between 5000 and 12000 in the con-tralateral C7 posterior or anterior division. Thepercentages of muscle function greater thanM3 for each corresponding motor donor were

84 percent for the accessory nerve, 52 percentfor the ipsilateral intercostal nerves, 45 percentfor the cervical plexus motor donors, 40 per-cent for the phrenic nerve, and 51 percent forthe contralateral C7. One should notice that,despite the higher number of motor fibers inthe C7 root anterior and posterior divisions ascompared with the accessory nerve, only halfthe muscles neurotized by the contralateral C7root achieved a muscle grade above M3, ascompared with the 84 percent recoveryachieved in the supraspinatus muscle by neu-rotization from the accessory nerve. This canbe explained by the fact that most neurotiza-tions75 from the XI cranial nerve were directend-to-end repairs, whereas the neurotizationsby means of the contralateral C7 necessitatedthe use of long interposition nerve grafts. Fur-thermore, direct neurotization of a nerve lead-ing to a single target, such as the suprascapularnerve, is preferential to neurotization of anerve that leads to multiple muscle targets,such as the median nerve.

In cases of combined-root rupture avulsion(C5/C6 rupture and C7/C8/T1 avulsion or C5rupture and C6/C7/C8/T1 avulsion), the C5root, especially in a prefixed plexus, can beused for reconstruction of the musculocutane-ous nerve and part of the median nerve. Again,the suprascapular nerve is reconstructed fromthe distal accessory nerve. In these cases inwhich the lower roots are avulsed (C8 and T1),hand function is not expected to recover spon-taneously, and thus the ulnar nerve can beharvested as a free or pedicled vascularizedgraft and used for reconstruction. The benefitsof a vascularized nerve graft are achieved, in-cluding its use in a scarred bed. Moreover, itprovides the best milieu for rapid axonalgrowth. Also, the possibility of scarring at thedistal coaptation site is minimized because offaster regeneration and better graft vasculariza-tion.

Our strategy regarding the use of the ulnarnerve as a vascularized trunk graft is the follow-ing: If C5 and C6 or a large C5 root (as in aprefixed plexus) are ruptured, but the proxi-mal stumps are in continuity with the spinalcord, with simultaneous lower root avulsions,then during the initial brachial plexus recon-struction, the ipsilateral ulnar nerve is used asa free or pedicled vascularized graft to recon-struct the musculocutaneous, the median, and,on occasion, the radial nerve. The vascularizedulnar nerve graft is based on the superior ulnar


collateral artery and vein. If the C6 root is alsoavulsed and the C5 root is flat, pale, and elec-trically nonresponsive, making it inappropriatefor reconstruction, then ipsilateral extraplexusdonors are used, including partial hypoglossal,partial phrenic, distal accessory, and ipsilateralintercostal nerves. In these cases, the musculo-cutaneous nerve is reconstructed with a mini-mum of three intercostal nerves, the distal ac-cessory is given to the suprascapular nerve, andthe axillary nerve is neurotized with additionalintercostals or with other extraplexus motordonors. During the second stage, the ulnarnerve is harvested as a vascularized interposi-tion nerve graft to neurotize the median nervefrom the anterior division of the contralateralC7 root. The triceps branch and/or the radialnerve can be neurotized from motor fibersfrom the posterior division of the contralateralC7 root. This plan of reconstruction minimizesthe possibility of co-contractions of the bicepsand triceps, allowing for easier patient re-education during the rehabilitation phase. Inour patient population, co-contractions be-tween antagonistic muscles are very rare, be-cause of careful preoperative and intraopera-tive planning. However, when co-contractionsoccur, they are diagnosed easily with the clini-cal examination and with needle electromyo-graphy and are usually treated with negativebiofeedback training.

During the first or second stage of brachialplexus reconstruction, we try to leave twobanked nerve grafts at the medial and lateralsides of the antecubital fossa for future free-muscle transplantation. In cases of global rootavulsion (five avulsions), intraplexus donorsobviously are not available, and all reconstruc-tion is carried out from extraplexus donors.The primary goal remains the restoration ofshoulder function and stability, elbow flexion,and hand sensation. Banked nerves should beleft for future free-muscle transfer for handreanimation if the patient is judged to be agood candidate. Global avulsion of the bra-chial plexus carries the worst prognosis amongall types of plexus injuries.

The morbidity from harvesting the variousextraplexus donors can be minimized by tak-ing only the number of donor fibers needed,such as in partial phrenic or partial hypoglossaltransfers, which are used in combination withan end-to-side coaptation and by using intra-plexus root donors that have adequate overlapfrom neighboring myotomes, such as the con-

tralateral C7 root. Thus, when the contralateralC7 root (either the posterior or anterior divi-sion) is harvested, the senior author performsselective neurectomies, to match the cross-sectional area of the interposition cross-chestnerve graft (saphenous nerves or vascularizedulnar nerve).71 Even if one needs to harvest asmuch as 80 percent of the anterior or posteriordivision, motor morbidity can be avoided. Inour patient population of nearly 50 cases withcontralateral C7 transfers, we encounteredtransient hypoesthesia only in the hand, in thedistribution of C7 dermatome, which disap-peared by 3 months. At 6 months, there was nodiscernible motor or sensory deficit in the con-tralateral normal limb. Similarly, when harvest-ing the phrenic nerve, the senior author pre-fers to perform a partial neurectomy combinedwith end-to-side coaptation, thus preservingthe innervation to the diaphragm. Moreover,we avoid harvesting both the phrenic and in-tercostal nerves in the same patient to preventthe possibility of respiratory compromise.

At the completion of surgery and before ex-

FIG. 1. A 37-year-old man who sustained a right brachialplexus injury after falling from a vertical ladder and landingon his right shoulder. The patient could not abduct andexternally rotate his shoulder.


tubation, a custom-made brace with a halo isapplied to the patient. This brace keeps thearm abducted 45 degrees and in anterior flex-ion. The halo prevents lateral movement of thehead, reducing the risk of injury to the coap-tation sites. This type of immobilization is ex-tremely important, especially in cases in which

the repairs have been executed very proximallynear the transverse foramina. The brace is re-moved 6 to 8 weeks later, and the patient’s armis placed in a sling. The borders of the treatedarea are marked with methylene blue, and spe-cific instructions are given to the nursing per-sonnel and the family not to touch this area so

FIG. 2. Preoperative brachial plexus chart of the patient in Figure 1.

FIG. 3. (Left) Intraoperative diagnosis. Scarring of the upper trunk and the suprascapularnerve. Both structures were hard on palpation. (Right) Microneurolysis of the upper trunk andsuprascapular nerve (SS), posterior cord (PC), and lateral cord (LC). The fascicles werebulging following the microneurolysis. ax, axillary nerve; MC, musculocutaneous nerve; M,contribution to median nerve.


that the underlying coaptations will not betraumatized accidentally.

Complications of first-stage reconstructionare rare and include mainly localized woundinfection and, very rarely, pneumothorax dur-ing intercostal nerve harvesting. Seroma for-mation after intercostal nerve harvesting, as aresult of the extensive dissection, has been ob-served at times and can be prevented by com-pression dressing and frequent observation.

POSTOPERATIVE CARE

Physical therapy with passive range of mo-tion and slow-pulse electrical stimulation76 arestarted with the removal of the brace. Ultra-sound therapy for scar manipulation is alsorecommended for separating the nerve heal-ing from the overlying skin wound. Physicaltherapy begins at 8 weeks postoperatively withpassive range of motion and manual massageover the treated areas. Of great importance isthe training of the patient and the physicaltherapist in all aspects of the new neurotiza-tions. For instance, if intercostal nerves havebeen used to reinnervate the musculocutane-ous nerve, the patient needs to perform theValsalva maneuver while attempting to flex the

elbow to achieve stronger contraction. Our ex-perience shows that following such neurotiza-tions in motivated patients, central nervous sys-tem neurons exhibit extreme plasticity, andthe patient quickly embraces the new nervepathways. Furthermore, the motor cortex map-ping resembles the physiological biceps area.77

The physical therapist should also be aware ofdeveloping co-contractures in the reinnervatedmuscles and treat them aggressively with nega-tive or positive biofeedback as needed.

The first follow-up visit is arranged at 6months after surgery. The patient is instructedto monitor any improvement in function andto document specific dates of recovery. Duringthe first follow-up visit, a needle electromyogramis obtained. At 12-month follow-up, the patient isevaluated fully for return of function, and plansfor secondary reconstructions are made.

Specific Aims of Reconstruction

When deciding which target muscles are to bereinnervated, the following rules apply:

1. Stability of the shoulder is a very impor-tant goal. The return of function in thesupraspinatus and deltoid muscles is con-

FIG. 4. Four months postoperatively the patient has excellent external rotation, shoulderabduction, and elbow flexion.


sidered of prime importance. Unstable andsubluxated shoulders very often cause ipsi-lateral neck pain due to the gravitationaleffect of the flail arm. Shoulder fusion is

never advocated in our center, especially ifno adequate scapulothoracic motion ispresent.78

2. One of the most important functions of

FIG. 5. Postoperative brachial plexus chart of the patient in Figure 1.

FIG. 6. A 26-year-old man from Greece sustained a left brachial plexus palsy secondaryto a motorcycle accident. He presented with a totally flail left upper extremity. Note thescars in the upper arm from the open reduction with internal fixation of the humerusand the prominent muscle wasting.


the upper extremity is bringing the handto the mouth, for obvious reasons (nour-ishment). Unless elbow flexion is re-stored, even hands that have been sparedfrom injury are useless. Therefore, resto-

ration of the biceps is of great importanceto the overall function of the upper ex-tremity, and during surgery we try to pro-vide the musculocutaneous nerve with thebest motor donors. To ensure good bi-


FIG. 8. Exploration of the supraclavicular and infraclavicular plexus 4 months after theinjury. The intraoperative diagnosis included ruptured roots (C5 to T1) with neuromaformations at the trunk, cord, and infraclavicular levels. The radial (R), ulnar (U), andmedian (M) nerves were also injured at the mid-humeral level with extensive scar for-mation. SS, suprascapular nerve; AX, axillary nerve; MC, musculocutaneous nerve.


ceps function, we also have to ensure thatthere are no contractures at the elbowlevel that might hinder its full range ofmotion. Furthermore, during the initialnerve reconstruction, one or two of theupper intercostal nerves are dedicated toneurotize the ipsilateral latissimus dorsimuscle as a reserve muscle unit to eventu-ally enhance the power of a weakenedrecovering biceps. If the biceps restora-tion is complete, then the reinnervatedlatissimus can be used as a pedicled mus-cle flap to strengthen the function of theipsilateral triceps muscle.

3. The senior author strongly believes thatreinnervation of the triceps is also ofprime importance because it can give sta-bility to the elbow joint. It is also true thatmany patients, even if they regain power-ful wrist extension or finger flexion, arenot able to use their fingers optimally indaily activities because their elbow is un-stable.79 Many authors,80,81 however, donot attempt to neurotize the triceps, butactually many times sacrifice the muscleand use it as a pedicled flap to substitutefor biceps function.

4. The median nerve also needs to be re-stored for both sensory protection82 andfinger flexion (the latter is especially truefor short denervation times). Sensory pro-tection of the hand is of prime impor-

tance in global brachial plexus injuriesand can be achieved by neurotization ofthe median nerve from sensory intercostalnerves or from supraclavicular sensorynerves.

5. In case of lower root involvement, it isprudent to leave banked nerves at theelbow level, neurotized either by ipsilat-eral donors or by the contralateral C7root, using the selective contralateral C7technique introduced by Terzis71 for fu-ture free-muscle transplantation for handreanimation.

The above-mentioned priorities in restor-ative microsurgery, especially if the denerva-tion time is short, can achieve shoulder stabili-zation, a certain degree of abduction andexternal rotation, strong elbow flexion, elbowextension, hand protective sensation, and, inselected cases, reanimation of the hand.

Free-Muscle Transplantation

Functioning free-muscle transplantation hasbecome a reconstructive option in cases of de-layed patient presentation with long denerva-tion time, or in cases with multiple-root avul-sions, especially when the lower roots of theplexus that innervate the hand are involved.Furthermore, this microsurgical technique canbe extremely useful when primary reconstruc-tion of the biceps or forearm musculature has

FIG. 9. The reconstruction included extensive microneurolysis of C5, C6, C7, C8, and T1;reconstruction of the radial (R) and axillary (AX) nerves from the posterior cord (PC) withinterposition nerve grafts; nerve graft reconstruction of the suprascapular nerve (SS) from C5spinal nerve; and reconstruction of the distal median (M) and musculocutaneous (MC) nervesfrom their more proximal parts with sural nerve grafts.


not yielded satisfactory results. Currently, free-muscle transfers are performed routinely forrestoration of elbow flexion83,84 and reanima-tion of the hand.85–88 A variety of nerve andmuscles can be used for these purposes. In ourcenter, the latissimus dorsi (ipsilateral if sparedfrom the injury or reinnervated or contralat-eral) and the rectus femoris muscles are con-sidered the optimal muscles to provide elbowflexion or elbow extension. Nerves to neurotizethese muscles can be banked nerves from ipsi-lateral motor donors, such as the eleventh cra-nial nerve, or intercostal nerves or cross-chest

nerve grafts from the contralateral anterior orposterior divisions of the C7 root. With such atransfer, a functional outcome of at least M3strength (antigravity) can be achieved in themajority of the cases (78 percent).83 Similarly,the gracilis muscles can be transferred to theforearm as free-muscle flaps to restore fingerflexion or extension or to substitute for intrin-sic muscle function. Electrophysiologic assess-ment of the free muscle can show contractionsas early as 2 to 4 months,86 and full range ofcontraction can be expected at approximately6 to 12 months. Of great importance is thecondition of the corresponding joints; thus,before any muscle transfer, the adjacent jointsneed to be free of contractures, otherwise onecould end up with a strong muscle unit but noresultant joint excursion. The success of a free-muscle transfer relies on the muscle trans-ferred, the donor nerve selection, and the com-pliance of the patient. Operative techniqueand postoperative care are equally importantfactors in achieving success. Placement of themuscle at optimal tension and adequate flapcoverage, along with early aggressive passivemotion exercises, allow for grip strength of 50percent of normal, in case of the transplanta-tion of a gracilis muscle to the forearm forfinger flexion.89

Other Secondary Procedures

These procedures are used as adjuvant pro-cedures in patients who have already had bra-chial plexus reconstruction with appropriateneurotizations, to augment specific functions.They are also of great importance in patientswith long denervation times, who are not goodcandidates for primary microsurgical recon-struction. These include tendon transfers,pedicled muscle transfers, joint fusions, and avariety of osteotomies.

During tendon transfers, the decision has tobe made to sacrifice one function to replace orenhance another function. The tendons to betransferred need to be adequately innervatedand are selected by clinical examination andneedle electromyogram. A common tendonprocedure is transfer of the clavicular and ac-romial insertion of the trapezius with or with-out fascia lata to the deltoid insertion on thehumerus to enhance abduction and reversesubluxation.90 Other transfers around theshoulder area include transfer of the sternocla-vicular part of the pectoralis major to the an-terior deltoid for anterior flexion or transfer to

FIG. 10. Six years postoperatively, the patient is able tofully abduct (above) and anteriorly flex his arm (center). He hasfull external rotation (below).


the native biceps for elbow flexion.91 The pec-toralis minor can also be transferred to thebiceps muscle to strengthen elbow flexion.92 Aclassic maneuver is also the rerouting of thelatissimus dorsi tendon around the humerus to

restore dynamic external rotation. In the expe-rience of the senior author (J.K.T.), this par-ticular transfer yields better results in childrenthan in adults. At the elbow level, Steindlerflexorplasty, reattachment of the flexor mass of

FIG. 11. Six years postoperatively, the patient has full elbow flexion (left and center). He is able to grasp and lift a chair abovehead level (right).



the antebrachium proximally onto the hu-merus, can be used also to enhance elbowflexion, alone or simultaneously with othertransfers.93 The latissimus dorsi muscle hasbeen used extensively for restoration of elbowflexion as a pedicled myocutaneous flap.75,94,95

Reanimation of the hand and wrist can alsobe improved with secondary musculotendi-nous transfers using local reinnervated tar-gets or targets that have escaped injury. Com-mon transfers include the flexor carpiulnaris or flexor carpi radialis to the extensordigitorum communis for strengthening fin-ger extension, and the extensor carpi radialisbrevis, extensor carpi radialis longus, or bra-chioradialis to the abductor pollicis longusor extensor pollicis longus for abduction orextension of the thumb. Profundoplasty, inwhich the missing portion of the flexor digi-torum profundus is connected to the func-tioning part of the flexor digitorum profun-dus, can be also performed for enhancementof finger flexion.

Rotational osteotomies of the humerus help

improve external rotation by 30 degrees andare carried out in collaboration with our ortho-pedic colleagues. Rotational osteotomies of theradius have also been used in our center tocorrect supination or pronation deformities.Wrist fusions are usually performed simulta-neously with a free-muscle transfer for handreanimation. This allows a far more accuratesetting of tension in the transferred muscle.

Despite reports in the literature,78,96 we donot advocate fusion of the shoulder as a ther-apeutic modality. On the contrary, we believethat the primary microreconstruction of thebrachial plexus, especially if it is done in atimely manner, will restore satisfactory shoul-der stability and function in the majority of thepatients.75

PAIN MANAGEMENT IN BRACHIAL PLEXUS INJURIES

Injury to the brachial plexus can cause se-vere pain. The incidence varies with each studybut affects about 10 to 20 percent of patientswith brachial plexus injuries and almost 40

FIG. 13. A 13-year-old boy sustained a right global brachialplexus palsy following a motor vehicle accident. The patientpresented with a totally flail right upper extremity.

FIG. 14. A preoperative myelogram shows two largepseudomeningoceles at C8 and T1 levels and a smaller oneat C7 level.


percent of patients with avulsions, especiallywhen the lower C8 and T1 roots are involved.97

This is caused primarily by the high content of

sensory fibers in the lower roots. However, theincidence of intractable pain associated withlower root avulsion in our series relatively is low


FIG. 16. (Left) The patient underwent an exploration of his supraclavicular and infraclavicularright brachial plexus 11 months after the injury. The intraoperative diagnosis was avulsions of C6 toT1 roots. (Center and right) Reconstruction included harvesting of the (1) right ulnar nerve as a freevascularized nerve graft, (2) bilateral sural nerves, and (3) ipsilateral intercostals T5 to T7; recon-struction of the median (M) and musculocutaneous (MC) nerves from C5 with the vascularized ulnarnerve graft; reconstruction of the axillary nerve (Ax) from C5 with sural nerve grafts; direct end-to-endcoaptation of the suprascapular nerve (SS) with the accessory nerve (Acc); reconstruction of thetriceps and thoracodorsal nerves from ipsilateral intercostals; and coaptation of one intercostal nerveto a sural nerve that was left as a banked nerve to the right upper extremity for future free-muscletransplantation for hand reanimation. R, radial nerve.


(12 out of 263 patients).75 Pain usually starts afew days after the pain from the initial traumasubsides and can be intractable. It is commonlydescribed as continuous, burning, and com-pressing and is frequently located in the hand.Pain can be aggravating and can raise manypsychosocial issues for the patient and his orher immediate family. The few studies of painfollowing brachial plexus injury98,99 show thatthis type of pain is very difficult to treat. Mostpatients are initially treated with various nar-cotics, tricyclic antidepressants, antipsychoticdrugs, and sympathetic blocks. It has beenshown, however, that there is a statistical cor-

relation and temporal relationship betweenthe reduction in pain and returning of motorfunction100,101; this is also supported by obser-vations in our patient population.76 In twocases of intractable pain that were encounteredin our series, the intensity of the pain was suchthat the senior author recommended to thepatients to undergo a dorsal root entry zoneprocedure first, before proceeding with theformal brachial plexus reconstruction. Thebenefit of following this sequence of events isthat, if this procedure is successful, the patientcan focus on his rehabilitation without distrac-tion from the pain. It also allows the recon-

FIG. 17. Six years postoperatively, the patient demonstrates 90-degree abduction (above, left),90-degree external rotation (above, right and center, left), and full elbow flexion (center, right) andis able to carry objects with his hand (below).


structive microsurgeon access to the most ac-curate preoperative diagnosis of the status ofthe roots. The dorsal root entry zone (DREZ)procedure was introduced in 1979 by Nasholdand colleagues,102 whose long-term follow-upstudy of their results showed that with this neu-roablative procedure, 54 percent of patientswith avulsion injuries were afforded good painrelief after 34 months. Other studies showedincreased numbers of pain relief as high as 79percent.103 Despite evidence in the literaturepertaining to limb amputation as a choice forrehabilitation and pain control,104 we stronglydisagree with this procedure, because pain as-sociated with preganglionic injury of the bra-chial plexus is not relieved by amputation. It isa well-known fact that the best treatmentagainst amputation neuromas is replantationof the digit. In our patient population, we havenever had to perform limb amputation forpain control. On the contrary, the early explo-ration and reconstruction of the brachialplexus, in the majority of the cases, not onlyimprove the function of the arm but also re-lieve the postavulsion pain and the pain relatedto the instability of the shoulder. In our patientpopulation, following exploration and recon-struction of the brachial plexus, relief of thepain occurred in the majority of the patients.For all types of injury, the mean pain score wassignificantly lower (p , 0.05) than it was beforethe surgery.75

CASE REPORTS

Case 1: Supraclavicular Traction InjuryB.E., a 37-year-old white man, fell from a vertical ladder on

concrete and landed on his right shoulder. He presented withdeltoid, supraspinatus, and infraspinatus paralysis and wasunable to abduct and externally rotate the arm (Figs. 1 and2). The function of the hand was normal. Preoperatively healso had mild weakness of his biceps muscle and numbnessaround the shoulder area. The patient did not suffer otherassociated injuries but complained of severe pain. Denerva-tion time was 2 months. He was initially operated on inanother center, where minimal exploration of the plexus wasperformed that revealed thickening of the suprascapularnerve. No further treatment was carried out at that time. Afterreferral to our center, the patient underwent an explorationof the supraclavicular plexus that revealed extensive harden-ing and scarring, primarily around the upper and middletrunks, with involvement of the suprascapular nerve. Exten-sive microneurolysis with a diamond knife under the oper-ating microscope was performed. Bulging of the releasedfascicles was noted immediately (Fig. 3). Postoperatively, thepatient has excellent return of abduction, external rotation,and anterior flexion of the arm (Figs. 4 and 5). He returnedto his previous occupation 5 months after surgery.

Case 2: Multiple Supraclavicular and Infraclavicu-lar Ruptures

V.A., a 26-year-old white man from Greece, was involved ina motorcycle accident. He sustained a complete left brachialplexus palsy. Associated injuries included open fracture of thehumerus, treated with an external fixator; fractures of theradius and ulna, treated with open reduction with internalfixation; and fractures of the wrist and second metacarpal,which were reduced with Kirschner wires. Two months later,the external fixator was removed and the humeral fracturewas fixed with internal plating and autologous bone grafting.Four months after the injury, the patient was referred to ourcenter. A preoperative CT/myelogram revealed no avulsionsof all cervical roots, and a preoperative electromyogramshowed multiple-root ruptures. Clinically, the patient had noabduction, external rotation, elbow flexion, or elbow exten-sion, and hand function was graded as 0 (Fig. 6). He also hadtotal anesthesia below the elbow level and severe hypesthesiaabove the elbow (Fig. 7). Total denervation time was 4months. The patient underwent exploration of the supracla-vicular and infraclavicular plexus and the entire arm. Therewere four levels of injury: at the supraclavicular, subclavicular,and infraclavicular areas, and in the arm. Intraoperative find-ings included (1) rupture of the upper and middle trunkinvolving roots C5, C6, and C7 with a large neuroma presentand (2) a large neuroma underneath the clavicle, corre-sponding to the suprascapular nerve and the posterior cord,with sparing of the thoracodorsal nerve. In the infraclavicularregion, the axillary, median, and musculocutaneous nerves wereruptured, and multiple neuromas and excessive scarring werepresent. The radial, median, and ulnar nerves were rupturedand extensively scarred at the mid-humeral level, resulting inlarge nerve gaps; this injury was associated with a left humeralfracture (Fig. 8). The reconstruction included the following(Fig. 9):

• Extensive microneurolysis of C5, C6, C7, C8, and T1 roots.• Reconstruction of the suprascapular nerve with two sural

nerve grafts (6 cm) from the C5 root.• Reconstruction of the axillary nerve from the posterior

cord with three sural nerve grafts (12 cm).• Reconstruction of the radial nerve with four sural nerve

grafts (13 cm) from the posterior cord.• Reconstruction of the median nerve involved three sural

nerve grafts 35 cm long, from the proximal stump of themedian nerve in the infraclavicular area to the distalstump of the median nerve in the elbow region.

• Reconstruction of the musculocutaneous nerve involvedthree sural nerves grafts 15 cm long, from the exit of themusculocutaneous nerve from the lateral cord to its entryinto the biceps muscle.

Six years postoperatively, the patient has anterior shoulderflexion of 120 degrees (pectoralis major 41), abduction of 90degrees (supraspinatus and deltoid graded as 4), latissimusdorsi of 41, external rotation of 90 degrees (infraspinatus31), full elbow flexion (biceps 4), full elbow extension (tri-ceps 4), and protective sensation in the hand with a usefulgross grip (Figs. 10 through 12).

Case 3: Multiple-Root AvulsionsA.T., a 13-year-old white boy, was an unbelted passenger

when he was ejected from a car during a motor vehicle ac-cident. Associated injuries included a Le Fort II fracture,fractures of the radius and ulna, and soft-tissue hematomas.


He was referred to our center with right global brachialplexus palsy. Clinically, the patient had only some trapeziusfunction. All other muscle groups were graded as 0. Thepatient also had a total anesthetic arm (Figs. 13). A preop-erative electromyogram indicated a global plexopathy withavulsions of C7, C8, and T1 roots. A preoperative CT/my-elogram showed pseudomeningoceles at three levels (C7, C8,and T1) and absent rootlets from C6 to T1 (Fig. 14). The totaldenervation time was 11 months (Fig. 15). The patient un-derwent an exploration of the right supraclavicular and in-fraclavicular plexus. The intraoperative diagnosis was avul-sions of C6 to T1 roots (Fig. 16, left). The reconstruction of thebrachial plexus included the following (Figs. 16, center and right):

• Harvesting of: the ulnar nerve as a vascularized free graft;bilateral sural nerves; and T5, T6, and T7 ipsilateral inter-costals.

• Neurotization of the suprascapular nerve directly fromthe distal accessory nerve.

• Direct coaptation of the thoracodorsal to the T5 intercos-tal nerve.

• Direct neurotization of T6 to the triceps branch.• Coaptation of T7 to a sural nerve graft that was tunneled

as a banked nerve to the right upper extremity for futurefree-muscle transfer.

• Reconstruction of the musculocutaneous nerve and themedian nerve through a free vascularized ulnar nervegraft from C5.

• Reconstruction of the axillary nerve from C5 with twosural nerve grafts.

Six years following the reconstruction of the plexus, thepatient is able to abduct his arm 90 degrees, externally rotateit 90 degrees, and fully flex and extend the elbow with strength

(biceps and triceps were graded as M4). He is also able to graspand lift objects with his right hand (Figs. 17 and 18).

RESULTS

Results of different neurotizations have beenpublished by many authors. At least 2 to 3 yearsof follow-up are required to evaluate results.These outcome studies should be reported ac-cording to the different reconstructive meth-ods and techniques used for restoration ofeach function.61

A serious outcome study performed in ourinstitution involved 263 patients operated onby the senior author from 1978 to 199675; 204had adequate follow-up. From this series, it isclear that intraplexus donors consistentlyyielded better results. This is probably true notonly because of the higher fiber numbers en-countered in the proximal stumps of the rup-tured roots but also because it is a more “phys-iologic” reconstruction, which makes thepostoperative recovery of function easier. Mostauthors agree with this statement105 and believethat, even with only two roots available forreconstruction, a good functional outcome canbe expected.

Based on the outcomes when extraplexusdonors were used, direct neurotization of the



suprascapular nerve from the accessory nerveyielded 75 percent good and excellent re-sults.75 The spinal accessory nerve has alsobeen used for restoration of elbow flexion inavulsion injuries, with coaptation to the mus-culocutaneous nerve. Good function of the bi-ceps at the M31 level has been reported formore than 70 percent of the patients.106,107

Intercostal nerves are probably superior toall other extraplexus donors when they areused for reconstruction of the musculocutane-ous or triceps nerves. Different studies havereported good function of the biceps (M31and above) after neurotization with intercos-tals for 63 to 81.8 percent of patients.108,109 Themean time for recovery of function with thisneurotization has been reported to be 12months. On the contrary, the recovery of thedistal radial nerve and the motor part of themedian nerve is not expected to be good. How-ever, the intercostal nerves can restore protec-tive sensibility to the hand in a large number ofpatients. This experience on the use of theintercostal nerves is comparable with the se-nior author’s experience.75

In our center, use of the selective contralat-eral C7 technique as a contralateral intraplexusdonor is strongly advocated.71 When the ante-rior division of this root is connected througha vascularized ulnar nerve graft to the mediannerve or the musculocutaneous nerve, goodresults can be expected in almost 60 percent ofthese patients.75,110 Poorer results can be ex-pected for the radial nerve. In global avulsions,because recovery from the ulnar nerve cannotbe expected, the ulnar nerve can be harvestedas a vascularized nerve graft based on the su-perior ulnar collateral artery.111 This techniquecan yield better results as compared with theuse of sural or saphenous nerve cross-chestgrafts, owing to the rich blood supply of thevascularized ulnar nerve and to the great num-ber of fibers that it can carry. In our practice,we routinely use the contralateral C7 root as amotor donor for banked cross-chest nervegrafts for selected targets or future use forfree-muscle transfers. In the senior author’shands, as has also been reported by other sur-geons,112 only minor tingling and hypesthesiacan be expected at the donor-side hand thatresolve within 6 months.

The ipsilateral cervical plexus motor donorsusually do yield only modest functional results,mainly because of the small number of fibers.On the contrary, the phrenic nerve can pro-

vide acceptable elbow flexion at the M31 levelfor almost 84 percent of patients within 1 yearof neurotization and with no respiratory defi-cit.113 However, one should be cautious ofavoiding extensive ipsilateral intercostal nerveharvesting if the phrenic nerve has alreadybeen injured.

In global avulsions and in late presentations,secondary procedures, such as local or free-muscle transfers, are required to improve theoverall function. In general, better outcomescan be expected for young adults with shortdenervation times.

CONCLUSIONS

The treatment of brachial plexus lesions iscomplicated and requires a thorough under-standing of the brachial plexus macroanatomy,microanatomy, and function. Also, persistencefrom both the surgeon and the patient isneeded for achieving the best results. The clin-ical examination should be thorough and care-fully documented. In addition, radiographicevaluation of the plexus by means of a CT/myelogram and electrophysiologic studies in-cluding a Lamina test76 can assist in establish-ing a preoperative diagnosis. Realizing theexpected goals of the reconstruction and dis-cussing these goals extensively with the patientare of great importance. Subsequently, the re-storative microsurgeon can establish the strat-egies for the microreconstruction of the bra-chial plexus.

The preoperative diagnosis of the lesion canbe confirmed intraoperatively with a combina-tion of electrical stimulation of all roots anddistal plexus components and histochemistry.Working closely with the anesthesiologist andmaintaining the patient under light anesthesiaare also very important. Elevation of the vitalsigns while resecting a root will provide valu-able information and will determine whetherthe involved roots are avulsed or in continuitywith the spinal cord. Roots in continuity follow-ing resection to healthy fascicles can be usedfor motor donors. In general, intraplexus do-nors yield better results, mainly because of thehigher content of axon numbers and also be-cause they provide a more “physiologic” type ofreconstruction, which requires less retrainingfor the patient. Extraplexus donors, such as theaccessory nerve, the intercostal nerves, thephrenic nerve, or the cervical plexus nerves,are used mainly when multiple roots areavulsed. The higher the number of roots


avulsed, the more the need for harvesting ex-traplexus motor donors for neurotizations. Ex-traplexus donors are of extreme value and canyield good functional results for specific typesof reconstruction, i.e., intercostal nerves tomusculocutaneous neurotization or accessorynerve to suprascapular nerve neurotization.

The factors affecting the functional outcomehave also been studied. The denervation time,i.e., the period between the original traumaand the surgical reconstruction, is the mostimportant. The sooner the patient presents forbrachial plexus reconstruction, the better theoutcome.75 The age of the patient is a secondimportant factor. Younger patients are ex-pected to have better chances of functionalnerve regeneration and this is shown by mostoutcome studies; however, older patients withbrachial plexus lesions should be also exploredand their brachial plexus reconstructed. De-spite the complexity of the microsurgical pro-cedures and the prolonged recovery period,the surgical correction of brachial plexus inju-ries has become routine in recent years. Ad-vances in microsurgical techniques and postop-erative rehabilitation can eliminateamputation of the arm as a surgical option andcan facilitate the return to society of a highnumber of patients who until recently were notconsidered a productive part of the workforce.

Julia K. Terzis, M.D., Ph.D.Microsurgical Research CenterEastern Virginia Medical School700 Olney RoadRoom 2056Norfolk, Va. 23510REFERENCES

1. Flaubert, A. C. Memoire sur plusiers cas de luxationsdans les efforts pour la reduction ont ete suivisd’accidents graves. Repertoir Generale d’Anatomie et dePhysiologie Pathologique 3: 55, 1827.

2. Thorburn, W. Secondary suture of the brachial plexus.Br. Med. J. I: 1073, 1900.

3. Clark, L. P., Taylor, A. S., and Prout, T. P. Study onbrachial birth palsy. Am. J. Med. Sci. 130: 670, 1905.

4. Taylor, A. S. Brachial birth palsy and injuries of similartype in adults. Surg. Gynecol. Obstet. 30: 494, 1920.

5. Seddon, H. J. The use of autogenous grafts for therepair of large gaps in peripheral nerves. Br. J. Surg. 35:151, 1947.

6. Seddon, H. J. Nerve grafting. J. Bone Joint Surg. (Br.) 45:447, 1963.

7. Choi, P. D., Novak, C. B., Mackinnon, S. E., and Kline,D. G. Quality of life and functional outcome follow-ing brachial plexus injury. J. Hand Surg. (Am.) 22: 605,1997.

8. Midha, R. Epidemiology of brachial plexus injuries ina multitrauma population. Neurosurgery 40: 1182, 1997.

9. Penkert, G., Carvalho, G. A., Nikkhah, G., Tatagiba, M.,Matthies, C., and Samii, M. Diagnosis and surgery ofbrachial plexus injuries. J. Reconstr. Microsurg. 15: 3,1999.

10. Millesi, H. Interfascicular grafts for repair of periph-eral nerves of the upper extremity. Orthop. Clin. NorthAm. 8: 387, 1977.

11. Narakas, A. O. The treatment of brachial plexus inju-ries. Int. Orthop. 9: 29, 1985.

12. Kline, D. G., and Judice, D. J. Operative managementof selected brachial plexus lesions. J. Neurosurg. 58:631, 1983.

13. Andrew, T., and Wallace, W. A. Do brachial plexusinjuries occur at initial impact in motor-cyclists? Br.Med. J. 1: 1668, 1978.

14. Anonymous. Brachial plexus injuries in motorcyclists(Editorial). Br. Med. J. 280: 1242, 1980.

15. Barnes, R. Traction injuries in the brachial plexus inadults. J. Bone Joint Surg. (Br.) 31: 10, 1949.

16. Samardzic, M. M., Rasulic, L. G., and Grujicic, D. M.Gunshot wounds to the brachial plexus. J. Trauma 43:645, 1997.

17. Desai, D. C., Uribe, A., and Lachman, T. Brachialplexus injury due to compression: An alternate mech-anism of injury. Case report and review of the litera-ture. Am. Surg. 63: 487, 1997.

18. Nakamichi, K., and Tachibana, S. Intraneural gan-glion of the brachial plexus. J. Hand Surg. (Br.) 23: 1:123, 1998.

19. Arnold, P. M., Oldershaw, J. B., McDonald, L. W., andLanger, B. G. Myxomatous cyst of the brachial plex-us: Case report. J. Neurosurg. 73: 782, 1990.

20. Barbier, O., Malghem, J., Deleare, O., Vande Berg, B.,and Rombouts, J. J. Injury to the brachial plexus bya fragment of bone after fracture of the clavicle. J. BoneJoint Surg. (Br.) 79: 534, 1997.

21. Miller, D. S., and Boswick, J. A., Jr. Lesions of thebrachial plexus associated with fractures of the clavi-cle. Clin. Orthop. 64: 144, 1969.

22. Jupiter, J. B., and Leffert, R. D. Nonunion of the clav-icle: Associated complications and surgical manage-ment. J. Bone Joint Surg. (Am.) 69: 753, 1987.

23. Hansky, B., Murray, E., Minami, K., and Korfer, R. De-layed brachial plexus paralysis due to subclavian pseu-doaneurysm after clavicular fracture. Eur. J. Cardiotho-rac. Surg. 7: 497, 1993.

24. Van Alphen, J. C., Niekus, T. J., Koning, J., and Jansen,W. B. J. False aneurysm of the common carotid ar-tery and injury of the brachial plexus due to traumainflicted by an air gun pellet. J. Trauma 41: 1044, 1996.

25. Demetriades, D., Theodorou, D., Cornwell, E., III, et al.Penetrating injuries of the neck in patients in stablecondition: Physical examination, angiography, orcolor flow Doppler imaging. Arch. Surg. 130: 971, 1995.

26. Kroll, D. A., Caplan, R. A., Posner, K., Ward, R. J., andCheney, F. W. Nerve injury associated with anesthe-sia. Anesthesiology 73: 202, 1990.

27. Po, B. T., and Hansen, H. R. Iatrogenic brachial plexusinjury: A survey of the literature and of pertinent cases.Anesth. Analg. 48: 915, 1969.

28. Bhardwaj, D., and Peng, P. An uncommon mechanismof brachial plexus injury: A case report. Can. J. Anaesth.46: 173, 1999.

29. Van der Werkin, C., and De Vries, L. S. Brachial plexusinjury in multitraumatized patients. Clin. Neurol. Neu-rosurg. 95 (Suppl.): S30, 1993.


30. Coene, L. N. Mechanisms of brachial plexus lesions.Clin. Neurol. Neurosurg. 95 (Suppl.): S24, 1993.

31. Stevens, J. H. Brachial plexus injuries: By J. H. Stevens,M.D., 1934. Clin. Orthop. 237: 4, 1988.

32. Roaf, R. Lateral flexion injuries of the cervical spine.J. Bone Joint Surg. (Br.) 45: 36, 1963.

33. Kendall, H. O., Kendall, F. P., and Wadsworth, G. E.Muscles, Testing and Function, 2nd Ed. Baltimore: Wil-liams & Wilkins, 1971.

34. Landi, A., and Copeland, S. Value of the Tinel sign inbrachial plexus lesions. Ann. R. Coll. Surg. Engl. 61:470, 1979.

35. Dykes, R. N., and Terzis, J. K. Spinal nerve distribu-tions in the upper limb: The organization of the der-matome and afferent myotome. Philos. Trans. R. Soc.Lond. B. Biol. Sci. 293: 509, 1981.

36. Leffert, R. D. Clinical diagnosis, testing, and electro-myographic study in brachial plexus traction injuries.Clin. Orthop. 237: 24, 1988.

37. Cobby, M. J., Leslie, I. J., and Watt, I. Cervical myelo-graphy of nerve root avulsion injuries using water-soluble contrast media. Br. J. Radiol. 61: 673, 1988.

38. Cooke, J., Cooke, D., and Parsons, C. The anatomyand pathology of the brachial plexus as demonstratedby computed tomography. Clin. Radiol. 39: 595, 1988.

39. Nagano, A., Ochiai, N., Sugioka, H., et al. Usefulnessof myelography in brachial plexus injuries. J. HandSurg. (Br.) 14: 59, 1989.

40. Murphey, F., Hartung, W., and Kirklin, J. W. Myelo-graphic demonstration of avulsing injury of the bra-chial plexus. Am. J. Roentgenol. 58: 102, 1947.

41. Walker, A. T., Chaloupka, J. C., de Lotbiniere, A. C.,Wolfe, S. W., Goldman, R., and Kier, E. L. Detectionof nerve rootlet avulsion on CT myelography in pa-tients with birth palsy and brachial plexus injury aftertrauma. A.J.R. Am. J. Roentgenol. 167: 1283, 1996.

42. Sedel, L. The results of surgical repair of brachialplexus injuries. J. Bone Joint Surg. (Br.) 64: 54, 1982.

43. Millesi, H. Surgical management of brachial plexusinjuries. J. Hand Surg. (Am.) 2: 367, 1977.

44. Hashimoto, T., Mitomo, M., Hirabuki, N., et al. Nerveroot avulsion of birth palsy: Comparison of myelo-graphy with CT myelography and somatosensoryevoked potential. Radiology 178: 841, 1991.

45. Vielvoye, G. J., and Hoffmann C. F. E. Neuroradio-logical investigations in cervical root avulsion. Clin.Neurol. Neurosurg. 95 (Suppl.): S36, 1993.

46. Blair, D. N., Rapoport, S., Sostman, H. D., and Blair,O. C. Normal brachial plexus: MR imaging. Radiol-ogy 165: 763, 1987.

47. Nakamura, T., Yabe, Y., Horiuchi, Y., and Takayama, S.Magnetic resonance myelography in brachial plexusinjury. J. Bone Joint Surg. (Br.) 79: 764, 1997.

48. Volle, E., Assheuer, J., Hedde, J. P., and Gustorf-Aeck-erle, R. Radicular avulsion resulting from spinal in-jury: Assessment of diagnostic modalities. Neuroradiol-ogy 34: 235, 1992.

49. Carvalho, G. A., Nikkhah, G., Matthies, C., Penkert, G.,and Samii, M. Diagnosis of root avulsions in trau-matic brachial plexus injuries: Value of computerizedtomography myelography and magnetic resonanceimaging. J. Neurosurg. 86: 69, 1997.

50. Hems, T. E., Birch, R., and Carlstedt, T. The role ofmagnetic resonance imaging in the management oftraction injuries to the adult brachial plexus. J. HandSurg. (Br.) 24: 550, 1999.

51. Gasparotti, R., Ferraresi, S., Pinelli, L., et al. Threedimensional MR myelography of traumatic injuries ofthe brachial plexus. A.J.N.R. Am. J. Neuroradiol. 18:1733, 1997.

52. Roger, B., Travers, V., and Laval-Jeantet, M. Imagingof posttraumatic brachial plexus injury. Clin. Orthop.237: 57, 1988.

53. Bufalini, C., and Pesatori, G. Posterior cervical elec-tromyography in the diagnosis and prognosis of bra-chial plexus injuries. J. Bone Joint Surg. (Br.) 51: 627,1969.

54. Bonney, G., and Gilliatt, R. W. Sensory nerve conduc-tion after traction lesion of the brachial plexus. Proc.R. Soc. Med. 51: 365, 1958.

55. Murase, T., Kawai, H., Masatomi, T., Kawabata, H., andOno, K. Evoked spinal cord potentials for diagnosisduring brachial plexus surgery. J. Bone Joint Surg. (Br.)75: 775, 1993.

56. Kline, D. G., and Hudson, A. R. Diagnosis of rootavulsions. J. Neurosurg. 87: 483, 1997.

57. Holland, N. R., and Belzberg, A. J. Intraoperative elec-trodiagnostic testing during cross-chest C7 nerve roottransfer. Muscle Nerve 20: 903, 1997.

58. Landi, A., Copeland, S. A., Parry, C. B., and Jones, S. J.The role of somatosensory evoked potentials andnerve conduction studies in the surgical managementof brachial plexus injuries. J. Bone Joint Surg. (Br.) 62:492, 1980.

59. Turkof, E., Millesi, H., Turkof, R., et al. Intraoperativeelectroneurodiagnostics (transcranial electrical mo-tor evoked potentials) to evaluate the functional statusof anterior spinal roots and spinal nerves during bra-chial plexus surgery. Plast. Reconstr. Surg. 99: 1632,1997.

60. Celli, L., and Rowesta, C. Electrophysiologic Intraop-erative Evaluation of the Damaged Root in Tractionsof the Brachial Plexus. In J. K. Terzis (Ed.), Microre-construction of Nerve Injuries. Philadelphia: Saunders,1987.

61. Millesi, H. Brachial Plexus Injuries: Management andResults. In J. K. Terzis (Ed.), Microreconstruction of NerveInjuries. Philadelphia: Saunders, 1987.

62. Carson, K. A., and Terzis, J. K. Carbonic anhydrasehistochemistry: A potential method for peripheralnerve repair. Clin. Plast. Surg. 12: 227, 1985.

63. Riley, D. A., Ellis, S., and Bain, J. Carbonic anhydrasehistochemistry reveals subpopulation of myelinatedaxons in the dorsal and ventral roots of rat spinalnerves. Soc. Neurosci. Abstracts 7: 257, 1981.

64. Gruber, H., and Zenker, W. Acetylcholinesterase: His-tochemical differentiation between motor and sen-sory nerve fibres. Brain Res. 51: 207, 1973.

65. Engel, J., Ganel, A., Melamed, R., et al. Choline ace-tylcholinesterase for differentiation between humanmotor and sensory nerve fibers. Ann. Plast. Surg. 4:376, 1980.

66. Gu, Y.-D. Functional motor innervation of brachialplexus roots: An intraoperative electrophysiologicalstudy. J. Hand Surg. (Br.) 22: 258, 1997.

67. Capek, L., Clarke, H. M., and Zuker, R. M. Endoscopicsural nerve harvest in the pediatric patient. Plast. Re-constr. Surg. 98: 884, 1996.

68. Malessy, M. J., and Thomeer, R. Evaluation of inter-costal to musculocutaneous nerve transfer in recon-structive brachial plexus surgery. J. Neurosurg. 88: 266,1998.


69. Narakas, A. O., and Hentz, V. R. Neurotization in bra-chial plexus injuries: Indications and results. Clin. Or-thop. 237: 43, 1988.

70. Gilbert, A. Neurotisation by Contralateral PectoralNerve. In Proceedings of the 10th Symposium on the Bra-chial Plexus, Lausanne, Switzerland, 1992. Taken fromChuang, D. Neurotization procedures for brachialplexus injuries. Hand Clin. 11: 633, 1995.

71. Terzis, J. K. Selective contralateral C7 technique fordevastating brachial plexus lesions (in preparation).

72. Gu, Y. D., and Ma, M. K. Nerve transfer for treatmentof root avulsion of the brachial plexus: Experimentalstudies in a rat model. J. Reconstr. Microsurg. 7: 15, 1991.

73. Chen, L., and Gu, Y. D. An experimental study ofcontralateral C7 root transfer with vascularized nervegrafting to treat brachial plexus avulsion. J. Hand Surg.(Br.) 19: 60, 1994.

74. Terzis, J. K., and Wu, Y. Microanatomy of neurotiza-tion procedures for restorative microsurgery of bra-chial plexus injury (in preparation).

75. Terzis, J. K., Vekris, M. D., and Soucacos, P. Outcomesof brachial plexus reconstruction in 204 patients withdevastating paralysis. Plast. Reconstr. Surg. 104: 1221,1999.

76. Liberson, W. T., and Terzis, J. K. Contribution of Clin-ical Neurophysiology and Rehabilitation Medicine inthe Management of Brachial Plexus Palsy. In J. K.Terzis (Ed.), Microreconstruction of Nerve Injuries. Phil-adelphia: Saunders, 1987. Pp. 555–570.

77. Malessy, M. J., Thomeer, R. T., and van Dijk, J. G.Changing central nervous system control followingintercostal nerve transfer. J. Neurosurg. 89: 568, 1998.

78. Emmelot, C. H., Nielsen, H. K., and Eisma, W. H.Shoulder fusion for paralyzed upper limb. Clin. Or-thop. 340: 95, 1997.

79. Doi, K., Shigetomi, M., Kaneko, K., et al. Significanceof elbow extension in reconstruction of prehensionwith reinnervated free-muscle transfer following com-plete brachial plexus avulsion. Plast. Reconstr. Surg.100: 364, 1997.

80. Berger, A., Flory, P. J., and Schaller, E. Muscle transferin brachial plexus lesions. J. Reconstr. Microsurg. 6: 113,1990.

81. Akasaka, Y., Hara, T., and Takahashi, M. Free muscletransplantation combined with intercostal nerve cross-ing for reconstruction of elbow flexion and wrist ex-tension in brachial plexus injury. Microsurgery 12: 346,1991.

82. Ihara, K., Doi, K., Sakai, K., et al. Restoration of sen-sibility in the hand after complete brachial plexusinjury. J. Hand Surg. (Am.) 21: 381, 1996.

83. Chuang, D. C., Carver, N., and Wei, F. C. Results offunctioning free muscle transplantation for elbowflexion. J. Hand Surg. (Am.) 21: 1071, 1996.

84. Doi, K., Sakai, K., Fuchigami, Y., and Kawai, S. Recon-struction of irreparable brachial plexus injuries withreinnervated free-muscle transfer. J. Neurosurg. 85:174, 1996.

85. Manktelow, R. T., and McKee, N. H. Free muscle trans-plantation to provide active finger flexion. J. HandSurg. (Am.) 3: 416, 1978.

86. Manktelow, R. T. Functioning Free Muscle Transfers.In D. P. Green (Ed.), Operative Hand Surgery, 3rd Ed.New York: Churchill Livingstone, 1993. Pp. 151–164.

87. Manktelow, R. T., and Zuker, R. M. The principles of

functioning muscle transplantation: Applications tothe upper arm. Ann. Plast. Surg. 22: 275, 1989.

88. Doi, K., Hattori, Y., Kuwata, N., et al. Free muscletransfer can restore hand function after injuries of thelower brachial plexus. J. Bone Joint Surg. (Br.) 80: 117,1998.

89. Manktelow, R. Functioning Muscle Transplantation tothe Arm. In J. K. Terzis (Ed.), Microreconstruction ofNerve Injuries. Philadelphia: Saunders, 1987. Pp. 267–281.

90. Ruhmann, O., Wirth, C. J., and Gosse, F. Trapeziustransfer in deltoid paralysis (in German). Orthopade26: 634, 1997.

91. Narakas, A. O. Muscle transpositions in the shoulderand upper arm for sequelae of brachial plexus palsy.Clin. Neurol. Neurosurg. 95 (Suppl.): S89, 1993.

92. Tsai, T. M., Kalisman, M., Burns J., and Kleinert, H. E.Restoration of elbow flexion by pectoralis major andpectoralis minor transfer. J. Hand Surg. (Am.) 8: 186,1983.

93. Beaton, D. E., Dumont, A., Mackay, M. B., and Richards,R. R. Steindler and pectoralis major flexorplasty: Acomparative analysis. J. Hand Surg. (Am.) 20: 747,1995.

94. Stern, P. J., and Caudle, R. J. Tendon transfers forelbow flexion. Hand Clin. 4: 297, 1988.

95. Zancolli, E., and Mitre, H. Latissimus dorsi transfer torestore elbow flexion: An appraisal of eight cases.J. Bone Joint Surg. (Am.) 55: 1265, 1973.

96. Ruhmann, O., Gosse, F., Wirth, C. J., and Schmolke, S.Reconstructive operations for the paralyzed shoulderin brachial plexus palsy: Concept of treatment. Injury30: 609, 1999.

97. Bruxelle, J., Travers, V., and Thibeaut, J. B. Occur-rence and treatment of pain after brachial plexusinjury. Clin. Orthop. 237: 87, 1988.

98. Parry, C. B. Pain in avulsion lesions of the brachialplexus. Pain 9: 41, 1980.

99. Narakas, A. O. The Effects of Pain in ReconstructiveMicrosurgery in 160 Patients with Traction and orCrush Injury to the Brachial Plexus. In J. Siegfried, M.Zimmermann, and R. Baumgartner (Eds.), Phantomand Stump Pain. Berlin: Springer-Verlag, 1981. Pp.126–147.

100. Berman, J. S., Birch, R., and Anand, P. Pain followinghuman brachial plexus injury with spinal cord rootavulsion and the effect of surgery. Pain 75: 199, 1988.

101. Berman, J., Anand, P., Chen, L., et al. Pain relief frompreganglionic injury to the brachial plexus by lateintercostal nerve transfer. J. Bone Joint Surg. (Br.) 78:759, 1996.

102. Friedman, A. H., Nashold, B. S., Jr., and Bronec, P. R.Dorsal root entry zone lesions for the treatment ofbrachial plexus avulsion injuries: A follow-up study.Neurosurgery 22: 369, 1988.

103. Samii, M., and Moringlane, J. R. Thermocoagulationof the dorsal root entry zone for the treatment ofintractable pain. Neurosurgery 15: 953, 1984.

104. Wilkinson, M. C., Birch, R., and Bonney, G. Brachialplexus injury: When to amputate? Injury 24: 603, 1993.

105. Sedel, L. The Management of Supraclavicular Le-sions: Clinical Examination, Surgical Procedures, Re-sults. In J. K. Terzis (Ed.), Microreconstruction of NerveInjuries. Philadelphia: Saunders, 1987. Pp. 385–392.

106. Songcharoen, S., Mahaisavariya, B., Chotigavanich, C.,et al. Spinal accessory neurotization for restoration


of elbow flexion in avulsion injuries of the brachialplexus. J. Hand Surg. (Am.) 21: 387, 1996.

107. Allieu, Y., Privat, J. M., and Bonnel, F. Paralysis in RootAvulsion of the Brachial Plexus: Neurotization by theSpinal Accessory Nerve. In J. K. Terzis (Ed.), Microre-construction of Nerve Injuries. Philadelphia: Saunders,1987. Pp. 415–423.

108. Nagano, A., Tsuyama, N., Ochiai, N., et al. Directnerve crossing with the intercostal nerve to treat avul-sion injuries of the brachial plexus. J. Hand Surg. (Am.)14: 980, 1989.

109. Waikakul, S., Wongtragul, S., and Vanadurongwan, V.Restoration of elbow flexion in brachial plexus avul-sion injury: Comparing spinal accessory nerve transferwith intercostal nerve transfer. J. Hand Surg. (Am.) 24:571, 1999.

110. Gu, Y. D., Chen, D. S., and Zhang, G. M. Long-termfunctional results of contralateral C7 transfer. J. Re-constr. Microsurg. 14: 57, 1998.

111. Terzis, J. K., and Breidenbach, W. The Anatomy of theFree Vascularized Nerve Grafts. In J. K. Terzis (Ed.),Microreconstruction of Nerve Injuries. Philadelphia: Saun-ders, 1987. Pp. 101–116.

112. Liu, J., Pho, R. W., Kour, A. K., et al. Neurologic deficitand recovery in the donor limb following cross-C7transfer in brachial-plexus injury. J. Reconstr. Microsurg.13: 237, 1997.

113. Gu, Y. D., and Ma, M. K. Use of phrenic nerve forbrachial plexus reconstruction. Clin. Orthop. 323: 119,1996.

Self-Assessment Examination follows onpage 1123.


Self-Assessment Examination

The Surgical Treatment of Brachial Plexus Injuries in Adultsby Julia K. Terzis, M.D., Ph.D., and Konstantinos C. Papakonstantinou, M.D., Ph.D.

1. TINGLING UPON TAPPING OF THE SUPRACLAVICULAR AREA USUALLY MEANS:A) Healing fracture of the clavicle.B) Pseudoaneurysm of the subclavian artery.C) Fluid collection.D) Root connectivity with the spinal cord.

2. A POSITIVE HORNER’S SIGN MEANS:A) Injury to the upper roots of the brachial plexus.B) Normal hand function.C) The patient has a concomitant ophthalmologic problem.D) Lower-root avulsions.

3. PSEUDOMENINGOCELES IN A CT/MYELOGRAM MOST LIKELY MEAN:A) The roots most likely are connected to the spinal cord.B) The roots most likely are totally or partially avulsed.C) The interventional radiologist injured the root.D) Intact dural sheath.

4. END NEUROMAS OF THE ROOTS WHEN FOUND SHOULD BE:A) Stimulated.B) Excised and grafted.C) Sent for biopsy.D) All of the above.

5. EXTRAPLEXUS MOTOR DONORS INCLUDE:A) The contralateral intercostal nerves.B) The saphenous nerves.C) The ipsilateral cervical motors and accessory nerve.D) None of the above.

6. GOALS OF BRACHIAL PLEXUS RECONSTRUCTION IN GLOBAL PARALYSIS INCLUDE:A) Shoulder stabilization.B) Protective sensation in the hand.C) Elbow flexion.D) All of the above.

7. REINNERVATION OF THE IPSILATERAL LATISSIMUS DORSI MUSCLE IS IMPORTANT FOR:A) Cosmetic reasons.B) Stabilization of the scapula.C) Transfer to the arm as a pedicled muscle to strengthen elbow animation.D) Restoration of finger flexion as a free-muscle transfer.

8. FREE MUSCLES FOR SUCCESSFUL RESTORATION OF ELBOW FLEXION INCLUDE:A) The contralateral latissimus dorsi and the rectus femoris muscle.B) The adductor longus muscle.C) The gracilis muscle.D) None of the above.

9. PAIN IN BRACHIAL PLEXUS INJURIES:A) Is psychological in origin and should be treated with counseling.B) Can be severe in high-velocity penetrating injuries.C) Always requires surgical treatment.D) Is associated with upper-root avulsion.

10. THE BEST PROGNOSIS AFTER BRACHIAL PLEXUS INJURY CAN BE EXPECTED:A) In older patients.B) In multiple-root avulsions.C) In long denervation times.D) None of the above.

To complete the examination for CME credit, turn to page 1241 for instructions and the response form.

brachial plexus injury in adults

Documents

flexor digitorum profundus

pedicled vascularized graft

magnetic resonance imaging

senior author prefers

brachial plexus elements

interposition nerve grafts

interposition nerve grafting

brachial birth palsy