Peripheral Nerve Monitoring During Operative Procedures · tive monitoring (IOM) requires a skilled...

P

No one in

The idea

Periph

B

nvolved in the

America

s and opinion

heral NOpe

Brian A. Cr

e planning of Aut

C

an Associatio

2621 S

ns in this Mon

Nerveerativerum, MD a

this CME actithors/faculty

CME is availab

Copy

on of Neurom

Superior Dr N

nograph are sothose o

e Mone Procnd Jeffrey

ivity have anyhave nothing

ble 2/1/2010

yright: 200

muscular and

NW Rochest

olely those ofof the AANE

nitorincedureA. Stromm

y relevant finag to disclose.

- 2/1/2013

07

d Electrodiag

ter, MN 559

f the author anEM

ng Dues men, MD

ancial relatio

gnostic Medi

01

nd do not nec

Product

uring

nships to disc

icine

cessarily repre

: MM49

close.

esent

CMPr CoMalsuthinInTheldi LeU

ReExexD AcThCoboon HoOpuvi

ME Informatioroduct: MM49

ourse DescriptMonitoring of plowing for rapurgeon in the his neurophysinvolved in intrantended Audiehis course is ectrodiagnostisorders.

earning Objectpon conclusion

1. describ2. describ3. assess t

elease Date: 2xpiration Datexpiration date.uration/Comp

ccreditation anhe American Aouncil for Conoth sections ofnly the credit c

ow to Obtain Cnce you have urchased this pew a transcr

on 9 ‐ Peripheral N

tion peripheral nervpid assessmenoverall approaological informaoperative pernce intended for ic medicine w

tives n of this prograbe the uses of pbe techniques otechnical prob

/1/2010 e: 2/1/2013. Yo pletion Time: 2

nd DesignationAssociation of ntinuing Medicf this enduringcommensurate

CME reviewed the product. Answript of your

Nerve Monitor

ve function is nt of the integach to treatmemation in an aipheral nerve m

Neurologists,with the inten

am, participanperipheral nervof peripheral nblems associate

our request to

2 hours

n Statements Neuromuscul

cal Education tg material for ae with the exte

materials, youwer the questioCME by log

ring During Op

important durgrity of the roent of these leaccurate and monitoring and

, Physiatrists, t to improve

ts should be abve stimulation nerve stimulatied with the ope

receive AMA P

ar and Electroto provide cona combined mant of their part

u can obtain Cons and click sugging into ww

perative Proced

ring the surgicoots, plexus, aesions. There arapid fashion. d describes its

and others wthe quality o

ble to: and intraoperon during surgerative monito

PRA Category 1

odiagnostic Mntinuing medicaximum of 2 Aticipation in th

CME credit by ubmit. Once yww.aanem.org

dures

cal treatment nd nerves. Thare, however, This study asuse in the mo

who practice of medical car

rative monitorigical interoperaoring of periph

1 Credits™ mu

edicine (AANEcal education MA PRA Categhe activity.

clicking on thyour answers hg and clicking

of peripheral he results of tmany technicassesses the eqre common cli

neuromusculre to patients

ing during opeative monitorineral nerves.

ust be submitte

EM) is accredifor physiciansgory 1 creditsTM

e link in the ehave been submg View Profi

nerve and plehis monitoringal challenges tquipment and nical scenarios

ar, musculosks with muscle

rative procedung.

ed on or befor

ited by the Acs. The AANEMM. Physicians s

e‐mail receivedmitted, you wile and then

xus lesions, g assist the to providing techniques s.

keletal, and and nerve

ures.

re the credit

ccreditation designates hould claim

d when you ll be able to My CME.

AANEM MONOGRAPH ABSTRACT: Monitoring of peripheral nerve function is important duringthe surgical treatment of peripheral nerve and plexus lesions, allowing forrapid assessment of the integrity of the roots, plexus, and nerves. Theresults of this monitoring assist the surgeon in the overall approach totreatment of these lesions. There are, however, many technical challengesto providing this neurophysiological information in an accurate and rapidfashion. This study assesses the equipment and techniques involved inintraoperative peripheral nerve monitoring and describes its use in the morecommon clinical scenarios.

Muscle Nerve 35: 159–170, 2007

PERIPHERAL NERVE STIMULATION AND MONITORINGDURING OPERATIVE PROCEDURES

BRIAN A. CRUM, MD,1 and JEFFREY A. STROMMEN, MD2

1 Department of Neurology, Mayo Clinic College of Medicine, 200 First Street, SW, Rochester, Minnesota 55905, USA2 Department of Physical Medicine and Rehabilitation, Mayo Clinic College of Medicine, Rochester, Minnesota, USA

Accepted 19 October 2006

Electrophysiological monitoring of peripheralnerves during surgery is an extremely valuable pro-cedure that provides vital, real-time information tothe surgical team. Preoperative electrophysiologicaltesting provides the surgeon with valuable data toassist with decision-making; however, there is infor-mation that simply cannot be obtained from thesestudies, and intraoperative studies help bridge thisgap. Monitoring of peripheral nerves during surgeryhas been described and studied for nearly 40years21,22 and has been the subject of several re-views14,30,31,33 as well as a monograph.2 Intraopera-tive monitoring (IOM) requires a skilled and knowl-edgeable team of physicians and technologists whoensure rapid and accurate acquisition and interpre-tation of electrophysiological data. It also requires agood working relationship with the entire surgicalteam. Communication with the surgeons is vital inassisting decision-making between divergent surgicalplans. Accurate and timely communication with the

anesthesiologist is also important as certain anes-thetic agents can have detrimental effects on elec-trophysiological studies. Having a peripheral nerveIOM plan prior to surgery is ideal to maximize theefficiency and utility of the studies.

The uses for peripheral nerve IOM monitoringinclude: (1) identifying peripheral nerves; (2) localiz-ing pre-existing disease processes along the course of anerve; (3) determining the functional continuity acrossa pre-existing lesion, prior to detection by other elec-trophysiological means; (4) determining likelihood ofnerve root avulsion; (5) identifying targets for nervebiopsy; and (6) monitoring and thereby preventingdamage to intact nerves during surgery.

This review provides background on the equip-ment and techniques used for intraoperative periph-eral nerve and plexus monitoring and an interpre-tation of the findings obtained thereby. The use ofnerve conduction studies (NCSs), needle electro-myography (EMG), and evoked potentials (EPs) isconsidered. For the purpose of this review, the term“peripheral nerve” applies to nerve distal to the spi-nal cord, and also distal to the intervertebral fora-men (spinal root/nerve, plexus, peripheral nerve).The application of these electrodiagnostic (EDX)techniques is described for specific clinical scenar-ios. The use of IOM involving cranial nerves hasrecently been reviewed and is not addressed.12

TECHNIQUES

Nerve Conduction Studies. NCSs are performed withstimulation and recording of motor, sensory, or

Abbreviations: ADM, abductor digit minimi; CMAP, compound muscle ac-tion potential; EDX, electrodiagnostic; EEG, electroencephalogram; EMG,electromyography; EP, evoked potential; IOM, intraoperative monitoring;NCS, nerve conduction study; NAP, nerve action potential; MEP, motorevoked potential; MRI, magnetic resonance imaging; MUAP, motor unit ac-tion potential; SEP, somatosensory evoked potential; SNAP, sensory nerveaction potential; UNE, ulnar neuropathy at the elbowKey words: brachial plexopathy; intraoperative monitoring; intraoperativenerve stimulation; motor evoked potentials; nerve root avulsion; somatosen-sory evoked potentialsCorrespondence to: S. C. Stucky; e-mail: [email protected]

© 2006 American Association of Neuromuscular & Electrodiagnostic Medi-cine. Published by Wiley Periodicals, Inc.Published online 8 December 2006 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/mus.20707

Peripheral Nerve Stimulation MUSCLE & NERVE February 2007 159

mixed nerves, yielding a nerve action potential(NAP). Potentials can also be recorded over muscle[compound muscle action potentials (CMAPs)] orcentrally over spine or scalp (EPs). Many factorsdictate which of these (or what combination) areused, and this, in turn, dictates the type of equip-ment necessary and the regimen of anesthesia thatshould be utilized.

Intraoperative peripheral nerve stimulation isusually performed directly on the surgically exposednerve. A bipolar stimulator is held in place by thesurgeon with the cathode directed toward the re-cording electrodes. The interelectrode distance isusually 3 mm, although the distance is dictated bythe size of the nerve. A larger nerve such as thesciatic nerve demands a larger interelectrode dis-tance, such as 7 mm. The proper orientation of thestimulating electrodes is important, and visual con-firmation of correct placement by the IOM physicianor technologists in the operating room is useful. Ifthe bipolar orientation is not possible, monopolarstimulation can be performed with the cathode onthe nerve and the anode some distance away. Thisshould generally be avoided, however, as stimulationcannot be focused precisely, and the current mayspread to adjacent nerves as well as farther down thenerve being studied. Bipolar stimulation results inless current spread and is therefore preferred. An-other possible technique is using a tripolar stimulat-ing electrode.22,34 In tripolar stimulation, a singlecathode is situated between two anodes. This furtherfocuses the site of stimulation while minimizing thestimulus artifact. The size of the stimulating elec-trodes should be in concert with the size of thenerve. Special stimulating electrodes, some withhooks or very small pointed tips, often need to beused with this technique. During stimulation, thenerve should be elevated out of the surgical field toavoid contact with excessive fluid, which effectivelyreduces the stimulus received by the underlyingnerve. Our stimulating electrodes are made of asilver solder alloy and are gas-sterilized or autoclavedafter surgical procedures.

The amount of stimulation required to depolar-ize the underlying axon is less than that needed forpercutaneous stimulation. Excessive stimulationmust be avoided in order to limit current spreaddown the stimulated nerve and to other nearbynerves and muscles.

Excessive stimulation also increases the stimulusartifact, which becomes a larger problem as the dis-tance between stimulating and recording electrodesis lessened. A square-wave pulsed stimulus with aduration of 0.05 ms and intensity of only a few

milliamperes (1–5) is usually sufficient for depolar-ization of the axonal membrane in a supramaximalfashion. Of note, short-duration stimulation of thistype is more likely to preferentially activate motoraxons.36 For constant-voltage stimulation, intensitiesof 25–50 V are used.25 It is important to rememberthat, in diseased nerves, higher stimulus intensitymay be needed to reach threshold. It is our practiceto increase the stimulus intensity up to 20–25 mA ifno response is obtained initially. These higher inten-sities, however, result in increased stimulus artifactand likely evoke volume-conducted responses ratherthan true NAPs (see below). Since the size of theNAP is small (microvolts), averaging one to five re-sponses is often helpful (although when a responseis present, it is usually visible after the first stimulus).When there is no visible twitch from the muscle andno elicited electrical response, it is imperative toensure that stimulation is indeed occurring. Thestimulating electrodes then are placed on a nearbyknown functioning nerve or muscle to assess fortwitch. Even in the setting of neuromuscular block-ing agents, direct stimulation of muscle should resultin a twitch.

NAP recording takes place along the course ofthe stimulated nerve. The distance between the cath-ode and the active recording electrode should be atleast 4 cm to minimize stimulus artifact. The inter-electrode distance is 3–5 mm, with a wider separa-tion used when there is a longer distance betweenthe cathode and recording electrode. This allows forthe longer traveling wave to pass completely underthe active electrode before passing under the refer-ential electrode. The farther the separation, how-ever, the more likely it is that extraneous noise andstimulus artifact will differ at the two recording elec-trodes and will not be rejected as common modesignals by the differential amplifier. The nerveshould be lifted out of the surgical field as at the siteof stimulation. The recording electrodes are held inplace by the surgeon and visual confirmation of thecorrect orientation (active electrode toward thecathode) should be obtained. The size of the record-ing electrodes should match the recording nerve. Athree-pronged electrode with a ground–active–refer-ential arrangement has also been used. Electrodescan be used for either stimulating or recording,depending on whether they are connected to theamplifier or stimulator. Bipolar point electrodes aregenerally used for stimulation and the curved hookelectrodes for recording. The ground electrode is aflat metal plate that is placed under the patient andis separate from the ground used for cautery.

160 Peripheral Nerve Stimulation MUSCLE & NERVE February 2007

Since the NAP waveforms contain frequencies inthe 1-kHz range, it is appropriate to use filter set-tings of 5–10 Hz for the low-frequency (high-pass)filter and 2–3 kHz for the high-frequency (low-pass)filter. Occasionally, it is helpful to increase the high-frequency filter to better separate the NAP waveformfrom stimulus artifact.34 The NAP amplitude is usu-ally less than 100 �V, so a gain of 20–50 �V perdivision is utilized. The latency will depend on thelength between stimulating and recording elec-trodes. A simple guide is 1 ms per 5-cm distance(assuming 50-m/s conduction velocity). A time baseof 0.5 ms per division is reasonable, but should beincreased for longer distances.

When the goal of intraoperative peripheral nervestudies is to determine the functional continuity orthe exact location of a peripheral nerve lesion, stim-ulation and recording will need to be performed oneither side of the lesion. The orientation may varydepending on the accessibility to the nerve. Whenassessing for continuity, it is important to realize thatthe presence of only a few large myelinated axonscan produce a response with a relatively normalconduction velocity, latency, and threshold; there-fore, it is most useful to assess the amplitude of theNAP to determine the number of functioning axonsacross a lesion. A NAP proves the existence of a largenumber (over 4000) of functioning, medium-sized,myelinated axons.21 Stimulation is usually per-formed proximal to the lesion, with recording distal.When localizing lesions, recording electrodes aregenerally placed proximally and the stimulating elec-trodes are moved from proximal to within, and thendistal to, the suspected lesion, assessing for a changein morphology of the waveform when a step is madeinto or across the lesion.

Recording CMAPs has the advantage of amplifi-cation of responses as each axon innervates andactivates hundreds to thousands of muscle fibers.Amplitudes are measured in millivolts, as opposed tomicrovolts in NAP and EP recordings. CMAP record-ing is performed with surface (as in routine NCSs),subcutaneous, or intramuscular electrodes. Subcuta-neous recordings are performed with electroen-cephalographic (EEG) needles placed above or intothe muscle of interest. Fine, longer, intramuscularwires placed with the use of a hollow needle can alsobe utilized. Both intramuscular and subcutaneousrecordings limit the size of the recording area,thereby reducing extraneous noise in the recording.Intramuscular recordings record from a small part ofa muscle and therefore reflect activity in a fraction ofthe axons that might be stimulated intraoperatively.They also tend to introduce more extraneous noise

than subcutaneous EEG electrodes and are not suit-able for IOM if the amount of innervation to amuscle is important to measure. This technique ismost often utilized for deep muscles or muscles thatare smaller or where selective recording is difficult(e.g., rhomboids, laryngeal muscles). SubcutaneousEEG electrodes are favored at the authors’ institu-tion given the ease of placement and quiet record-ings and the ability to better quantify the number offunctioning axons.

Evoked Potentials. Responses from stimulation ofperipheral nerve can also be recorded from the spi-nal cord and cortex as somatosensory evoked poten-tials (SEPs). Stimulation of peripheral nerve will de-polarize both motor and sensory axons; however,selective orthodromic recording from central sen-sory pathways ensures that only the large-fiber/dor-sal-column pathway is assessed, just as it does withpercutaneous stimulation in routine SEPs. Since thegoal for intraoperative SEPs (in most cases) is toassess nerve root continuity, stimulation is per-formed as close to the intervertebral foramen aspossible. The cathode is directed proximally andrecording is performed from the cervical spine levelvia either a nasopharyngeal electrode or from a nee-dle electrode placed directly on the lamina in thecervical spine. Scalp EEG electrodes are placed atC3� and C4� with Fz as a reference (international10–20 system). These responses are small in ampli-tude, so many stimuli must be averaged—typically20–50 stimulating at 1.1–1.9 Hz. Stimulus intensity istypically between 10 and 20 mA. The presence of acentral response (scalp or cervical spine) indicatesthe continuity of the dorsal root in cases whereavulsion is questioned.15,26 Although this does notdirectly test the ventral root, its separate continuity isoften assumed when a response is obtained.26 Lackof a response argues for dorsal nerve root avulsion ordisruption, especially when NAPs can be recordedfrom the corresponding spinal nerve or plexus ele-ment. In a pure preganglionic lesion affecting thedorsal roots, the cell body and peripheral sensoryaxons are still intact and a peripheral NAP would beexpected, usually with normal conduction velocity.

Motor evoked potentials (MEPs) from transcra-nial electrical stimulation can be recorded peripher-ally at the intervertebral foramen, at the same loca-tion as stimulation for SEPs. Anodal stimulation isutilized with a short-duration (0.05 ms), rapid-rise-time stimulus using subcutaneously placed EEG elec-trodes at C3 and C4. Single or several (2–5) stimuliwith an interstimulus interval of 1 ms are given withan intensity of 200–600 V. Direct nerve recording is


performed with the bipolar electrodes describedabove, usually placed onto the spinal nerve as closeto the exit from the intervertebral foramen as possi-ble. A response indicates continuity of the ventralroot, whereas absence of a response suggests rootavulsion or nonfunctioning axons.35 Neuromuscularparalysis is often employed to eliminate volume-conducted muscle artifact from neck and proximalarm muscles. Excessive stimulus artifact can also be aproblem given the short distance between stimulat-ing and recording electrodes. Occasionally, it is nec-essary to move the recording electrodes distally ontobrachial plexus elements to obviate this artifact. Thepolarity of the stimulus can also be reversed andseveral stimuli averaged in an attempt to reduce thestimulus artifact by phase cancellation. The MEPlatency is typically in the neighborhood of 8 ms whenrecording from the cervical roots.

MEPs can also be recorded from muscle using asurface, subcutaneous, or intramuscular electrode.The size of a response in this situation may havemore to do with the distance between any depolar-izing muscle fibers and the recording electrodesthan with the actual number of functioning axons.Although a response does indicate continuity or re-innervation to the particular muscle, it cannot provethe continuity at individual roots: even with lack ofcontinuity of one root (e.g., C5), a MEP could still berecorded over a muscle (e.g., biceps) due to inner-vation by another root (C6). Also, a MEP recordedfrom muscle may be due to only minimal reinnerva-tion of a few axons that may not result in meaningfulrecovery. An absent MEP recorded over muscle doesnot disprove continuity or reinnervation across alesion as there may be regenerating axons that havenot yet reached the muscle. An absent muscle re-sponse cannot distinguish between a pre- and a post-ganglionic lesion. The most accurate ways of predict-ing nerve (especially root) continuity and axonalregeneration are the procedures of recording MEPsover the nerve roots and evaluating for proximalNAPs.

In both SEP and MEP recordings, it is desirableto have a “normal” control to confirm the correctfunctioning of all equipment and the reliability ofthe study. This ideally would entail study of a spinalnerve known to be functional by clinical examina-tion, radiological studies, or preoperative electro-physiological studies, with the expectation of SEPand MEP responses. An absent response would indi-cate a technical problem that would need to beremedied before further testing is undertaken.Along the same lines, when there is concern regard-ing a false-positive response (e.g., a volume-conducted

muscle response in the MEP study), recording of aspinal nerve from a known avulsed ventral root ishelpful; a response there confirms a likely false-pos-itive response from the other root. A neuromuscularblocking agent could also be used to make this dis-tinction. If the waveform in question disappears ordecreases in size, a volume-conducted response ismost likely. Given the limited surgical exposure inmost cases, however, this control neurophysiologicalassessment can be difficult or impossible to accom-plish. As an alternative, the contralateral limb can betested by stimulation of the median nerve at thewrist, checking for SEPs over the cervical spine andscalp. A MEP can be recorded over a limb muscle[biceps or abductor digit minimi (ADM)]. The re-cording electrodes over the ADM can also be used torecord a CMAP with peripheral (wrist) stimulationin the monitoring of neuromuscular blockade.

In choosing which technique (SEP or MEP) shouldbe used for monitoring purposes, it is important torealize that, from a surgical reconstruction perspective,the continuity of the ventral roots is the most function-ally important factor to determine. Functioning ventralroots can be used as the proximal stump for nervegrafting. The loss of the ventral roots as a graftingvehicle will necessitate other transposition proceduresutilizing nerve or nerve–muscle transfers from otherterritories.29,31 The continuity of the dorsal root doesnot guarantee continuity of the ventral root, and viceversa.26 A mismatch (partially avulsed dorsal or ventralroots) was noted in 11% of roots studied by laminec-tomy.8 In most of these instances, the ventral root wasavulsed with an intact dorsal root. A combination ofthese two IOM techniques (SEP and MEP), therefore,may be ideal.3,14

Needle Electromyography. Monitoring needle EMGactivity during surgery can give relatively noninva-sive, real-time information regarding the status ofmotor axons. A recording electrode placed in a mus-cle can be used to identify abnormal activity, namelyneurotonic discharges. Neurotonic discharges arehigh-frequency bursts of motor unit action poten-tials (MUAPs) either firing briefly or in more pro-longed trains.9 These bursts or trains are made up ofMUAPs firing at 30–100 Hz. Neurotonic dischargesare caused by mechanical irritation to axons, includ-ing traction, stretch, manipulation, or saline irriga-tion. They must be distinguished from irregular vol-untary MUAPs occurring under light anesthesia orfrom other electrode or surgical (i.e., cautery) arti-facts. Although neurotonic discharges are sensitiveindicators of nerve irritation, their presence doesnot always indicate damage to axons and their ab-


sence does not guarantee lack of damage. Neuro-tonic discharges are, for example, common in somespine surgeries, although postoperative radiculopa-thy is rare.11 It is important to note that sharp tran-section of a nerve may produce no neurotonic dis-charges. Neurotonic discharges are less likely to beproduced after mechanical stimulation in previouslydamaged nerves. Neurotonic discharges can still berecorded with neuromuscular blocking agents pro-ducing up to 75% block, as measured by CMAPamplitude.16

Free-running recording can be achieved withsubcutaneous EEG needle electrodes, often refer-encing a nearby muscle to limit the number of chan-nels required. One channel would have medial gas-trocnemius referenced to anterior tibialis, the nextvastus medialis referenced to rectus femoris, and soon. Abnormal activity may not, therefore, be local-ized precisely to one muscle. If such localization isvital, then each channel should be made to repre-sent separate muscles, with two electrodes placed ineach muscle. Fine-wire electrodes can also be used,with the active recording surface being a small baredtip. This is most useful for deep or small muscles(rhomboids, laryngeal muscles), especially when amore selective recording from a single muscle isvital. As the fine-wire electrodes record from asmaller area of muscle, EEG electrodes are preferredto maximize the chance of detecting neurotonicdischarges.

TECHNICAL PROBLEMS

Acquiring accurate information quickly is importantbecause it guides surgical decision-making. Severalpotential technical problems must be understood.Low nerve temperature is inevitable in IOM, leadingto slowed conduction velocities and higher responseamplitudes. Since temperature cannot be increasedduring IOM, the effect of low temperature mustsimply be kept in mind. Most analysis during IOM ismonitoring for the presence or absence of a poten-tial or a change in a potential at a nearby recordingsite. Cool temperatures during IOM will unlikelyhave a significant effect on these parameters. Lowsystemic blood pressure can reduce SEP and MEPamplitudes. Peripheral ischemia from a blood pres-sure cuff can also affect studies. If a tourniquet is inplace for more than 60 minutes, it should be re-leased for at least 20 minutes before beginning IOMstudies. Since the operating room is an electricallyhostile environment, it is imperative to rememberthat surgical instruments, beds, machines, and lightsall contribute 60-Hz interference. Limiting fluores-

cent lights, electrical motors, or cautery devices dur-ing recording is helpful.

Anesthesia can have a major detrimental impacton IOM, especially the use of inhalational agentsthat suppress cortical excitability. For SEPs, this neg-atively impacts scalp recordings more than cervicalspine or nasopharyngeal recordings. Anestheticagents reduce the effectiveness of transcranial elec-trical stimulation in initiating a MEP. NAP record-ings, however, are minimally affected by anesthesia.When recording directly from muscle, neuromuscu-lar blocking agents should be minimized or not usedat all; these agents may, however, be desirable forNAP, SEP, or MEP studies in which muscle artifactmust be eliminated. Our preferred anesthetic regi-men is intravenous narcotic and propofol with use ofmedium-acting neuromuscular blocking agents.Once adequate responses are obtained, low-level ha-logenated agents can be used. Both low-level inhala-tional agents and intravenous midazolam produceamnesia for the surgical procedure.

Stimulus artifact can be a challenge. Ensuringthe electrodes are lifted out of a wet field is impor-tant. The lowest stimulus intensity at the shortestduration possible should be used to achieve supra-maximal stimulation. The distance between stimula-tion and recording electrodes can be increased ifexposure in the surgical field permits this. Record-ing electrodes can be arranged in a monopolar fash-ion with G2 in adjacent tissue perpendicular to G1.Muscle artifact caused by volume conduction fromnearby sources can also distort the NAP. Properorientation of the electrodes and grounding must beassured. Neuromuscular blocking agents can be uti-lized to eliminate this artifact as well.

PERIPHERAL NERVE LESIONS

Peripheral nerve IOM is often performed to deter-mine continuity across an injured segment of nerveor for precise localization of a peripheral nerve le-sion.

In describing focal peripheral nerve injuries,there are two main classification systems. One, pro-posed by Sunderland,32 uses anatomical distinctions,whereas the other, by Seddon,28 utilizes the func-tional status of the nerve, that is, whether there isneurapraxia, axonotmesis, or neurotmesis. In neura-praxia, a functional block exists to conduction of theaction potential along the nerve; the ultimate prog-nosis is favorable given the preservation of axonaland neural architecture. In axonotmesis, disruptionof the axon occurs with some degree of intact neuralstructure, but the endoneurium is preserved. Recov-


ery requires regrowth of axons back to their targetorgan through the endoneurial tubes, following wal-lerian degeneration. Neurotmetic lesions involveboth the axonal and neural structure, making re-growth of the axon nearly impossible. For up to 4–7days after a nerve injury, the distal portion of theinjured nerve will be electrically excitable. After that,wallerian degeneration will occur back to the level ofthe nerve injury. Axonal regeneration then may pro-ceed from the lesion, moving distally at a rate ofabout 1 mm per day.

More complicated is the reinnervation of the endorgan, which must occur for functional recovery.After axons reach a muscle, there may be a delay ofseveral weeks to a month before reinnervation canbe detected. This is usually first detected by electro-physiological means (nascent MUAPs), then by visi-ble voluntary contraction.23 Depending on the dis-tance from the lesion to the end organ, a variableperiod of time must pass before clinical and electro-physiological signs of reinnervation become appar-ent. By the time it can be concluded that there is noreinnervation, the opportunity for a surgical ap-proach to reinnervation (i.e., nerve grafting) may belost since the best results are obtained when surgeryis performed in the first 6–12 months. With a prox-imal sciatic lesion, 6–12 months or more would berequired to conclude there is no reinnervation tothe anterior compartment of the leg from a clinicalor standard electrophysiological measurement. Atechnique for determining the likelihood of reinner-vation (or continuity) after a peripheral nerve lesionat an earlier time-point is therefore useful for guid-ing surgical intervention when it will be most suc-cessful.

Surgical decision-making in peripheral nerve le-sions is based on the determination of functioningaxons across a lesion, or continuity. This may beproven by clinical examination, as some preservedmotor or sensory function in the distribution of thenerve in question indicates a lesion in continuity.Preoperative electrophysiological assessment alsocontributes to this determination as the finding ofvoluntary MUAPs implies a lesion in continuity. Ifneither of the aforementioned clinical or electro-physiological findings are present, the lesion isthought to be “complete” clinically (i.e., there is nocontinuity proved) and often a surgical repair (nervegrafting) is considered. This is a challenge becausethere may not be the luxury of waiting for theseclinical or electrophysiological signs of reinnervationto occur or not occur, before deciding on surgery. Aspreviously discussed, there is a time window in whichsurgical intervention is more likely to be successful,

and this is typically shorter than the time it takes todetect signs of reinnervation (if it were to occur). Inthis setting, only intraoperative assessment can pro-vide a timely determination of whether there is func-tional continuity across a lesion. Visual inspection isof great importance as nerve thickening and neu-roma formation may suggest a lesion without conti-nuity, although there is no visual way to reliablydetermine physiological continuity of axons, espe-cially regenerating ones. The presence of a NAPacross a lesion, however, remains the gold standardby which many determine nerve continuity and,therefore, the type of surgical approach. It is clearthat a NAP can be obtained in the setting of a lesionthat appears complete clinically by routine preoper-ative electrophysiological testing.22 Lesions in conti-nuity are treated with neurolysis and are not grafted;they have a high likelihood of having a good out-come. Lesions with no NAP transmitted are thoughtto have no chance of spontaneous recovery and arethus treated by grafting or nerve transfer.22,31

Nerve stimulation and recording is performed asdescribed above. Stimulation is proximal to the le-sion, with recording distally. Ideally, the recordingelectrodes are placed just proximal to, then in, andfinally distal to the lesion. Recording a NAP overintact nerve (proximal to the lesion) helps as a con-trol to ensure functioning of the whole IOM system.When a NAP is recorded across a lesion, continuityof at least 4000 axons is likely and a surgical proce-dure limited to neurolysis is typically performed.21 Ina study by Kline et al., approximately 90% of patientsexperienced a good outcome.22

If there is no NAP across a lesion, there areessentially three possibilities. First, there may becomplete axonal disruption with no regeneratingaxons across the lesion. This would necessitate resec-tion and a nerve grafting or transfer procedure.Second, there may be the potential for axonal regen-eration that is being stunted by compression of fas-cicles by scarring within the nerve itself. An internalneurolysis can then be performed, allowing out-growth of these regenerating axons. This carries withit some risk for neural damage and disruption ofregenerating axons and is generally not performed.Unfortunately, distinction between these two possi-bilities is impossible electrophysiologically and mustbe made by the surgeon, taking into considerationthe appearance of the nerve.22,27 In some settings,although a NAP is recorded across the lesion, part ofthe nerve appears significantly injured and fascicularNAP recordings are helpful. Single fascicles are stim-ulated; those that demonstrate continuity are left


alone and those that do not are treated with primaryrepair or fascicular grafting.33

A third explanation for the lack of a NAP acrossa lesion is conduction block or neurapraxia. Thepreoperative electrophysiologic evaluation shouldhelp distinguish this, since signs of recent denerva-tion would be minimal to absent, and CMAP ampli-tudes evoked distal to the lesion would be normal.

Neurapraxia that does not completely block allconduction through a lesion will lead to responsesbeing recorded with both proximal and distal stim-ulation (across a lesion). Responses from the proxi-mal site may be longer in duration (temporal disper-sion), lower in amplitude (partial conductionblock), and arrive at the recording electrodes with alonger latency than expected (focal slowing). Inch-ing is a technique of stimulation at short, incremen-tal steps across a lesion while the change in morphol-ogy and latency of the evoked waveforms recordeddistally is assessed.5 When inching at 1-cm incre-ments, assuming a conduction velocity of 50 m/s,the onset of each successive waveform should beseparated by 0.2 ms. As the main goal of inching is tolocalize more precisely focal lesions, it is useful in theintraoperative setting when this localization is notprovided by preoperative electrophysiological stud-ies.

Given the above information, it is clear that pe-ripheral nerve IOM can be helpful when surgicallyapproaching a peripheral nerve lesion in which thelocalization or continuity of the lesion is in doubt.The approach to and utility of IOM of the mostcommon mononeuropathies is considered in whatfollows.

Ulnar Nerve Entrapment at the Elbow. Ulnar neurop-athy at the elbow (UNE) is the second most commonentrapment mononeuropathy, trailing only medianneuropathy at the wrist (carpal tunnel syndrome).While IOM is not often used for UNE surgery, it canbe useful in certain instances. The two main anatom-ical sites of compression at the elbow are the cubitaltunnel, formed by the two heads of the flexor carpiulnaris muscle, and the retroepicondylar groove be-tween the medial epicondyle and the olecranon. Thesurgical approach to compression at these two levelsis different: simple decompression with entrapmentat the cubital tunnel, and ulnar nerve transpositionor epicondylectomy if entrapment is more proximalin the retroepicondylar groove.

Preoperative electrophysiological studies mayconfirm UNE but not localize the lesion to one ofthese sites.13,23 Factors that contribute to this includevariability in anatomical location of these two poten-

tial entrapment sites, selective involvement of fasci-cles in the ulnar nerve, technical difficulties withoverstimulation (especially at the below-elbow site),and anastomotic nerve connections. The degree ofintraoperative electrophysiological abnormalities isoften more severe than expected based on routinepreoperative studies.22 Intraoperative studies mayeven show abnormalities when routine nerve con-duction studies across the elbow are normal.17 Addi-tionally, compression may be at a more distal sitethan expected.6,7 Intraoperative ulnar nerve studiesare therefore useful when localization is needed toguide the type and site of surgical intervention.

The ulnar nerve is usually easily exposed surgi-cally. If possible, stimulation and recording shouldbe performed on a normal portion of the nerve toensure a working system. Then, recording proxi-mally, the stimulating electrodes are placed at inter-vals closer, within, and distal to the area in question.An assessment is made for changes in amplitude,latency, and morphology as stimulation moves intothe abnormal segment of nerve. Based on intraop-erative ulnar nerve studies, the site of compression ismost commonly at or just proximal to the retroepi-condylar groove/medial epicondyle.2,17,22 In fact, a30-year study at Louisiana State University demon-strated compressive ulnar neuropathies localized tothe epicondyle level in over 97% of cases.17 Cubitaltunnel localization was seen more frequently inother studies.5,7 Based on these findings, the physi-cian should begin stimulating across the epicondylarregion. If no clear abnormalities are seen, the seg-ment of nerve across the cubital tunnel should beassessed. If these sites are normal, it is important torealize that the ulnar nerve can rarely be compressedat sites distal to the cubital tunnel, within or as itexits the flexor carpi ulnaris.4,7

Median Neuropathy at the Wrist. Median neuropa-thy at the wrist is the most common upper-extremityentrapment neuropathy. Routine electrophysiologi-cal studies are excellent at confirming the clinicaldiagnosis of carpal tunnel syndrome.1 As in UNE,intraoperative studies are rarely performed given therelative ease of the decompression procedure andthe lack of uncertainty in localization.

In one approach to intraoperative median nervestudies, stimulation of the nerve was performed in5-mm increments proximal to, through, and distal tothe carpal tunnel with surface recording over thethenar eminence.2 The most abnormal segment withrespect to focal slowing or conduction block corre-sponded to the segment of the most abnormal-appearing nerve. In the remainder of cases, the slow-


ing involved two or more 5-mm segments. The site ofthe most abnormal conduction was within the first10–20 mm just distal to the proximal border of theflexor retinaculum. Immediately upon release of themedian nerve in the carpal tunnel, latencies havebeen noted to improve10 or remain the same.37

Another technique has been described in whichthe median nerve is stimulated in the region of thecarpal tunnel with recording of the sensory digitalbranches on the third digit.24 This technique hasdemonstrated that the most abnormal segment(most conduction slowing and amplitude reduction)is the distal part of the carpal tunnel. This alsocorrelates with the area of highest intracarpal tunnelpressure measurements.

Common Peroneal Neuropathy at the Knee. Thecommon peroneal nerve can be compressed or dam-aged as it traverses the fibular head at the knee.Similar to median neuropathy at the wrist and UNE,localization with routine electrophysiological studiesis usually possible. Intraoperative peroneal nervestudies can be performed for localization and todetermine nerve continuity.2 One large series of sur-gically treated common peroneal neuropathies re-vealed that often there is no transmission of a NAPacross the lesion, leading to nerve graft repairs. Mostpatients in this series of 318 cases had suffered trau-matic peroneal neuropathies. Nontraumatic caseswith compression or entrapment at the knee (51 of318) were more likely to have recordable NAPs (42of 51) and thus undergo neurolysis as opposed tonerve grafting.18

USE IN BRACHIAL PLEXUS RECONSTRUCTION

Given the complexity of brachial plexus injuries interms of anatomy and type of injury, intraoperativeelectrophysiological monitoring is essential to en-hance clinical outcome. Brachial plexus injuries aretypically classified as preganglionic, postganglionic,or a combination of both. The continuity of thespinal root is likely the most important factor insurgical planning because root avulsion is nonrevers-ible with no chance for recovery spontaneously orwith primary anastomosis or grafting procedures. Ina suspected postganglionic injury, IOM is essential todetermine the presence of axonal continuity at atime when CMAPs or voluntary MUAPs cannot bedetected due to lack of distal axonal regeneration.

Preoperative evaluation is important to deter-mine the degree of both vertical (root or plexuslevel) and horizontal (i.e., preganglionic vs. postgan-glionic) involvement, but available methods are not

always adequate to make this determination. Factorssuggestive of preganglionic injury include: wingingof the scapula due to serratus anterior weakness;presence of Horner’s syndrome; or pseudomeningo-cele seen with myelography or magnetic resonanceimaging (MRI). Unfortunately, myelography andMRI may not be able to clearly identify root avulsionin some cases.8 The presence of sensory nerve actionpotentials (SNAPs) on routine NCSs in a flail limb issuggestive of a preganglionic lesion, but the absenceof these sensory responses may represent either apostganglionic process or a mixed process with ad-ditional root avulsion. The presence of prominentfibrillation potentials in cervical paraspinal musclesalso suggests root avulsion, but cannot determinethe precise root level. After surgical exposure, visualinspection may reveal the anatomical or structuralintegrity of a spinal root or nerve. If, however, anavulsed root remains intradural or, as is frequentlythe case, is scarred, one may be misled into assumingfunctional continuity of that spinal root or nerve. Inthis case the functional continuity of the axons can-not be determined without IOM techniques.

SEPs have been the mainstay of determining rootcontinuity. The presence of reproducible cervical orcortically recorded potentials is indicative of intactlarge-fiber sensory axons but does not provide infor-mation regarding motor axons. During surgery,baseline SEPs are recorded after stimulation of ei-ther an intact ipsilateral nerve or root or, alterna-tively, of the contralateral limb (usually mediannerve), to assure the recording system is workingproperly. The absence of these responses from scalprecordings can be seen with excessive inhalationalagents, although the cervical potentials should beminimally affected. The cervical potentials are gen-erally less reliable, however, given the possibility ofmuscle artifact and the short distance between stim-ulation and recording sites, making stimulus artifactmore of a confounder. Interaction with the anesthe-siologist prior to beginning these recordings is crit-ical to assure the best possible scalp responses. Aftersurgical exposure, direct root stimulation is per-formed under the surgeon’s direction, with record-ing over the scalp and cervical spine. In the recentpast, continuity at the root level, based on the pres-ence of a reproducible SEP, was believed to correlatewith the continuity of the motor axons in the ventralroot as well. In many cases, however, this proved notto be true in that grafting procedures led to noperipheral reinnervation. For this reason, followingSEP studies, the ventral root axons are assessed bymeans of MEP studies. If no reproducible responsescan be obtained when recording at the root level,


this suggests a lack of continuity of the motor axonsback to the anterior horn cell, which would elimi-nate direct root grafting or anastomosis.3,35 In thissetting, alternative surgical approaches, such as neu-rotization and tendon or muscle transfers, may beperformed. A reproducible MEP indicates ventralroot continuity, which allows the surgeon to considernerve grafting procedures. A potential pitfall in MEPrecording is a volume-conducted response originat-ing from nearby muscles rather than from the nerveitself. Use of a short-acting paralytic agent is neces-sary in this setting: volume-conducted muscle re-sponses will be suppressed, whereas MEPs directlyfrom the nerve are unaffected.35 The assessment ofSEPs and MEPs is then carried out on the remainingproximal roots or stumps that are not clearly avulsedon visual inspection.

Once it is determined whether the nerve rootsare in continuity with the spinal cord, attention turnsto defining functional conduction through the pe-ripheral segments that appear to have been injured.Although gross or microscopic inspection can givethe surgeon some indication of fascicular preserva-tion, IOM studies are critical to determine both thecontinuity across the lesion and the proximal extentof the process. This is best performed with NAPs. If,on visual inspection, there is a scarred but anatomi-cally intact segment, neurolysis only is performedwhen a NAP is present. If there is no reproducibleNAP, the lesion is generally resected, followed bynerve repair or grafting procedures. Determiningthe proximal extent of the resection is critical in thata functional fascicular structure in the proximalstump without intervening scar tissue will allow thegreatest chance for distal growth and reinnervation.In a pure postganglionic lesion, a NAP should bepresent with stimulation and recording proximal tothe lesion. Sequential recording is then performedmoving distally at approximately 1–2-cm intervalsuntil the response is lost, thus identifying the mostproximal extent of functioning axons. At that pointthe resection is performed followed by the graftingprocedure of the surgeon’s choice. It is important tocombine NAP with SEP and MEP recording becausea NAP could still be generated by sensory axons inthe setting of a ventral root avulsion, and this wouldlead to a futile attempt at regeneration if a directgrafting procedure was performed. Finally, once anerve lesion is identified, it may be clear that certainfascicles are disrupted whereas others have pre-served conduction. Internal neurolysis with split re-pairs may be necessary in this setting and NAP re-cordings to specific fascicles are critical for asuccessful outcome.31

Illustrative Case Report

The following case presents the use of these IOMtechniques. A 42-year-old man presented with a flailleft arm after a snowmobile accident. Six monthsafter the injury, complete loss of motor and sensoryfunction persisted in the arm. On clinical examina-tion voluntary contraction was possible only in therhomboid, trapezius, and serratus anterior musclesand diffuse sensory loss was present throughout thelimb. Stretch reflexes were absent. A Tinel’s sign waspresent in the left supraclavicular region, radiatingto the thumb. There was no evidence of Horner’ssyndrome. Routine NCSs of the left arm revealedabsent median and ulnar motor responses and alow-amplitude median SNAP. The lateral antebra-chial sensory response was absent. Needle examina-tion showed dense fibrillation potentials with noMUAPs activated in muscles in the left upper limbplus infraspinatus. Rhomboids were normal as werethe middle cervical paraspinals. Prominent fibrilla-tion potentials were noted in the low cervicalparaspinal muscles. These findings were consistentwith a severe left pan-brachial plexopathy withmixed preganglionic and postganglionic injury andprobable complete root avulsion affecting the lowersegments (C8, T1, and possibly C7). There was likelyat least partial preservation of the C5 root, althoughthis was difficult to predict with absolute certainty asin some cases the rhomboids may be innervated byC4. A computed tomography myelogram was consis-tent with left C8 and T1 nerve root avulsions. It wasfelt that he likely had intact outflow from C5 and C6.The continuity of C7 was uncertain, but given theserratus anterior activation, this was likely to be in-tact. Reconstructive surgery was indicated with IOMto determine root continuity and assist with surgicalplanning.

After surgical exposure, visual inspection sug-gested that the C5, C6, and C7 roots were intact,probably with significant postganglionic injury. TheC8 and T1 roots were visibly avulsed. IntraoperativeSEP testing with stimulation of the C5 and C6 rootsshowed reproducible responses (Fig. 1A and B). Mo-tor evoked potentials were also present on C5, C6,and C7 roots while under a short-acting paralyticagent (Fig. 1C–E). This confirmed that there wasboth motor and sensory root continuity at C5 and C6and at least motor root continuity at C7. Attentionthen turned to determining whether there was con-tinuity across the injured brachial plexus. No repro-ducible response could be recorded across the up-per or middle trunk segments, but a NAP waspresent stimulating and recording proximal to this


level (Fig. 1F). Based on these findings the followinggrafting procedures were performed: C5 to axillarynerve; C6 to musculocutaneous nerve; and C7 toradial nerve. Also performed was a transposition of

the contralateral C7 root to the left median nerve viaa vascularized ulnar nerve graft. In this particularcase, IOM confirmed continuity of the C5–C7 roots,which were then used as grafting vehicles, allowingthe contralateral C7 root to be used in an attempt toachieve more distal reinnervation. It also helped todefine the proximal extent of the nerve injury.

Expected Findings with Various Injuries

Figure 2 shows various injuries along with the ex-pected IOM findings. Figure 2A and B shows ventralroot avulsion, eliminating a grafting procedure, evenwith the presence of an SEP, as in Figure 2B. Acomplete postganglionic injury is depicted in Figure2C. The SEP and MEP are present, whereas a NAPacross the plexus is not present. Stimulation andrecording in the proximal plexus yields a NAP, iden-tifying the proximal extent of the lesion. In thissetting, nerve grafting/transfers or end-to-end anas-tomosis would be appropriate. If a NAP is presentacross elements of the brachial plexus (Fig. 2D),only neurolysis of that segment would be performed.A mixed lesion with sensory root avulsion and severepostganglionic injury is shown in Figure 2E. In thiscase, the utility of performing both MEP and SEPstudies is demonstrated, as absence of a SEP withouttesting of the MEP could be interpreted as repre-senting a low likelihood of a successful grafting pro-cedure when, in reality, this is likely to be mostbeneficial.

USE IN SELECTING FASCICLES FOR BIOPSY

Rarely, given a focal or multifocal process, will anerve biopsy be required of a proximal or nontradi-tional nerve (i.e., not a whole sural or superficialperoneal). Targets include nerve root, brachialplexus, or fascicles of proximal or distal nerves.These biopsies are useful, leading to diagnoses suchas lymphoma, focal inflammatory neuropathy, orsarcoidosis, all of which have varying treatments andprognoses. Clinical examination, preoperative elec-trophysiological testing, and imaging studies with3-Telsa MRI can all localize pathology to certainnerves or parts of nerves (i.e., the peroneal divisionof the sciatic nerve). During surgery, however, whenfaced with actually removing sections of these nerves,it is crucial to remove the part of the nerve with thegreatest chance of pathological diagnosis and leastchance of causing a new postoperative deficit. Visualassessment made by the surgeon upon exposure ofthe nerve at the time of biopsy is helpful. Electro-

FIGURE 1. Intraoperative recordings during brachial plexus ex-ploration and reconstruction. SEP recording over the scalp andneck, respectively, with direct intraoperative C5 root stimulation(A) and C6 root stimulation (B). The presence of the responsesupports C5 and C6 sensory root integrity. Transcortical electricalstimulation with direct MEP recording over the C5 root (C), the C6root (D), and the C7 root (E) supporting integrity of C5, C6, andC7 motor roots. Nerve action potential with stimulation and re-cording proximal to the lesion (F); no reproducible responsecould be recorded across the lesion.


physiological testing can further add to this impor-tant assessment.

Just prior to biopsy, direct stimulation is applied tothe individual fascicles in question. An assessment ismade by the presence or absence of a downstreamtwitch in muscles innervated by the nerve, or by record-ing from muscle (surface or EEG needle electrodes) ornerve (as a NAP) as described earlier. This allows forthe identification of functioning and nonfunctioningfascicles. In this manner, a biopsy can be performedwith the highest diagnostic yield and lowest risk.

USE IN OPERATING ON NERVE TUMORS

Peripheral nerve tumors are rare. These are broadlyseparated into neural sheath tumors and nonneuralsheath tumors. Examples of the former include be-

nign entities such as neurofibromas (with or withoutneurofibromatosis type I) and schwannomas. Sarco-mas (neurogenic, fibrosarcoma, spindle cell, syno-vial, or perineurial sarcomas) are malignant neuralsheath tumors. Nonneural sheath tumors includeganglion cysts, lipomas, hypertrophic neuropathy(although some of these may be inflammatory), vas-cular tumors, and desmoid tumors. Metastatic carci-nomas can affect nerve; most commonly this is froma lung or breast primary tumor. The location forthese peripheral nerve tumors varies, with the bra-chial plexus and upper extremity being most com-mon, followed by the lower extremity and, uncom-monly, the lumbosacral plexus.19,20

The use of NAP recordings can be helpful intra-operatively. A stimulator can be used to localize

FIGURE 2. Expected intraoperative recording with various lesions of the roots or plexus. (A) Complete ventral and dorsal avulsion. Preservationof nerve action potential (NAP) due to intact sensory fibers, with MEP and SEP absent. (B) Ventral root disruption with preservation of dorsalroot. Preservation of NAP and somatosensory evoked potentials (SEP) with root stimulation but MEP absent. (C) Complete postganglioniclesion. Preservation of SEP, MEP and NAP proximal to lesion. No NAP recorded across the lesion, indicating no evidence of functional axonsacross the lesion. (D) Severe postganglionic lesion but with some regeneration through the injured segment. Note the presence of all responses,with the exception of the compound muscle action potential amplitude, given there has not yet been end-organ reinnervation. (E) Mixedpreganglionic and postganglionic lesions.


peripheral nerve if the architecture or anatomy isconfusing. Functioning fascicles are identified in or-der to protect them if possible, and thereby limitpostoperative neurological deficit. In most cases,complete tumor removal takes precedence and fas-cicles may need to be sacrificed. Fortunately, fasci-cles with tumor involvement are usually nonfunc-tioning.31

References

1. American Association of Electrodiagnostic Medicine, Ameri-can Academy of Neurology, American Academy of PhysicalMedicine and Rehabilitation. Practice parameter for electro-diagnostic studies in carpal tunnel syndrome: summary state-ment. Muscle Nerve 2002;25:918–922.

2. Brown WF, Veitch J. AAEM minimonograph #42: Intraoper-ative monitoring of peripheral and cranial nerves. MuscleNerve 1994;17:371–377.

3. Burkholder LM, Houlden DA, Midha R, Weiss E, VennettilliM. Neurogenic motor evoked potentials: role in brachialplexus surgery. Case report. J Neurosurg 2003;98:607–610.

4. Campbell WW. AAEE case report #18: Ulnar neuropathy inthe distal forearm. Muscle Nerve 1989;12:347–352.

5. Campbell WW. The value of inching techniques in the diag-nosis of focal nerve lesions. Inching is a useful technique.Muscle Nerve 1998;21:1554–1556.

6. Campbell WW, Pridgeon RM, Sahni SK. Entrapment neurop-athy of the ulnar nerve at its point of exit from the flexor carpiulnaris muscle. Muscle Nerve 1988;11:467–470.

7. Campbell WW, Pridgeon RM, Sahni SK. Short segment incre-mental studies in the evaluation of ulnar neuropathy at theelbow. Muscle Nerve 1992;15:1050–1054.

8. Carvalho GA, Nikkhah G, Matthies C, Penkert G, Samii M.Diagnosis of root avulsions in traumatic brachial plexus inju-ries: value of computerized tomography, myelography andmagnetic resonance imaging. J Neurosurg 1997;86:69–76.

9. Daube JR. Assessing the motor unit with needle electromyo-graphy. In: Daube JR, editor. Clinical neurophysiology. Phil-adelphia: FA Davis; 1996. p 271.

10. Eversmann WW, Ritsick JA. Intraoperative changes in motornerve conduction latency in carpal tunnel syndrome. J HandSurg Am 1978;3:77–81.

11. Gunnarsson T, Krassioukov AV, Sarjeant R, Fehlings MG.Real-time continuous intraoperative electromyographic andsomatosensory evoked potential recordings in spinal surgery:correlation of clinical and electrophysiologic findings in aprospective, consecutive series of 213 cases. Spine 2004;29:677–684.

12. Harper CM. Intraoperative cranial nerve monitoring. MuscleNerve 2004;29:339–351.

13. Harper CM. Peripheral nervous system monitoring. In: DaubeJ, editor. Clinical neurophysiology. Philadelphia: FA Davis;1996. p 465–466.

14. Harper CM. Preoperative and intraoperative electrophysi-ologic assessment of brachial plexus injuries. Hand Clin 2005;21:39–46.

15. Hashimoto T, Mitomo M, Hirabuki N, Miura T, Kawai R,Nakamura H, et al. Nerve root avulsion of birth palsy: com-parison of myelography with CT myelography and somatosen-sory evoked potential. Radiology 1999;178:841–845.

16. Holland NR. Intraoperative electromyography. J Clin Neuro-physiol 2002;19:444–453.

17. Kim DH, Han K, Tiel RL, Murovic JA, Kline DG. Surgicaloutcomes of 654 ulnar nerve lesions. J Neurosurg 2003;98:993–1004.

18. Kim DH, Murovic JA, Tiel RL, Kline DG. Management andoutcomes in 318 operative common peroneal nerve lesions atthe Louisiana State University Health Sciences Center. J Neu-rosurg 2004;54:1421–1429.

19. Kim DH, Murovic JA, Tiel RL, Moes G, Kline DG. A series of146 peripheral non-neural sheath nerve tumors: 30-year ex-perience at Louisiana State University Health Sciences Cen-ter. J Neurosurg 2005;102:256–266.

20. Kim DH, Murovic JA, Tiel RL, Moes G, Kline DG. A series of397 peripheral neural sheath tumors: 30-year experience atLouisiana State University Health Sciences Center. J Neuro-surg 2005;102:246–255.

21. Kline DG, Hackett ER, May PR. Evaluation of nerve injuries byevoked potentials and electromyography. J Neurosurg 1969;31:128–136.

22. Kline DG, Happel LT. Penfeld lecture. A quarter century’sexperience with intraoperative nerve action potential record-ing. Can J Neurol Sci 1993;20:3–10.

23. Kline DG, Hudson AR, Zager E. Selection and preoperativework-up for peripheral nerve surgery. Clin Neurosurg 1992;39:8–35.

24. Luchetti R, Schoenhuber R, Alfarano M, Deluca S, DeCiccoG, Landi A. Carpal tunnel syndrome: correlations betweenpressure measurement and intraoperative electrophysiologi-cal nerve study. Muscle Nerve 1990;13:1164–1168.

25. Nelson KR. Use of peripheral nerve action potentials forintraoperative monitoring. Neurol Clin 1988;6:917–933.

26. Oberle J, Antoniadis G, Kast E, Richter HP. Evaluation oftraumatic cervical nerve root injuries by intraoperative evokedpotentials. Neurosurgery 2002;51:1182–1190.

27. Oberle J, Antoniadis G, Rath SA, Richter HP. Value of nerveaction potentials in the surgical management of traumaticnerve lesions. Neurosurgery 1997;41:1337–1344.

28. Seddon HJ. Three types of nerve injury. Brain 1943;66:237–288.

29. Shin AY, Spinner RJ, Steinmann SP, Bishop AT. Adult trau-matic brachial plexus injuries. J Am Acad Orthop Surg 2005;13:382–396.

30. Slimp JC. Intraoperative monitoring of nerve repairs. HandClin 2000;16:25–36.

31. Spinner RJ, Kline DG. Surgery for peripheral nerve and bra-chial plexus injuries or other nerve lesions. Muscle Nerve2000;23:680–695.

32. Sunderland S. The anatomy and physiology of nerve injury.Muscle Nerve 1990;13:771–784.

33. Terzis JK, Dykes RW, Hakstian RW. Electrophysiological re-cordings in peripheral nerve surgery: a review. J Hand SurgAm 1976;1:52–66.

34. Tiel RL, Happel LT Jr, Kline DG. Nerve action potentialrecording method and equipment. Neurosurgery 1996;39:103–109.

35. Turkof E, Millesi H, Turkof R, Pfundner P, Mayr N. Intraop-erative electroneurodiagnostics (transcranial electrical motorevoked potentials) to evaluate the functional status of ante-rior spinal roots and spinal nerves during brachial plexussurgery. Plast Reconstr Surg 1997;99:1632–1641.

36. Veale JL, Mark RF, Rees S. Differential sensitivity of motorand sensory fibres in human ulnar nerve. J Neurol NeurosurgPsychiatry 1973;36:75–86.

37. Yates SK, Hurst LN, Brown WF. Physiological observations inthe median nerve during carpal tunnel surgery. Ann Neurol1981;10:227–229.


Peripheral Nerve Monitoring During Operative Procedures · tive monitoring (IOM) requires a skilled...

Documents

Transcript of Peripheral Nerve Monitoring During Operative Procedures · tive monitoring (IOM) requires a skilled...