Jpn J Rehabil Med 2005; 42: 89-98

2004年／第 41回 日本リハビリテーション医学会 学術集会／東京 Invited Lecture

Traumatic Injury to Peripheral

Lawrence R Robinson, MD

isolated nervous system injury, but may also

Epidemiology of Peripheral often accompany CNS trauma, not only

Trauma compounding the disability, but making recog- nition of the peripheral nerve lesion problem- Traumatic injury to peripheral nerves results atic. Of patients with peripheral nerve injuries, in considerable disability across the world. In about 60% have a . peacetime, peripheral nerve injuries commonly Conversely, of those with traumatic brain result from trauma due to motor vehicle acci- injury admitted to rehabilitation units,10-34% dents, and less commonly from penetrating have associated peripheral nerve injuries. trauma, falls and industrial accidents. Out of It is often easy to miss peripheral nerve injuries all patients admitted to Level I trauma centers, in the setting of CNS trauma. Since the neuro- it is estimated that roughly 2-3% have periph- logic history and examination is limited, early eral nerve injuries. If plexus and root hints to a superimposed peripheral nerve lesion injuries are also included,the incidence is about might be only flaccidity,areflexia,and reduced

5%. movement of a limb.

In the upper limb,the nerve most commonly Peripheral nerve injuries are of significant reported injured is the radial nerve,followed by import as they impede recovery of function and ulnar and median nerves. Lower limb return to work, and carry risk of secondary peripheral nerve injuries are less common,with disabilities from falls,fractures,or other secon- the sciatic most frequently injured,followed by dary injuries. An understanding of the classifi- peroneal and rarely tibial or femoral nerves. cation,pathophysiology and electrodiagnosis of

Fractures of nearby bones are commonly as- these lesions is critical to the appropriate diag- sociated,such as humeral fractures with radial nosis, localization and management of periph- neuropathy. eral nerve trauma. In wartime,peripheral nerve trauma is much more common and much of our knowledge Classification of Nerve Injuries about peripheral ,repair and recov- ery comes from experience derived in World There are two predominant schemes that

War I and II, and subsequent wars. have been proposed for classification of periph- Peripheral nerve injuries may be seen as an eral nerve traumatic injuries; that of Seddon

Used by permission, copyright 2000AAEM. This manuscript is based on the monograph “Traumatic Injury to

Peripheral Nerves”published by the AAEM in Muscle& Nerve 2000; 23: 863-873. The monograph can be

purchased from AAEM at (507)-288-0100. Professor and Chair, Department of Rehabilitation Medicine, University of Washington, Seattle, WA 98195

リハビリテーション医学 VOL.42 NO.2 2005年 2月 Lawrence R ROBINSON

Table 1 Classification systems for nerve injury

Seddon Sunderland Pathology Prognosis Classification Classification

Neurapraxia First Degree Injury or Ischemia Excellent recovery in weeks to months

Axonotmesis Disrupted Good to poor, depending upon integrity of

Variable Stromal Disruption supporting structures and distance to muscle

Second Degree Axons Disrupted Good, depending upon distance to muscle

Endoneurial Tubes Intact

Perineurium Intact

Epineurium Intact Third Degree Axons Disrupted Poor

Endoneurial Tubes Disrupted Axonal misdirection

Perineurium Intact Surgery may be required

Epineurium Intact Fourth Degree Axons Disrupted Poor

Endoneurial Tubes Disrupted Axonal misdirection

Perineurium Disrupted Surgery usually required

Epineurium Intact Fifth Degree Disrupted No spontaneous recovery

Endoneurial Tubes Disrupted Surgery required

Perineurium Disrupted Prognosis after surgery guarded

Epineurium Dis rupted Adapted from Dillingham. and that of Sunderland(Table 1). The for- scar tissue that axonal regrowth is impossible. mer is more commonly used in the literature. Examples are s harp injury, some traction

Seddon has used the terms , axono t- injuries or injection of noxious drugs. Progno- mesis, and neurotmesis to describe peripheral sis for spontaneous recovery is extremely poor nerve injuries.Neurapraxia is a comparative- without surgical intervention. Sunderland ly mild injury with motor and sensory loss but uses a more subdivided scheme to describe no evidence of . The peripheral nerve injuries, with five groups nerve distally conducts normally. Focal instead of three. demyelination and/or ischemia are thought to be the etiologies of the conduction block. Electrodiagnosis: Timing of Changes and

Recovery may occur within hours,days,weeks, Determining Degree of Injury or up to a few months. Axonotmesis is common- ly seen in crush injuries. The axons and their 1. The Compound Motor Action Potential myelin sheaths are broken,yet the surrounding Neurapraxia: In purely neurapraxic lesions, stroma (i.e.the ,perineurium,and the compound muscle action potential(CMAP) epineurium) remains partially or fully intact. will change immediately after injury,assuming

Wallerian degeneration occurs, but subsequent one can stimulate both above and below the site axonal regrowth may proceed along the intact of the lesion (Fig.1). When recording from endoneurial tubes. Recovery ultimately distal muscles and stimulating distal to the site depends upon the degree of internal disorgani- of the lesion, the CMAP should always be zation in the nerve as well as the distance to the normal since no axonal loss and no Wallerian end organ. Sunderland’s classification (below) degeneration has occurred. Moving stimulation further divides this category. Neurotmesis proximal to the lesion will produce a smaller or describes a nerve that has been either complete- absent CMAP, as conduction in some or all ly severed or is so markedly disorganized by fibers is blocked. It should be remembered that

Jpn J Rehabil Med VOL.42 NO.2 2005 Traumatic Injury to Peripheral Nerves

Fig.1 Diagrammatic representation of changes in Fig.2 Diagrammatic representation of changes in

the compound muscle action potential the compound muscle action potential (CMAP) after axonotmesis and neurot- (CMAP)after neurapraxia. mesis. amplitudes normally fall with increasing dis- tance between stimulation and recording ; “few days”thereafter, the CMAP and motor hence there is some debate about how much of conduction studies look the same as those seen a drop in amplitude is sufficient to demonstrate in a neurapraxic lesion. Nerve segments distal conduction block. Amplitude drops exceeding to the lesion remain excitable and demonstrate

20% over a 25cm distance or less are clearly normal conduction while proximal stimulation abnormal; smaller changes over smaller dis- results in an absent or small response from tances are likely also suggestive of an abnor- distal muscles . Early on,this picture looks the mality. In addition to conduction block,partial same as conduction block and can be confused lesions also often demonstrate concomitant with neurapraxia. Hence neurapraxia and slowing across the lesion. This slowing may be axontomeis can not be distinguished until due to either loss of faster conducting fibers or sufficient time for Wallerian degeneration in all demyelination of surviving fibers. All these motor fibers has occurred, typically about 9 changes in the CMAP will generally persist days post injury(Fig.2). until recovery takes place,typically by no more As Wallerian degeneration occurs,the ampli- than a few months post injury. Most important- tude of the CMAP elicited with distal stimula- ly, the distal CMAP will never drop in ampli- tion will fall. This starts at about day3and is tude in purely neurapraxic injuries, since no complete by about day 9. Neuromuscular axon loss or Wallerian degeneration occurs and junction transmission fails before nerve ex- the distal nerve segment remains normally citability.Thus in complete axonotmesis at excitable. day 9, one has a very different picture from

Axonotmesis and Neurotmesis : Electro- neurapraxia. There are absent responses both diagnostically, complete axonotmesis (equiva- above and below the lesion. Partial axon loss lent to Sunderland grades 2,3and 4)and com- lesions will produce small amplitude motor plete neurotmesis look the same, since the responses, with the amplitude of the CMAP difference between these types of lesions is in roughly proportional to the number of surviving the integrity of the supporting structures,which axons. One can compare side-to-side CMAP have no electrophysiologic function. Thus amplitudes to estimate the degree of axon loss, these lesions can be grouped together as though inherent side to side variability of up to axonotmesis for the purpose of this discussion. 30-50% limits the accuracy of the estimate. Immediately after axonotmesis and for a Using the CMAP amplitude to estimate the

リハビリテーション医学 VOL.42 NO.2 2005年 2月 Lawrence R ROBINSON

2. Compound or Action

Potentials

Neurapraxia: The sensory nerve action

potential (SNAP)and compound nerve action

potential(CNAP)will show changes similar to

the CMAP after focal nerve injury. In the

setting of neurapraxia,there is a focal conduc- tion block at the site of the lesion,with preser- ved distal amplitude. However,the criteria for

establishing conduction block in sensory nerve

Fig.3 Diagrammatic representation of changes in fibers are substantially different than that for

the compound muscle action potential the CMAP. When recording nerve action

(CMAP)after mixed lesion. potentials, there is normally a greater drop in

amplitude over increasing distance between degree of surviving axons is also most reliable stimulating and recording electrodes, due to only early after injury,before axonal sprouting temporal dispersion and phase cancellation. has occurred. Use of this technique later after Amplitude drops of 50-70% over a 25cm dis- injury will tend to underestimate the degree of tance are not unexpected and it is less clear just axon loss. what change in amplitude is abnormal. A large

Mixed Lesions: Lesions which have a mix- focal change over a small distance is probably ture of axon loss and conduction block provide significant. Slowing may also accompany par- a unique challenge. These can usually be sorted tial conduction blocks, as for the CMAP. out by carefully examining amplitudes of th e Responses elicited with stimulation and record-

CMAP elicited from stimulation both above ing distal to the lesion are normal in pure and below the lesion and by comparing the neurapraxic injuries. amplitude with distal stimulation to that Axonotmesis and Neurotmesis: Immediate- obtained from the other side. The percentage ly after axonotmesis,the SNAP looks the same of axon loss is best estimated by comparing the as seen in a neurapraxic lesion. Nerve seg-

CMAP amplitude from distal stimulation with ments distal to the lesion remain excitable and that obtained contralaterally. Of the remaining demonstrate normal conduction while proximal axons, the percentage with conduction block stimulation results in an absent or small are best estimated by comparing amplitudes or response. Hence neurapraxia and axontomeis areas obtained with stimulation distal and prox- cannot be distinguished until sufficient time for imal to the lesion. Thus if a 1mV response is Wallerian degeneration in all sensory fibers has obtained with proximal stimulation, a 2mV occurred, typically about 11days post injury. response is obtained distally, and a 10mV It takes slightly longer for sensory nerve response is obtained with distal stimulation studies to demonstrate loss of amplitude than contralaterally, one can deduce that probably for motor studies,i.e.11days vs.9days,due to about 80% of the axons are lost, and of the the earlier failure of neuromuscular junction remaining 20%,half are blocked (neurapraxic) transmission compared to nerve conduction. at the lesion site. As mentioned above, this analysis is most useful only in the acute phase, 3. Needle before reinnervation by axonal sprouting Neurapraxia: The needle EMG examination occurs (Fig.3). in purely neurapraxic lesions will show neuro- genic changes in recruitment with debatable

Jpn J Rehabil Med VOL.42 NO.2 2005 Traumatic Injury to Peripheral Nerves abnormalities in spontaneous activity. As When the lesion is distal and the distal stump is mentioned earlier,there is debate as to whether short, it takes only 10-14days for fibrillations

fibrillation potentials are recorded after a pure- to develop. With a proximal lesion and a longer ly neurapraxic lesion. One study of peripheral distal stump (e.g. ulnar innervated hand mus- nerve lesions in baboons has failed to demon- cles in a brachial plexopathy), 21-30 days are strate fibrillations in purely neurapraxic required for full development of fibrillation lesions. On the other hand, study of purely potentials and positive sharp waves. Thus, neurapraxic lesions in rats,has suggested the electromyographer needs to be acutely

fibrillations occur in blocked,but not denervat- aware of the time since injury,so that severity ed,muscle fibers. There are limited reports of is not underestimated when a study is perform-

fibrillations in humans with apparently predom- ed early after injury and also so that develop- inantly neurapraxic nerve lesions,but it is ment of increased fibrillation potentials over difficult to know whether or not any axon loss time is not misinterpreted as a worsening of the had occurred in these patients, since nerve injury. conduction studies are not sensitive for detect- and positive sharp wave density ing minimal axon loss. Needle EMG is more are usually graded on a 1-4 scale. This is an sensitive for detecting motor axon loss than ordinal scale,meaning that as numbers increase nerve conduction studies,and hence it is easy to findings are worse. However it is not an inter- imagine situations in which nerve conduction val or ratio scale,i.e.4＋ is not twice as bad as studies are within normal limits, but needle 2＋ or 4 times as bad as 1＋. Moreover, 4＋

EMG detects minimal or mild axon loss. fibrillation potentials does not reflect complete

Independent of whether or not the needle axon loss, and in fact may represent only a

EMG demonstrates fibrillation potentials in minority of axons lost.Evaluation of recruit- neurapraxia,the most apparent change on nee- ment and particularly of distally elicited CMAP dle EMG will be changes in recruitment. These amplitude are necessary before one can decide occur immediately after injury. In complete on whether or not complete axon loss has lesions (i.e. complete conduction block) there occurred. will be no motor unit action potentials. In Fibrillation potential size will decrease over incomplete neurapraxic lesions, there will be time since injury. Krafthas demonstrated reduced numbers of motor unit action poten- that fibrillations initially are several hundred tials firing more rapidly than normal (i.e. microvolts in the first few months after injury. reduced or discrete recruitment). Recruitment However, when lesions are more than 1 year changes alone are not specific for neurapraxia old,they are unlikely to be over 100μV in size. or axon loss. will also decrease in number as

Since no axon loss occurs in neurapraxic reinnervation occurs,however this finding is not injuries, there will be no axonal sprouting and usually clinically useful for two reasons. First, no changes in MUAP morphology (e.g. dura- since a qualitative or ordinal scale of fibrilla- tion, amplitude or phasicity) anytime after tion density is typically used and an accurate injury. quantitative measurement of fibrillation density

Axonotmesis and Neurotmesis: A number is not available,comparison of fibrillation num- of days after an axon loss lesion, needle EMG bers from one examination to the other is not will demonstrate fibrillation potentials and posi- reliable9. Second, even in complete lesions, tive sharp waves. The time between injury and fibrillation density will eventually reduce since onset of fibrillation potentials will be dependent the muscle becomes fibrotic and the number of in part upon the length of distal nerve stump. viable muscle fibers falls; in this case, reduc-

リハビリテーション医学 VOL.42 NO.2 2005年 2月 Lawrence R ROBINSON tion in fibrillation numbers does not predict other muscles can be deceptive and could erro- recovery, but rather muscle fibrosis. neously suggest intact innervation. Fibrillations may also occur after direct Mixed Lesions: When there is a lesion with muscle injury,as well as nerve injury. Partanen both axon loss and conduction block, needle and Dannerhave demonstrated that patients EMG examination can be potentially mislead- after muscle biopsy have persistent fibrillation ing if interpreted in isolation. If, for example, potentials starting after 6-7days and extending a lesion results in destruction of 50% of the for up to 11 months. In patients who have original axons and conduction block of the undergone multiple trauma, coexisting direct other 50%, then needle EMG will demonstrate muscle injury is common and can be potentially abundant (e.g. 4＋) fibrillation potentials and misleading when trying to localize a lesion. no voluntary MUAPs. The electromyographer

When there are surviving axons after an should not then conclude that there is a com- incomplete axonal injury, remaining MUAPs plete axonal lesion,but should instead carefully are initially normal in morphology,but demon- evaluate the motor nerve conduction studies to strate reduced or discrete recruitment. Axonal figure out how much of the lesion is neuraprax- sprouting will be manifested by changes in ic and how much axonotmetic. The important morphology of existing motor units. Amplitude point here is to not take the presence of abun- will increase, duration will become prolonged, dant fibrillations and absent voluntary MUAPs and the percentage of polyphasic MUAPs will as evidence of complete . increase as motor unit territory increases. This process occurs soon after injury. Micro- Localization of Traumatic Nerve scopic studies demonstrate outgrowth of these Injuries nerve sprouts starting at 4 days after parti al denervation. Electrophysiologic studies The localization of peripheral nerve injuries utilizing single fiber EMG demonstrates is sometimes straightforward but is potentially increase in fiber density starting at 3weeks post complicated by a variety of possible pifalls. injury. Localization is usually performed by two

In complete lesions,the only possible mecha- methods: 1) detecting focal slowing or conduc- nism of recovery is axonal regrowth. The tion block on nerve conduction studies, or 2) earliest needle EMG finding in this case is the assessing the pattern of denervation on needle presence of small, polyphasic, often unstable EMG. motor unit potentials previously referred to as Localizing peripheral nerve lesions by nerve

“nascent potentials.” Observation of these conduction studies usually requires that there potentials is dependent upon establishing axon be a focal slowing or conduction block as one regrowth as well as new neuromuscular junc- stimulates above and below the lesion. To see tions and this observation represents the ear- such a change there must either be focal liest evidence of reinnervation, usually preced- demyelination or ischemia,or the lesion should ing the onset of clinically evident voluntary be so acute that degeneration of the distal movement. These potentials represent the stump has not yet occurred. Thus lesions with earliest definitive evidence of axonal reinnerva- partial or complete neurapraxia (due to either tion in complete lesions. When performing the demyelination or ischemia) can be well local- examination looking for new motor unit poten- ized with motor nerve conduction studies, as tials, one must be sure to accept only “crisp”, can very acute axonal injuries. nearby motor unit potentials with a short rise- In pure axonotmetic or neurotmetic lesions,it time, since distant potentials recorded from is more difficult if not impossible to localize the

Jpn J Rehabil Med VOL.42 NO.2 2005 Traumatic Injury to Peripheral Nerves lesion using nerve conduction studies. In such a EMG. Conceptually, if one knows the branch- case, there will be mild and diffuse slowing in ing order to various muscles under study, one the entire nerve due to loss of the fastest fibers, can determine that the nerve injury is between or there will be no response at all. Conduction the branches to the most distal normal muscle across the lesion site will be no slower than and the most proximal abnormal muscle. across other segments. In addition, provided There are, however, a number of potential enough time for Wallerian degeneration has problems with this approach. First,the branch- elapsed (i.e.at least 9days for motor fibers or ing and innervation for muscles is not necessar- 11 days for sensory fibers), there will be no ily consistent from one person to another. change in amplitude as one traverses the site of Sunderlandhas demonstrated a great deal of the lesion. Thus,pure axon loss lesions are not variability in branching order to muscles in the well localized along a nerve by nerve conduc- limbs, variability in the number of branches tion studies. going to each muscle, and variability in which

There are some cases in which indirect infer- nerve or nerves supply each muscle. Thus,the ences can be made about the location of purely typical branching scheme may not apply to the axonal lesions. For instance,if the ulnar motor patient being studied and consequently the response is very small or absent and the median lesion site can be misconstrued. motor response is normal,this implies an ulnar Second, the problem of muscle trauma and neuropathy rather than a lower brachial plexus associated needle EMG findings can be mislead- lesion. However,in such an instance,the site of ing. As mentioned earlier, direct muscle pathology along the ulnar nerve may not be trauma can result in positive sharp waves and well defined. fibrillations for months or longer after injury. Another indirect inference that can be made Practically speaking, this can result in errone- based upon sensory nerve conduction studies is ously proximal lesion sites,or error in diagnos- placement of the lesion at a pre vs. post gang- ing more than one lesion. For example,in the lionic location. Lesions that are proximal to setting of humeral fracture with radial neur- the dorsal root ganglion, i.e. at the pre- opathy, the triceps not infrequently demon- ganglionic level (proximal root, cauda equina, strates fibrillation potentials, due to direct spinal cord)tend to have normal sensory nerve muscle trauma. However,one could be mislead action potential amplitudes,even in the setting to localize the lesion to the axilla or higher of reduced or absent sensation. This is a rather than spiral groove,if the triceps findings particularly bad prognostic sign when seen in are not recognized to come from direct muscle the setting of possible root avulsion. On the rather than nerve injury. other hand, lesions occurring distal to the dor- Third, the problem of partial lesions can sal root ganglion have small or absent SNAPs make for misdiagnosis to more distal sites. In

(when these are recorded in the appropriate partial ulnar nerve lesions at the elbow, for distribution). Thus, SNAPs may be useful to example,the forearm ulnar innervated muscles differentiate root vs. plexus or other pre-vs. are often spared. This is thought at least post-ganglionic locations. A limitation,particu- partially due to sparing of the fascicles in the larly in partial lesions,is the wide variability in nerve that are preparing to branch to the flexor

SNAP amplitudes seen in normal individuals. digitorum profundus and the flexor carpi ulnar- Mixed pre-and post-ganglionic lesions are also is, i.e. they are in a relatively protected posi- potentially difficult to interpret. tion. This finding could lead one to inadvertent- The other major electrodiagnostic method of ly localize the lesion distally to the distal fore- determining the site of nerve injury is by needle arm or wrist. Similarly,a lesion involving the

リハビリテーション医学 VOL.42 NO.2 2005年 2月 Lawrence R ROBINSON median nerve in the arm (above the elbow)has the worst prognosis. Recovery depends solely been reported to cause findings only in the upon axonal regeneration which may or may anterior interosseous distribution.Intraneur- not occur, depending upon the degree of injury al topography needs to be considered when to the nerve. In many cases of complete axon making a diagnosis based on branching. loss it is not possible to know the degree of

nerve injury except by surgical exploration

Electrodiagnostic Evaluation of with or without intraoperative recording, or

Prognosis looking for evidence of early reinnervation

after the lesion. As a consequence, it is often

Determining the pathophysiology of a periph- recommended to wait 2-4months and look for eral nerve traumatic injury can help with evidence of reinnervation in previously com- estimating prognosis. Those injuries that are pletely denervated muscles near the site of the completely or largely neurapraxic have a good lesion.Those lesions that have some spon- prognosis for recovery within a few months taneous recovery are usually treated conserva- (usually up to three months post injury). tively since operative repair is unlikely to

Resolution of ischemia and remyelination improve upon natural recovery. Those with no should be complete by this time. evidence of axonal regrowth usually have oper- Mixed injuries typically have two or more ative exploration with possible grafting. phases of recovery. The neurapraxic compo- Acknowledgement nent resolves quickly as above and muscle fiber Paula Micklesen is thanked for production of hypertrophy can provide additional recovery, figures for this manuscript. but the axonal component is slower, since it References depends upon distal axonal sprouting and o n axonal regeneration from the site of the lesion. 1) Brandstater ME, Fullerton M : Sensory nerve con- duction studies in cervical root lesions. Can J Thus patients usually experience a relatively Neurol Sci 1983; 10: 152 rapid partial but incomplete recovery followed 2) Brown MC, Holland RL, Hopkins AG : Motor a slower further recovery. Sensory recovery nerve sprouting.Ann Rev Neurosci 1981; 4: 17-42 may proceed for a longer time than motor. 3) Buchthal F : Fibrillations : clinical electro- physiology.in Abnormal Nerves and Muscle Gener- Partial axon loss lesions usually represent ators (ed by Culp WJ,Ochoa J).Oxford University axonotmesis, though a partial neurotmesis Press, New York, 1982; pp632-662 (e.g. a laceration through part of the nerve) 4) Campbell WW, Pridgeon RM, Riaz G,et al: Spar- ing of the flexor carpi ulnaris in ulnar neuropathy cannot always be excluded in such cases. In at the elbow. Muscle Nerve 1989; 12: 965-967 axonotmesis,recovery will depend upon axonal 5) Cangiano A, Lutzemberger L, NicotraL : Non- sprouting and regeneration. Thus there will be equivalence of impulse blockade and denervation in

the production of membrance changes in rat some early recovery followed possibly by a 1977 691-706 later recovery if or when regenerating axons . J Physiol ; 273: 6) Chaudry V,Cornblath DR : Wallerian degeneration reach their end organs. The amplitude of the in human nerves: Serial electrophysiological

CMAP provides some guide to prognosis. In studies. Muscle Nerve 1992; 15: 687-693

7) Cosgrove JL, Vargo M, Reidy ME : A prospective facial nerve lesions, it has been demonstrated study of peripheral nerve lesions occurring in trau- that patients with CMAP amplitudes 30% or matic brain-injured patients. Am J Phys Med Re- more of the other side have an excellent out- habil 1989; 68: 15-17 come, those with 10-30% have good but not 8) Dillingham TR : Approach to Trauma of Periph- always complete recovery, and those with ＜ eral Nerves. 1998 AAEM Course C : Electro- diagnosis in Traumatic Conditions―Course Hand- 10% have a poor outcome. out. AAEM, Rochester, MN, 1998; pp 7-12 Complete axonotmesis and neurotmesis have

Jpn J Rehabil Med VOL.42 NO.2 2005 Traumatic Injury to Peripheral Nerves

9) Dorfman LJ : Quantitative clinical electro- 27) Miller RG : AAEE Minimonograph #28: Injury to

physiology in the evaluation of nerve injury and Peripheral Motor Nerves. Muscle Nerve 1987; 10:

regeneration. Muscle Nerve 1990; 13: 822-828 698-710 10) Dumitru D : Electrodiagnostic Medicine.Hanley& 28) Milner-Brown HS, Stein RB, Lee RG : Synchrnon- Belfus, Philadelphia PA, 1995; pp 341-384 ization of human motor units: Possible role of

11) Erminio F, Buchthal F, Rosenfalck P : Motor unit exercise and supraspinal reflexes.EEG Clin Neuro-

territory and muscle fiber concentration in paresis physiol 1975; 38: 245-254

due to peripheral nerve injury and anterior horn 29) Moritani T, de Vries HK : Neural factors versus

cell involvement. Neurology 1959; 9 : 657-671 hypertrophy in the time course of muscle strength

12) Fisher MA : Minimonograph #13: H reflex and F gain. Am J Phys Med 1979; 58: 115-130

wave: physiology and clinical indications. Muscle 30) Noble J, Munro, CA, Prasad VSSV, Midha R :

Nerve 1992; 15: 1223-1233 Analysis of upper and lower extremity peripheral

13) Fowler TJ, Danta G, Gilliatt RW : Recovery of nerve injuries in a population of patients with

nerve conduction after a pneumatic tourniquet : multiple injuries. J Trauma 1998; 45: 116-122 observations on the hind-limb of the baboon. J 31) Ochoa J, Fowler TJ, Gilliatt RW : Anatomical

Neurol Neurosurg Psychiatry 1972; 35: 638-647 changes in peripheral nerves compressed by a pneu- 14) Garland DE,Bailey S : Undetected injuries in head- matic tourniquet. J Anat 1972; 113: 433-455 injured adults. Clin Orthop Relat Res 1981; 155: 32) Partanen JV, Danner R : Fibrillation potentials

162-165 after muscle injury in humans.Muscle Nerve1982; 15) Gilliatt RW, Hjorth RJ : Nerve conduction during 5 (9S): S 70-S 73

Wallerian degeneration in the baboon. J Neurol 33) Rasminsky M,Sears TA : Internodal conduction in

Neurosurg Psychiatry 1972; 35: 335-341 undissected demyelinated nerve fibres. J Physiol 16) Gilliatt RW,Taylor JC : Electrical changes follow- 1972; 227: 323-350 ing section of the facial nerve. Proc R Soc Med 34) Seddon HJ : Nerve grafting. J Bone Joint Surg 1959; 52: 1080-1083 ［Br］1963; 45: 447-455 17) Gilliatt RW, Westgaard RH, Williams IR : Extra- 35) Seddon HJ : Surgical Disorders of the Peripheral

junctional acetylcholine sensitivity of inactive Nerves. 2nd Ed, Churchill Livingstone, New York,

muscle fibres in the baboon during prolonged nerve 1975; pp21-23

pressure block. J Physiol 1978; 280: 499-514 36) Selecki BR, Ring IT, Simpson DaA, Vanderfield

18) Gilliatt RW : Acute compression block. in The GK,Sewell MF.Trauma to the central and periph-

Physiology of Peripheral Nerve Disease (ed by eral nervous systems: Part II, a statistical profile

Sumner AJ). W.B. Saunders, Philadelphia, 1980; of surgical treatment in New South Wales 1977. pp 287-315 Aust NZJ Surg 1982; 52: 111-116 19) Guth L : Neuromuscular function after regenera- 37) Sillman JS,Niparko JK,Lee SS,Kileny PR : Prog- tion of interrupted nerve fibers into partially dener- nostic value of evoked and standard electromyogra- vated muscle. Exp Neurol 1962; 6: 129-141 phy in acute facial paralysis. Otolaryngol Head

20) Haymaker W, Woodhall B : Peripheral Nerve Neck Surg 1992; 107: 377-381

Injuries. W.B. Saunders, Philadelphia PA, 1953 38) Speidel CC : Studies of living nerves: growth

21) Hoffman H : Local reinnervation in partially adjustments of cutaneous terminal arborization. J

denervated muscle: a histophysiological study. Comp Neurol 1942; 76: 57-73 Aust J Exp Biol Med Sci 1950; 28: 383 39) Stone L, Keenan MA. Peripheral nerve injuries in

22) Kimura J, Machida M, Ishida T, et al: Relation the adult with traumatic brain injury. Clin Orthop

between size of compound sensory or muscle action Relat Res 1988; 233: 136-144

potential, and length of nerve segment. Neurology 40) Sunderland S : Nerves and Nerve Injuries.2nd Ed, 1986; 36: 647-652 Churchill Livingstone,New York,1978; pp133-138 23) Kline DG : Surgical repair of peripheral nerve 41) Tackman W,Radu EW : Observations of the appli-

injury. Muscle Nerve 1990; 13: 843-852 cation of electrophysiological methods in the diag- 24) Kraft GH : Fibrillation amplitude and muscle atro- nosis of cervical root compressions. Eur Neurol phy following peripheral nerve injury. Muscle 1983; 22: 397-404

Nerve 1990; 13: 814-821 42) ThesleffS : Physiological effects of denervation of

25) MacKinnon SE,Dellon AL : Surgery of the Periph- muscle. Ann NY Acad Sci 1974; 228: 89-103 eral Nerve. Thieme Medical Publishers Inc., New 43) Trojaborg W : Early electrophysiological changes

York, 1988 in conduction block. Muscle Nerve 1978; 1: 400- 26) Massey JM,Sanders DB : Single-fiber EMG demon- 403 strates reinnervation dynamics after nerve injury. 44) Weddell G,Glees P : The early stages in the degen- Neurology 1991; 41: 1150-1151 eration of cutaneous nerve fibers.J Anat 1941; 76:

リハビリテーション医学 VOL.42 NO.2 2005年 2月 Lawrence R ROBINSON

65-93 47) Wood MB : Surgical Approach to Peripheral Ner- 45) Wertsch JJ, Oswald TA, Roberts MM : Role of vous System Trauma. 1998 AAEM Course C :

intraneural topography in diagnosis and localiza- Electrodiagnosis in Traumatic Conditions?Course

tion in electrodiagnostic medicine. Phys Med Re- Handout. AAEM, Rochester, MN, 1998; pp 27-36 habil Clin N Am 1994; 5: 465-475 48) Yuska MA, Wilbourn AJ : Incidence of fibrillation

46) Wertsch JJ, Sanger JR, Matloub HS : Pseudo- potentials in “pure”conduction block mononeuro- anterior interosseous nerve syndrome. Muscle pathies. Muscle Nerve 1998; 21: 1572(abs)

Nerve 1985; 8: 69

Jpn J Rehabil Med VOL.42 NO.2 2005