surgery of peripheral nerve injury · knee or tearing of the roots of the brachial plexus in a...

$: SURGERY OF PERIPHERAL NERVE INJURY · knee or tearing of the roots of the brachial plexus in a birth injury. They may also complicate fracture or beproducedbylong-term irritation,$
THE

SURGERY OF PERIPHERAL NERVE INJURY

ByIAN AIRD, CH.M., F.R.C.S.(ED.)

(From the Department of Surgery, University of Edinburgh.)Professor of Surgery Elect, Post-Graduate Medical School, Hammersmith.

IntroductionPeripheral nerves may be injured by open

wounds (by bullets or other projectiles or by sharpweapons) or they may occur without skin injury.The closed injuries of the peripheral nerves may bedue to sudden direct violence, as in the case ofcontusion of the external popliteal nerve againstthe neck of the fibula, compression of the radialnerve by tourniquet, and damage to a nerve bypressure from an operating table, from a crutch, orfrom the arm of a chair; or they may be producedby injuries of a stretching kind, as in tearing of theexternal popliteal nerve in forcible abduction of theknee or tearing of the roots of the brachial plexusin a birth injury. They may also complicatefracture or be produced by long-term irritation, asin the ulnar neuritis due to a valgus deformity atthe elbow, the paralysis of the lowest fibres of thebrachial plexus due to pressure by a cervical rib,and the paralysis of the dorsal interosseous nerveby fascial pressure at the elbow.

Recently, interest has focussed on nerve injuriescomplicating war wounds, and experience of thesehas extended substantially knowledge of the pro-cesses of nerve, injury and nerve repair. Thephenomena of nerve regeneration after injury aremuch better understood now than formerly, anddetails of diagnosis and of operative techniquehave been improved. Even so, it would not beunfair to say that operative technique has made nofundamental advance during the recent war. Ourclinical knowledge of nerve injuries has increasedby inconsiderable accretions scattered over thewhole clinical field, and it would seem profitable toreview this against the whole background of ourknowledge of peripheral nerve injuries.The most important principle of the care of peri-

pheral nerve injuries in the second world war wasthat of the isolation of patients in specialisedcentres, where the surgeon can co-operate not onlywith the orthopaedic expert, the neurologist, andthe radiologist, but with the physiologist, thepathologist, the bacteriologist, the psychiatrist,and the expert in physical medicine. The patientsrequire repeated study by this team of specialists,

and constant opportunity for physio-therapy andoccupational therapy, before their final return toindustry. Even after surgical treatment ends,long-term follow-up is esseftial, and opportunitiesfor vocational training and guidance in choice of acareer are not the least important duties of thecommunity to men and women who have sustainedperipheral nerve injuries.

The Clinical Effects of Nerve InjuryNerve injury may produce disturbances of

motor, sensory, sudomotor, vasomotor, nutri-tional, and reflex function in the territory of theaffected nerve.

(a) Motor Effectsi. Loss of voluntary contractionWhen a nerve is completely divided, the muscles

supplied by it are denervated, and no longer con-tract under volition. The paralysis of the musclessupplied by the injured nerve may, however, beobscured by tricks, synergia, and the employmentof subsidiary muscles, and by anomalies of inner-vation. Notable examples of trick movement are:abduction of the fingers by the extensor muscles inulnar palsy; flexion of the knee by gracilis in sciaticpalsy; extension of the thumb by abductor pollicisbrevis in radial paralysis; extension of the inter-phalangeal joints by the interossei and lumbricals,and adduction of the fingers by their flexors inulnar paralysis. Notable examples of the use ofsubsidiary muscles are: flexion at the elbow by theflexors of the forearm when biceps and brachialisare paralysed, and flexion of the wrist by abductorpollicis longus in combined median and ulnarpalsy. An example of the trick effect of synergiais the extension of the wrist which may occur whena fist is made, even in complete extensor paralysis(Fig. i).Anomalous innervation of muscles is another

cause of failure to realise the completeness of alesion, as when thenar muscles with anomalousinnervation continue to contract voluntarily afterdivision of the median nerve. Highet's procaine

**

Protected by copyright.

on February 9, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.22.251.225 on 1 S

eptember 1946. D

ownloaded from

http://pmj.bmj.com/

POST-GRADUATE MEDICAL JOURNALblock method48 is invaluable for the detection ofskilful trick movements, and of anomalous inner-vation. A muscle which continues to act afterprocaine block of the neighbouring uninjurednerve suspected of innervating it anomalously,may be presumed to be innervated by undividedfibres of the injured nerve under study.A further difficulty in estimating the extent of

motor loss in a war injury arises in cases of multiplewounds of an extremity. If, in addition to awound of a peripheral nerve, muscles are widelydamaged or tendons divided, the share of responsi-bility to be allotted to nerve damage on the onehand, muscle or tendon loss on the other, maysometimes be apportioned only with great diffi-culty. The visible and palpable contraction of amuscle belly should be studied independently of itskinetic effect.

2. AtrophyA few months after injury the motor end-plates

disappear, and, as paralysis continues, the para-lysed muscles atrophy; their fibres shrink, lose theircross-striation, and ultimately disappear, to becompletely replaced by fibro-fatty tissue some-times as early as three years after injury, but, asYoung has shown in a case of Learmonth'sI33,sometimes as late as 26 years.3. Faradism

Certain electrical changes8 occur in the denervatedmuscle. The simplest and the oldest method ofstudying the electrical behaviour of a muscle is todetermine its response to faradism. A normalmuscle responds to faradism by contraction, andresponds also, but less readily, to galvanism. Thecontraction produced by application of the cathodeto the point of entry (motor point) of the nerve ofsupply .to a muscle is normally greater than thecontraction produced by the anode applied to thesame point (K.C.C. > A.C.C.). One or two weeksafter nerve injury, the nerve and muscle cease torespond to faradism, but for ten days thereaftergalvanism continues to give a normal responsebefore it becomes sluggish, vermicular and linger-ing. In this period an increasingly strong currentis required to produce contraction, and the A.C.C.may equal or be greater than the K.C.C. (polarreversal). These changes in the electrical responseof a denervated muscle are together known as theReaction of Degeneration (R.D.). The most signi-ficant change is the failure to contract to faradism,and the least significant is polar reversal. TheR.D. is said to be complete when faradism no longerelicits contraction, and when polar reversal hasoccurred; it is said to be partial when faradism ispresent, but when a stronger current is requiredthan that required for the neighbouring muscle of

the opposite side. A curious electrical reaction,the "paradoxical electrical reaction of ERB," issometimes seen if a nerve has been compressed fora time: faradic excitation may be induced in asegment distal to a lesion which blocks conduc-tion; stimulation above the block fails to elicitcontraction, while stimulation below the block iseffective.

4. ChronaximetryThe chronaxie of a muscle also alters after nerve

division, increasing as the muscle degenerates. Thechronaxie is the minimum time required to producea contraction by a voltage equal to twice thatrequired to produce a minimal response. In short,the chronaxie is twice the rheobase. Muscles varyin chronaxie from o o8 to 0 7 milliseconds, but thechronaxie is constant for muscles co-operating in agiven movement. Chronaximetry8 has been em-ployed to decide whether the fibres of nerve testedare intact or degenerated, and it has some signi-ficance in this respect, but its value is limited sincethe method is exact only if all the fibres of thenerve tested are either intact or degenerated;normal fibres are not detected if mixed with de-generated fibres.8 ShapiroIoo records a lengthen-ing of chronaxie in the antagonists of the dener-vated muscles.

5. Electro-myographyMuch more valuable than chronaximetry is

electro-myography. Adrian2 first recorded thespontaneous electrical activity of muscles measuredby electrodes placed on the skin over them orinserted directly into them, and the method hasbeen employed by Denny-Brown and Penny-backer22 and others.45 For some time after injury,voluntary muscle is electrically inactive; that is,no action currents can be detected when a concen-tric needle electrode connected with a cathode rayoscilloscope is inserted into the muscle.I2 (Agalvanic current may be sent down to the needle-electrode after its insertion to determine in whichmuscle it lies.) Twelve to twenty-eight days later,fibrillation makes its appearance-irregular finespikes of change in potential, totally unrelated toattempted voluntary contraction; these variationsin potential persist for as long as any contractiletissue remains, or until reinnervation occurs.Even if spontaneous fibrillation is lacking in a de-nervated muscle, it may be induced by heat.With the onset of fibrillation, the denervated

muscle becomes increasingly sensitive to acetyl-choline, and Denny-Brown and Pennybacker22have suggested that fibrillations might be due tocirculating acetyl-choline. Eccles25 found, how-ever, that curare lessened the sensitivity of dener-

226 September, I946P

rotected by copyright. on F

ebruary 9, 2020 by guest.http://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

SURGERY OF PERIPHERAL NERVE INJURYvated muscle to acetyl-choline without affectingfibrillation.

Langley believed that the fatigue of fibrillationwas responsible for the atrophy of denervatedmuscle, but, Solandt and Magladeryloz abolishedfibrillation by quinine and quinidine, withoutpreventing atrophy.

Fibrillation ceases in denervated muscle with thedisappearance of cross-striation, and the loss ofcontractility.6. Strength-duration curve

Ritchie87 has for preference employed thestrength-duration curve for the study of the inner-vation of muscle. A high strength of current isemployed for an exceedingly short duration andthe voltage is plotted against the duration of thecurrent. The resultant curve is the strength-duration curve. The height of the curve variesfrom time to time in normal muscle, but the shapeof the curve remains fairly constant. Denervatedmuscle gives a rapidly rising curve, entirely dif-ferent from that given by normal muscle, and theform approaches normal again as reinnervationproceeds. In electro-myography and in Ritchie'sstrength-duration curve method, a single observa-tion gives little valuable information, but repeatedobservations showing a progressive change inelectrical activity sometimes allow conclusions tobe drawn from them.

7. Muscle biopsyMuscle biopsy is of clinical value in furnishing

information of the histological state of denervatedand/or ischaemic muscle."'I3, I4 Further, sincenerve fibres are found in more than nine-tenths ofthe muscle sections studied, muscle biopsy is aform of nerve biopsy.I4 Soon after denervationthe pattern of muscle fibres becomes more open;the nuclei appear to increase in number and to begrouped together in clumps; and the cross-striationbegins to fade. The Schwann tubes, empty oftheir axons, can still be traced to the motor end-plates, which usually remain distinguishable forabout nine months after injury.

(b) Sensory Effectsi. AnaesthesiaImmediately after nerve division, there is loss to

light touch over a well-defined area, loss to pin-prick over a smaller and less well-defined area, andloss of pressure sensibility. Sensory no less thanmotor distribution of any given nerve varies frompatient to patient (Figs. 2 and 3), and may evendiffer on the two sides of the same individual. Thisis due sometimes to partial replacement of onenerve by another, sometimes to anastomosis be-

tween two nerve trunks. Anomalies of sensory,like those of motor, distribution may be detectedby Highet's procaine method.48 The clinicalmeasurement of sensory loss is less exact than theestimate of loss of motor function. 8 Pure touchis difficult to produce, for even von Frey's hairs,algometers, and copper rods maintained at aconstant temperature produce some deformity ofthe skin.85 BishopTo described an electrical appa-ratus which will stimulate single sensory spots onthe skin by high-voltage low-current spark dis-charge, but his is regarded as a laboratory ratherthan a clinical method.Each primary modality of sensation-pain,

touch, cold, warmth, and pressure-is subservedby a sensory receptor which. is specific for thatmodality and each receptor is connected with aspecific size of myelinated fibre together with anaccessory unmyelinated fibre which acts as a sort of"burglar alarm" by evoking the sensation of painwhen the stimulus reaches a level which may beharmful to the organism.o20 In clinical examina-tion of a peripheral nerve injury each modalityshould be separately studied, and its area of lossmapped out. The areas of total loss of the variousmodalities differ, their outlines forming concentricrings except in lesions of the brachial plexus. Painloss, for example, is less than touch loss in lesionsof nerve trunks, but greater in root lesions. Coldis intermediate, and warmth is the largest of all.The area of sensory loss varies with the nerve andwith the lesion, but large and small areas of anaes-thesia may be found after both high and lowdivision of a nerve.85

2. PainPain referred to the area of cutaneous distribu-

tion of a damaged nerve is much more common inpartial than in complete lesions; in the latter itseldom persists for more than three weeks. It maybe of any degree-trifling discomfort, minorcausalgia (often referable to fracture, foreign body,soft tissue damage, or wound infection) or truecausalgia.3. Overlap of dermatomes

After division of a peripheral nerve some re-covery of sensation occurs even without regenera-tion, as a manifestation of the normal overlap ofdermatomes. 80 Each nerve has an autonomousarea whose sensation it wholly and exclusivelysubserves, but around the autonomous area is anintermediate zone in which sensory function isshared with the nerve of the adjacent territory.Immediately after a nerve injury, this intermediatezone may be or appear to be anaesthetised, butafter a few days nerve fibres from adjacent nervesmay take over the sensory function of the inter-

Se.Ptember, I946 227P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

POST-GRADUATE .MEDICAL JOURNALmediate zone, and the area of sensory loss mayappear to shrink. Later still, and up to a tirqelimit of about fourteen days, new fibres fromadjacent nerves seem actually to grow into theautonomous from the intermediate zone,8I, 8, I03,123 and the area of sensory loss shrinks still further.This is probably a manifestation of the continuousdegeneration and regeneration that takes placenormally in the peripheral nerve plexus of theskin.121 The ingrowth is slower into the territoryof a nerve which has sustained a gunshot woundthan into that of a nerve which has been cleanlydivided.29 It explains vaunted cases of "completerecovery of sensation" a few days after suture ofthe median nerve at the wrist-an unusually smallmedian territory having been reinnervated fromthe intermediate zone around it.

(c) Sudomotor EffectsWhile, in partial lesions, sweating is often in-

creased, the hyperhidrosis sometimes extendingbeyond the area of sensory loss, after a completeinjury sweating ceases in the denervated part.Anhidrosis is perhaps the best index, since themost objective index, of denervation of skin.Usually anhidrosis covers a larger area tian doesanaesthesia.

Several clinical tests have been devised for deter-mining precisely and for recording the area ofanhidrosis. Perhaps the most valuable of these isGuttmann's Quinizarin Test.42 The patient isgiven 5 to Io grains of aspirin and a cup of coffeeto induce general perspiration, and the skin of theaffected part is then dusted with Guttmann'sPowder. This is a mixture of 28 gm. of sodiumquinizarin 2-6 disulphonate, 24 gm. of anhidroussodium bicarbonate and 48 gm. of rice starch (thestarch serves as a base and the bicarbonate pre-vents moistening during storage). The arm isplaced beneath a hot-air cradle or in a radiant-heatchamber, and the powder turns a deep reddish-purple if sweating occurs (Fig. 4). The sweat mustnot be allowed to run from a perspiring area to anarea of anhidrosis. Silverman and PowellIoI usetincture of ferric chloride diluted with three partsof alcohol, painted on the skin with a cotton-tippedapplicator; the skin is allowed to dry and tannicacid powder is then spread over the whole area.Sweating skin blackens first in tiny dots and thendiffusely from coalescence of the dots. The blackstain can be removed. later with five per centsolution of oxalic acid.

Closely linked with the phenomenon of anhi-drosis is that of skin resistance to an electric cur-rent. Richter and Katz86 showed that whensweating is normal skin resistance is low, whileanhidrotic skin has a high resistance. If an in-

different electrode is attached to the ear and anexamining electrode is moved over normal skinuntil there is no fluctuation of a connected am-meter, and if the examining electrode is themmoved to a suspected area, the ammeter swingssharply as the skin resistance rises and the area ofhigh resistance can be marked out.56

(d) Vasomotor EffectsRichards84 has described in detail the vasomotor

changes which occur in denervated skin. They areparticularly marked in median, ulnar and sciaticlesions, because in these nerves vasomotor fibresare particularly numerous. The first change is oneof vaso-dilatation from paralysis of the vaso-constrictor fibres, and the denervated skin is redand warm as it is after sympathectomy. Afterabout three weeks, the skin of the denervated areabecomes blue and cooler, approximates morenearly in temperature to its environment, andbecomes but little warmer when another part ofthe body is heated. The second stage, of coolcyanosis, appears to be due to loss of the afferentlimb of the vasomotor reflex84, perhaps intensifiedby immobilisation. After a complete lesion of aperipheral nerve, sympathetic block fails to give arise of skin temperature in the affected area.77

(e) Effects upon ReflexesEvery reflex, superficial or deep, is lost, any part

of whose arc crosses the point of injury on theaffected nerve.

(f) Nutritional ChangesNutritional changes46 may occur in any tissues

supplied by denervated nerve, but they are un-common in any but irritative and painful lesions.The skin loses its pits and wrinkles and becomespreternaturally smooth, thin, glossy, mottled,translucent and inelastic. Normal resistance totrauma is lost and indolent ulcers develop, especi-ally in denervated fingers, toes, sole or heel. Thereis loss of subcutaneous tissue and the tips ofanaesthetic digits may. shrink and taper. Depila-tion is the rule in denervated areas, but increasedrate of growth of hair may occur in irrita-tive lesions. The nails become distorted-brittle,ridged or curved. Hypertrophic subcutaneousfibrosis is uncommon, but by it the skin may inirritative lesions be thickened and thrown intofolds, and clubbing of the fingertips has sometimesbeen observed.

Muscles, as we have seen, atrophy progressivelyand ultimately disappear. Muscles and theirtendons shorten and deformities are rendered per-manent by contraction of fasciae and joint capsulesalso. Contracture and ankylosis of joints develop




j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

SURGERY OF PERIPHERAL NERVE INJURYmore readily in painful than in painless lesions, andare more common in distal than in proximal nerveinjuries.46 It is unlikely that these lesions are duesolely to immobilisation, for they are selective intheir incidence-in tibial nerve lesions the joinis ofthe foot suffer most often, while in commonperoneal lesions the ankle is the commonest seat ofstiffness. In median injuries the interphalangealand metacarpo-phalangeal joints suffer, and thereis often a partial metacarpo-phalangeal ankylosisof the thumb which permits movement only in theplane of the palm. The risk of contracture andankvlosis increases with age and with virulence orchronicity of associated sepsis. German andRussian authors have theorised lengthily aboutcontractures and explain them as a reflex pheno-menon, but their evidence of a reflex effect is notconvincing.The bones may undergo a progressive decalcifi-

cation which, too, is more rapid and severe inpartial and painful lesions. The decalcification ismost intense in the proximal and distal extremitiesof the phalanges, and the radiological appearancesresemble those of the post-traumatic osteo-porosisof Sudek. Median nerve lesions, particularlythose associated with causalgia, are most apt toproduce osteoporosis. The increased blood supplyof the bone incidental to the lessened demands ofwasting muscle has been blamed for this decalcifi-cation,46 but the voluntary immobility of thecausalgic hand certainly aggravates it. It is ofinterest that osteo-porosis, like causalgia, is morefrequent in injuries of the median than in lesionsof any other nerve.

The Clinical Signs of Recovery(a) Rate of ReinnervationWhen an injured nerve recovers, either spon-

taneously or as a result of suture, the axon tipsproceed in regeneration at fairly standard rates.There is first a latent period of about seven days,followed by a progressive advance of 0*25 milli-metres per day through the injured part, and thenof 3 or 4 millimetres a day in the distal stump ofthe nerve;37 if the nerve has been crushed only andits perineurium has not been divided, the latentperiod is shorter and the rate of advance is morerapid. Finally the growing axons reach their end-plates, and some time passes before connectionis established. Meantime, the axons have becomemedullated centrifugally at a rate which progres-sively slows as the periphery of the injured nerveis reached. There are thus really three temporalstages in reinnervation: an initial delay of threeweeks or more (dependent on the form of interrup-tion) at the lesion, the time of travel down thenerve at the rate of 3 or 4 millimetres a day, and

the period between the arrival of the fibres at thestructure they innervate and the return of function,and the date of expected recovery in any musclecan be roughly calculated. The wave front offunctional maturation as it proceeds down a limbfollows not a straight line but a parabola, rapid atfirst and slow as the distal extremity is reached.98The distal retardation may be due to the pro-gressive slowing of myelinisation as regenerationproceeds further and further from the parentstump on which it depends, and partly also to thegreater degree of atrophy of end organs and ofmuscles in the more distal parts of a denervatedextremity.

(b) Progress of Motor RecoveryThis can be demonstrated in a muscle before

voluntary contraction can be elicited in it, and theintroduction of more delicate electric methodsduring the recent war for the measurement ofmuscle activity has aimed at detection of reinner-vation before actual contraction can be induced.The first sign of reinnervation is a change in theelectro-myograph. As long as three months beforethe appearance of voluntary movement, fibrillationmay cease in the reinnervated muscle; large actionpotential variations are then observed, charac-teristic of motor unit activity.Iz They appear onthe screen on attempted voluntary contraction,and they are a sign that some return of voluntarypower can be expected in due course. No reliancecan be placed on an isolated observation, for thefinding of an odd patch of apparent reinnervationmeans only that a few motor fibres have arrivedand does not tell what is happening in the nerve asa whole. Repeated observations at short intervalsare therefore necessary.Three to nine weeks before the first appearance of

voluntary power, a change may be observed in thestrength-duration curve.87 This cannot be elicitedas early, as a rule, as the change in the electro-myograph, but an alteration in the shape of theS.D. curve when it is obtained is perhaps a morereliable index of reinnervation. Here again, atleast two observations are required to detect areturn to the normal shape of curve.A few weeks, or perhaps monthsI4 before the first

clinical signs of recovery, muscle biopsy will showthe presence of axons in the Schwann tubes ofintramuscular nerve fibres, if regeneration hasproceeded so far. If myelination is far advanced,the nerve fibres of regenerating nerve may differfrom normal fibres only in branching more freelyand ending at a distance from the end-plates. Ifno axons are visible when it is reasonable to expectthem there must either be a division of the nerveor other serious obstacle to regeneration.I4

229September, I946P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

POST-GRADUATE MEDICAL JOURNALFinally, voluntary movement returns with a lower-

ing of chronaxie, and the normal response to faradicstimulation. It is of interest that voluntary con-traction can frequently be detected before a faradicresponse can be obtained, and the use of electricalmethods, therefore, in no way absolves the clinicianfrom careful and repeated study of voluntarypower in the denervated muscles; muscle testingshould be performed at intervals of a week or afortnight before the earliest date of expected re-covery; the patient must be warm, alone with hisexaminer, fresh, and free from fatigue.

Highet,68 for convenience of recording, recog-nised six stages in the voluntary recovery ofmuscle:-

(o) No contraction;(I) Return of perceptible contraction in proximal

muscles;(2) Perceptible contraction in both proximal and distal

muscles;(3) All important proximal and distal muscles act

against resistance;(4) Synergic and isolated movements added to previous

recovery, and(5) Complete recovery.

(c) Progress of Sensory RecoveryWe have already mentioned that sensation may

be recovered early in the periphery of the dener-vated zone by an assumption of function in theintermediate area of two adjacent nerves by fibresfrom the area served by the undivided neighbour8sand that later, up to fourteen days after injury,new nerve fibres may actually grow in to theautonomous zone of the denervated area from theterminal fibres of adjacent nerves.122 Later, andpari passu with motor recovery, regenerationoccurs by the downgrowth of nerve fibres along thetrunk of the injured nerve. In the case of largenerve territories in the extremities, sensory re-covery begins proximally and spreads centri-fugally, but so far as nerves with a circumscribedproximal distribution such as the axillary are con-cerned, the zone of sensory loss shrinks concen-trically, for axons entering such an area from themain trunk reach it first at its periphery.I23 Thereinnervation of a whole thickness skin graftproceeds centripetally in the same way.As sensation returns, it presents three charac-

teristics, which were described by Trotter andDaviesI6, I"I7:-

(I) A higher threshold of stimulation is required to elicitsensation. If constant intensity of stimulation isemployed the first signs of recovery may be missed.This accounts for the common observation that painreturns before touch. Pain has normally a variablethreshold and pinprick is a variable stimulus, whilea wisp of wool is a constant stimulus which may failto elicit early signs of recovery. If von Frey hairs

are used, the intensity of stimulus can be steppedup, and recovery of touch is found to coincide withrecovery of pain, or to follow it very closely. It isinteresting to know that the higher threshold ofstimulation required is not due to a lessened conduc-tivity of the nerve; the excitability of growing nerveis increased rather than diminished.

(2) Sensation is intensified. The specific sensation,when it returns, is unusually vivid. This is especi-ally true as regards pain and cold. Touch does notlend itself to intensification, and warmth, whenintensified, tends to be felt as pain.

(3) Sensation is referred peripherally. In the earlystages of recovery when only the proximal part ofthe denervated area has regained sensation, astimulus applied to the denervated part is referredtowards or to its distal extremity. If the stimulusis painful, the patient may rub the distal part, andeven, as Richards8 has observed, obtain relief fromthe rubbing. Peripheral reference is easily shown inthe case of touch, cold and pain, but is difficult toelicit with warmth. As recovery proceeds towardsthe distal parts of the denervated area, sensationmay be referred proximally (Trotter and Davies,z 6)even to a point outside the denervated area alto-gether, such as the site of nerve suture. Richards8shas further demonstrated that peripheral referencemay be accompanied or followed by "cross-refer-ence"; in a recovering median lesion, a stimulusapplied to the index finger may be referred tothe middle finger or the thumb. Intensificationand reference are the most constant and persistentsigns of regeneration after suture, and may bepresent for many years even after acuity has reacheda high standard (Richards).

Pain, subserved by undifferentiated nerve endsof the "free-beaded" type, and by relativelyprimitive fibres (the unmyelinated and fine-myelinated types C and B of Fulton32) has always abetter chance of sensory recovery than the othermodalities, and may be the only form of sensationrecovered.

After a nerve injury of the type which hasleft neurilemma and Schwann cell tubes intact,recovery should approach Ioo per cent, asSchafer92 proved in his own little finger. Afternerve section and suture, recovery is alwaysimperfect, for after suture the chances that anaxon reaches its own Schwann-cell tube are remoteand the most that can be expected is that the axonwill find its way into a tube which will lead it to areceptor similar to that which it innervatedpreviously36; final grades of touch and thermalsensation, accurate localisation and two-point dis-crimination and stereognosis are never regained.Similarly, Richards found that intensification andreference were seldom elicited unless the Schwanntubes had been damaged at the point of injury,and he believes that reference occurs only whenthere has been some crossing of axons. 85 Sperryo04by crossing sensory nerves in the experimentalanimal has shown that localisation cannot beperfectly relearnt.For reasons which have already been given,

Se,btember, I946230P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

SURGERY OF PERIPHERAL NERVE INJURY IAN AIRD, CH.M., F.R.C.S.(ED.)

Cli·

:: e ' ?.1t . ;: ^,3Bi i Esa l.-...... ......................

??..... ......,......... ::S,#: *l--*..........................: ..

....ii^:.!:: :

...'l.:'!giii'~ l~~IIIIIF iiP::ibi- ,:,:·:: !l:;i::ji: .i ... .i·::·.

...

.: ::·:::::::^.is~·::::" :.

iweHat*::ii.·:ii·ii·

FIG. I.-Dorsiflexion of wrist when fingers are flexed, despite complete extensor paralysis.An example of the trick effect of synergia.

....

'" '..........

E."''::'"':'::rat -

FIG. 2.-Complete ulnar nerve lesion. Dorsum of palm showing unusually large sensory loss.



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/


:.:

*erllrrqMC

FIG. 3.-Complete ulnar nerve lesion. Dorsum and palm showing unusually large sensory loss.

i'.i~.:l~iiiilili~ii~iiii.:iliii:.:iiii i:..i~i.'iiii... .. .......:. .:..... ..... ..... .. ...... ..ii~i~iiih;s~iiii:'·i. I::·iiii.:-ili'i'"iP1 .......~ :.. i~%; ~iie~iaiii.···'laiiiii iif~iilri~r~........... .........·

illp~~iiezi Irm:".,:::;",

:··: .'.'ii...................::

:c.;i~:~Y~~~................. ....i:::·~*~:·:: ..............

:iiii:;ii :·· ii iiliiil:·si!: ....... ........ili~iiiiii:~"N' 3i '':

lill~iiill I?:"::::iii'i ·I :::: iir~iiiil:'·"'iii;;iilllriili :.HI~ii ;i~igiiii ii~: ii· ..:;;:::::··. ........ ...Ilii

En.? -'~zi -:·i~-11' _n:: ;"' MniUi :i :i

... ..

..i'i.riiliii iiiii lni iii:diill E::

:: '"'::"

..... .....~iiiil ::· ii...i;:::··:.

FIG. 4.-Guttman's sweating test. Anhidrosis of right palm in complete median lesion. Note wasting ofthenar eminence, and simian position of thumb, which is rotated so that its palmar surfaceis almost in the same plane as that of the palm of the hand.



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/

SURGERY OF PERIPH]

sensory recovery, particularly in the periphery ofa denervated area where it marches with an areawhich has retained normal sensation, is not alwaysgood evidence of regeneration,I6 and there may bea further curious and complicating factor. Nervefibres, small and usually non-medullated, maygrow from blood vessels into the peripheral stumpsof nerves for a sufficient length to produce a reflexresponse to stimulation and to give some sensoryeffect if the distal trunk of the damaged nerve isstimulated electrically at an operation performedunder local anaesthesia.55For the recording of sensory recovery Highet68

suggested five stages:-(o) Absence of sensibility in autonomous zone;(I) Recovery of deep cutaneous pain sensibility within

the autonomous zone;(2) Return of some degree of superficial cutaneous pain

and touch within the autonomous zone;(3) Return of superficial cutaneous pain and touch

throughout the autonomous zone with the disap-pearance of intensification;

(4) As for stage (3), plus recovery of two-point dis-crimination within the autonomous zone.

Tinel's Sign,II', I3 "le Signe du Fourmillement,"may conveniently be discussed here. Introducedas a sign of nerve regeneration in 1917, the distaltingling produced by tapping the injured nerve hasnow fallen into discredit as evidence of the progressof the growing axon. In many cases, it is true thatif the line of a nerve is tapped at different levelsuntil a point is found where a tingling sensation isreferred distally from the point of tapping the signwill be elicited progressively more distally alongthe regenerating nerve. Most workers during therecent war, however, have found again and againthat patients with Tinel's phenomenon, even whenit was carefully and conclusively elicited by tappingthe nerve trunk disto-proximally, were found atoperation to have widely separated nerve endswithout the slightest possibility of regeneration(Coleman,I8 Seddon,94 Pollock,82 and Cairns andYoungI6). Pollock and Davies failed to elicitTinel's sign in seven of fifty recovering nerves, andthey found it entirely absent in only seven out offifty complete lesions. Even an "advancingTinel's sign" may be found with complete ana-tomical division. Of recent writers, Nathan andRennie74 alone find any merit in Tinel's sign asevidence of regeneration, but even they regard anegative Tinel's sign as of no value, and accepteven an advancing sign with some reservation.

TYPES OF NERVE INJURYNerve injuries can be conveniently divided into the

following six types, based on the classification ofSeddon97:-

(I) Complete division (the complete neurotmesis ofSeddon), when the nerve is divided across in itswhole thickness;

(2) Incomplete division (partial neurotmesis), when the

September, 1946 ERAL NERVE INJURY 233nerve is partly severed, some portion remainingundivided;

(3) Axonotmesis (Seddon), when certain axis cylindersare disrupted within the nerve, without disturbanceeither of the Schwann tubes or of the neurilemma;

(4) Neurapraxia (Seddon), when nerve function is lostfrom traumatic demyelination of the larger fibres,without other damage to axons or to their cover-ings ;2 it may be partial or complete;

(5) Ischaemia of a segment of nerve or of segments ofmore than one nerve;

(6) Combinations of two or more of the last four of these.

"From a single examination it is absolutelyimpossible to differentiate a complete division(neurotmesis) from a complete lesion in continuity(axdnotmesis or neurapraxia). Moreover, one can-not predict whether the nerve will recover spon-taneouslyorrequire surgical treatment." (Pollock. 83)Nevertheless, a provisional diagnosis should beattempted when the patient is first seen, with thereservation that it may require correction ormodification later.

(I) It is simpler to exclude complete division(complete neurotmesis) from the diagnosis than toassert confidently that it is present. It is notpresent if some part of the affected nerve's functioncan be conclusively shown to be present. Con-versely, in complete division (complete neurot-mesis) there is immediate and total sensory, motor,sudomotor, and vasomotor paralysis over thewhole area of the affected nerve, but the sametotal loss of function attends complete axonotmesisand complete neurapraxia,.and unless the natureof the injury gives some immediate hint these canbe differentiated from neurotmesis and from eachother only when indications of recovery or de-generation appear with the passage of time. Thedevelopment of a palpable neuroma in a patientwith a total loss of function is strong evidence ofcomplete neurotmesis.

(2) In partial neurotmesis the changes are similarto those of complete neurotmesis, but affect an arealess than the whole area of distribution of theaffected nerve. Not all muscles innervated by thenerve cease to function, and sensory loss is onlypartial too, unless, as sometimes happens, all thesensory or all the motor fibres are in the partdivided. Sensory disturbance may take the formof a diminution in the intensity of sensation(hypoesthesia) without its complete loss in any onepart of the area supplied; this form of sensory lossis commonly seen in traumatic ulnar neuritis.Persistent hyperaesthesia, hyperalgesia, or pain issuggestive of a partial lesion; spontaneous pain isuncommon in a complete lesion, and, if it occurs,usually disappears after a few weeks. So, too,persistance or excess of sweating in the territory ofan injured nerve is indicative of a partial lesion.

In most cases of partial neurotmesis, a hardneuroma develops at the site of trauma, and if this



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/

POST-GRADUATE MEDICAL JOURNALis palpable the diagnosis is clear, and operation isindicated. t

Differentiation of partial from complete neurot-mesis is particularly difficult in a sensory nervewith a small territory of supply. Procaine block(Highet)48 has a special application here. If afterblock of the affected nerve anaesthesia is increased,or if after block of its neighbour or neighbourssensation remains, division may be regarded asincomplete.

(3) and (4) Axonotmesis and neurapraxia can bedistinguished from the first two forms of injury bythe absence of a palpable neuroma, and also bydissociation of sensation, if it occurs. In neura-praxia subjective numbness, paraesthesiae, andsome loss of position sense may occur withoutobjective sensory loss; the fibres for position senseare among those large calibre fibres whose myelinsheaths seem particularly sensitive to concussion.In general, however, only the passage of timedifferentiates neurapraxia from axonotmesis. Inthe former, fibrillation does not appear in theelectro-myograph, the strength, duration curveremains of constant shape, the reaction of degene-ration is still absent after three weeks, for Walleriandegeneration does not occur, muscle biopsy (occa-sionally justifiable in this connection if a suspectedneurapraxia is slow of recovery) shows a normalhistological picture, and recovery is early andquickly complete.Attempts have been made by contrast radiography

to show whether interruption in a nerve's con-tinuity is complete.88,o07 Perabrodil in salinemay be injected intraneurally; if it passes the siteof injury, conduction is likely to be recovered,-andthe injury may be assumed to be either neurapraxiaor axonotmesis; if the perabrodil fails to pass thesite of injury, it is doubtful whether recovery willoccur. The method is technically difficult, and notwithout risk of injury to the nerve investigated,and it is unlikely to become generally popular.

(5) The presence of ischaemia in a nerve or inthe muscles which it supplies may easily be over-looked. Pronounced contractures and loss ofvoluntary power in muscles other than those sup-plied by the injured nerve are strongly suggestiveof the presence of ischaemia, particularly if anysensory disturbance follows a glove or stockingpattern rather than the pattern of an area suplpliedby the injured nerve trunk. The possibility ofischaemia of a nerve or of the muscles supplied byit should be considered particularly if one of themain arteries is known to have been damaged or ifthe distal pulses are impalpable; ischaemic effectsare greatest when, in addition to damage to themain arterial trunk, collaterals are divided in anextensive wound. The presence of ischaermiarenders recovery unlikely even after an apparently

successful suture, and the frozen, contracted,atrophic, and discoloured appearance of thisdoubly affected extremity is a danger signalwarning against any attempt at nerve repair. Ifthere is any doubt about the degree of ischaemia inone of the muscles supplied by an injured nerve,muscle biopsy should be performed."I The histo-logical changes of ischaemia are characteristic evenwhen associated with those of denervation, andtimely biopsy may avoid an operation doomed tofailure and the deferred hope of a fruitless con-valescence."IA neurological defect associated with a traumatic

aneurysm usually appears full-blown at the timeof injury,27 and rarely recovers spontaneously; ifexploration is delayed for a month or two to allowfull expansion of collaterals, nerve and artery canbe dealt with simultaneously.

Ischaemia of nerves and of their territoriesoccurs, of course, in thermal lesions of the extremi-ties, and it occurs too in damage to the main vesselsof a limb, but a more insidious type of nerveischaemia has been described by Parkes76 and bySeddon and Holmes.99 The latter authors des-cribe a median nerve injury with all the manifesta-tions of median nerve interruption, with a divisionof the nerve and a neuroma on its proximal stumpin the upper third of the forearm. They found,however, that the distal segment of the nerveappeared at operation to be threadlike from thelevel of the lesion as far as the wrist, and histo-logical examination showed an excessive amountof collagen in each funiculus, almost entirelylimited to the endoneurium. The nerve alone hadsuffered, and the muscles were not ischaemic, andSeddon and Holmes concluded that the change inthe distal segment was probably due to a con-comitant occlusion of the anterior interosseusartery and its median branch. Parkes recorded asimilar type of lesion in both upper and lowerlimbs. Paralysis in nerve trunks occurred withoutnervous interruption demonstrable at operation.In the upper limb cases, the main vessel wasgenerally damaged, but in the lower limb the causewas usually a closed fracture with great swellingof the leg followed by numbness in the foot, themuscles and nerves suffering ischaemia frompressure. Nerve fibres do not all resist ischaemiaequally, and often there is a delay in the onset ofthe nerve disability; the unmyelinated C fibressubserving slow pain are the most resistant toasphyxia.

(6) Combined lesions offer great difficulty indifferential diagnosis. Only the passage of time,repeated and careful study, and a detailed clinicalrecord will permt the assessment of a combinedinjury in termsi of partial neurotmesis, partialaxonotmesis, partial neurapraxia and ischaemia.




j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

SURGERY OF PERIPHERAL NERVE INJURYThe co-existence of all four of these is admittedlyrare-when a missile passes through a main arteryand its companion nerve, dividing certain of itsfibres, disrupting the axons of others, and breakingin others again the continuity of their myelinsheaths-but a combination of neurapraxia andaxonotmesis, or of neurotmesis and axonotmesis isnot uncommon.Even more puzzling are instances of double

division of a nerve, or different forms of nervelesion at different levels, or the occurrence ofparalysis from sulphonamide or serum injection,or of pressure paralysis, or of unrelated non-traumatic nervous disease, in a wounded limb. Allthese demand accurate records for their confidentelucidation.

Level of InjuryWhen the type of lesion has been established, its

level must be decided from the site of injury andfrom the extent of anaesthesia and of paralysis. Innerve injuries associated with fracture, the level ofthe fracture, and consequently presumably of thenerve injury, is known. Considerable difficulty inascertaining the level of the lesion may be met if apatient has sustained multiple wounds. It may bedifficult to decide, for example, at what level inaxilla or arm the ulnar nerve has been damaged,and a wide exploration may be necessary if opera-tion is performed. If a neuroma is palpated or ifa foreign body is seen radiologically near the trackof the nerve, a strong suspicion of the level may beentertained.Even more treacherous are the circumstances

when a nerve injury occurs at two or more separatelevels. The ulnar nerve, for example, may bedoubly divided in a gunshot wound of the flexedelbow, and a nerve injury may occur in a lowerlimb and escape notice because of a concomitantcord lesion. In the latter case, physical examina-tion will often prove the double injury.65 Thepresence of a scar may raise the suspicion of apressure palsy below the higher lesion. Continuedparalysis of one muscle group, when other musclegroups have recovered, should again suggest adouble lesion. Disproportionate wasting of a lowerlimb after a spinal injury should attract attentionto a concomitant peripheral nerve lesion, as alsoshould the failure to elicit a mass reflex by stimu-lation within the cutaneous distribution of a peri-pheral nerve. If the nerve lesion is in the cordonly, no fibrillation action potentials are registeredin the myograph; evidence of fibrillation suggests acomplicating peripheral nerve injury. In an injuryof the cauda equina complicated by a peripheralnerve lesion below, fibrillation is present in musclessupplied by the injured nerve, but there is noanhidrosis in its cutaneous territory.

Treatment

(a) Immediate TreatmentIn a wounded man who has sustained a peri-

pheral nerve injury at the time of wounding, thefirst aim is to save life and to treat the mainwound. During the surgical debridement of thewound, a nerve may be found divided in it. Noformal suture should be performed. The nerveends may be approximated to each other by twonon-absorbable sutures or, if they cannot be soapproximated, a suture should be placed in thedivided perineurium at each end, so that thenerve can subsequently be found readily, or thedivided ends may be sutured to adjacent muscle orfascia to prevent their retraction. Care should betaken to avoid the application of any large quantityof sulphonamide directly to the injured nerve.Sulphonamide in small doses, such as are givenwhen a wound is lightly dusted with powder,probably does not damage a nerve, but nervoustissue is extremely susceptible to large local con-centrations of sulphonamide, and paralysis mayfollow the application of sulphonamide to thesurface of an undivided nerve trunk.54 Primaryor secondary suture of the skin wound should beperformed without regard to the nerve injury.During evacuation and early convalescence thelimb should be splinted, not only to afford immobi-lisation to wounded bone and soft tissues, but alsoto ensure relaxation of paralysed muscle and theavoidance of pressure on anaesthetised skin. More,perhaps, than any other injury, the nerve injuriesof warfare require careful documentation, for onthe clinical record may depend future successfuloperation. It is important to record the date andtime of injury, the nature of the trauma, thepresence of associated injuries (particularly tovessels, muscles and tendons), the site of entranceand exit wounds, the position of the limb when thewound was sustained, the type of local or radiatingpain, numbness and weakness after wounding, theappearance of the wound at operation, the colourand temperatures of affected and unaffected limbsand the presence of pulses, the presence or absenceof sweating, whether a tourniquet has been usedand for how long, and the precise operative pro-cedure undertaken.

(b) Definitive Treatmenti. Whether to operate

Neurotmesis, partial or complete, must be ex-plored. Axonotmesis, if not associated with neu-rotmesis, requires exploration only if regenera-tion is prevented by dense perineural scarring.Neurapraxia, unaccompanied by any other type oflesion, should not be explored.

235September, I946P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

POST-GRADUATE MEDICAL JOURNALThe clearest indications for operation are: (I) a

nerve already seen divided at a debridementoperation, and (2) a tender swelling on the course ofa nerve, at the site of an injury, and associatedwith motor or sensory loss in the distribution ofthat nerve; a palpable neuroma is certain evidenceof complete or partial neurotmesis.

If there is no neuroma palpable, the decision toexplore is postponed. Successive opportunitiesarise for reassessment of the developing situation.A neuroma may become palpable on the one handor, on the other, signs of recovery may appear.The appearance of the reaction of degeneration isnot in itself an indication for operation, and anadvancing Tinel's sign is not an indication againstit. For each affected muscle the expected time ofrecovery can be estimated, and failure of signs ofreinnervation in electro-myograph or strength-duration curve when these signs might reasonablybe expected is good reason for exploration.

In deciding whether or not a lesion is a neurot-mesis requiring operation, or an axonotmesiswhich will recover without operation, considerationmust be paid to the individual nerve. In the caseof injury of the ulnar nerve just above the elbowthere is not long to wait before signs of re-inner-vation appear in the flexor carpi ulnaris, and littleis lost by waiting; in injury of the sciatic nerve inthe buttock it may be many months before signsof recovery in the calf muscles, and the optimumtime for operation may be missed by procrastina-tion.Even if some recovery occurs, difficulty may

arise in deciding whether to operate if that re-covery ceases short of complete function. Thecontribution of the nerve injured to the totalfunction of the limb must be considered. There islittle point in exploring a median or posteriortibial nerve whose sensory function has largelyrecovered, or an ulnar nerve whose motor functionhas returned; yet a median nerve which has re-gained only motor function, or an ulnar which hasregained only sensation, may well merit explora-tion and perhaps resection and suture.Learmonth6o has advanced certain indications

for exploring an injured nerve irrespective of evi-dence of regeneration, with the object of ascertainingits exact condition and ensuring a good permanentbed. These indications include previous long-con-tinued sepsis in the wound; the situation of theinjured nerve at the bottom of a healed "trough"wound (this is particularly common in the case ofthe ulnar behind the elbow); a history of unusuallysevere bleeding from a muscle mass at the time ofinjury which may have resulted in local necrosisof muscle and late strangulation of such a nerve asthe median under pronator teres; closed injuries ofthe crushing type when such a nerve as the sciatic

is likely to be tightly gripped in a collar of fibroustissue formed from bruised hamstrings; concomitantinjury of the main artery at the same level asthe nerve; injury of more than one nerve of alimb at the same time; and, lastly, persistent painor persistent hyperhidrosis in the area of distribu-tion.To some extent, too, the decision to operate

depends upon the degree of success which is knownto attend suture of the particular nerve injured.In a complete lesion of the dorsal interosseousnerve the results of suture are in general so un-satisfactory that there is justification for the viewthat no useful purpose is served by performing it;even so, exploration may confirm that the lesion isneurotmesis and that tendon transplantation mayproceed at once. So, too, in a high complete lesionof the brachial plexus operation offers no advan-tage.2. When to OperateWhen it has been decided to operate, the time

for operation must be chosen. In a closed lesion,the most suitable time for intervention is probablyin the third or fourth week after the injury hasbeen sustained. Since the ability of the centralstump to send forth axons is not impaired if it besevered a second time more than a week and lessthan a year after its first division,55 there is noadvantage in immediate suture, and there may besome disadvantage in it. In the third or fourthweek, the outgrowth of Schwann cells from thedistal stump is at its most vigorous,', 55 and on thisoutgrowth of Schwann cells, attracted73 or at leastacceleratedI3I towards the proximal stump by akind of reversed neurotropism, but proceedingthroughout the whole distal segment as far as themyoneural junction, depends probably the numberof outgrowing axons which will find a path to theperiphery. At this time also, the slender peri-neurium is toughened by fibrosis and lends itselfto successful suture. A delay of one or two monthsprobably does not hinder the operation greatly, forregeneration is no slower after excision of neuromaand suture than after primary suture,55 but a delayof longer than five or six months will delay finalreinnervation and will lessen the degree of ultimatefunctional recovery. After about one hundred daysfrom the time of injury, the whole peripheralstump shrinks, and the Schwann tubes narrow,55while the endoneurium and perineurium becomethickened by collagen, so that there is incongruityof size at the suture line; relatively few growingaxons find Schwann tubes for their passage; theultimate size of the growing fibres remains small;medullation is slow, and final functional recovery isinferior in quality. Moreover, as time passes, thegap between nerve ends may increase, and the

Septemzber, I946236P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/


:~::*.:::;.. ·*·

....'iiii ". ..4:.:

.·;:

..;

lli?!ii~:I·:r::

..:· .rr

FIG. 5.-Median paralysis. Failure of opposition of right thumb. Note wasting of thenar eminence.

^~: irfi.^'

.~: i.R' ,} -=

.....·......... :;S, ::i:t~ii:-. I|

.o, ...' d .. R ...., . ,;·i::: ·. a.. s | *

.....~~:·:. lI

; .i·

..:.·......cil,i·.i·;,-| :. .*. ::i.. ".i ^

*~:~;: i"..;.

...:.:...

~~~~~~·.:· .........';'ii.i.' i. ............... .... ...;: . ..

FIG. 6.-Repair of ulnar nerve at elbow. Incision. (After Learmonth.)(Reproduced by kind permission of Professor Learmonth and the Journal "Surgery, Gynecology and Obstetrics")



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/


iiiiidiiiii~iiil~:!i.............

..........

dddamo:

... .... ... ..:·i::iiiif:l................~ii.

FIG. 7.-Repair of ulnar nerve at elbow. Neuroma and glioma exposed and cleared.

:....

............iiii~...

·· :;r:::~r:.:: II·:...

.....·....··...·..

.....

:..........iiii~ii

MHiiiii

::ii*iii ii!'ii~':ilii!gi~Wiii i"

FIG. 8.-Repair of ulnar nerve at elbow. Gauze drawn deep to common flexor origin to form canal;traction suture in neuroma.



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/

SURGERY OF PERIPHERAL NERVE INJURYproximal and distal stumps may become less elasticand extensible. SpurlingIo5 has shown that inAmerican wounded who sustained nerve injuries inthe European theatre of war, when operation wasperformed between the twenty-first and twenty-eighth days only I per cent of the patients hadgaps too great to be closed by suture; after theNorth African Campaign, when nerve injurieswere evacuated to the United States, Io per centof the gaps were too great to be closed by suture.German surgeons in general waited some months

before undertaking operation,53, II4 I5 apparentlybecause they confused neurapraxias with otherforms of nerve lesion,124 and they waited in thehope that each individual lesion would turn out tobe a neurapraxia. The Italians, on the other hand,operated onr the eighth to tenth day after injury23if there was complete paralysis of a large area, andif electrical reactions had already changed.

Operation should not, of course, be undertakenin the presence of an unhealed wound, or, indeed,for about fourteen days after healing; a healedwound should be massaged daily before operationto make sure that infection will not reawaken in it.Since the advent of penicillin and the increasedpopularity of delayed and early secondary suture,the average time of operation upon nerve injurieshas become substantially earlier.75 Learmonthand Wallace63 have shown how useful in thisrespect is a close collaboration between plastic andperipheral nerve units. Not only can open woundsbe closed by plastic methods, but dense scars canbe replaced by whole skin and fat, so that scarringafter repair of a nerve may be minimal, andstrangulation of the sutured nerve unlikely. Whena wound has been closed, or when a scar has beenreplaced by skin transplant, the incision for thesubsequent nerve repair should be placed along oneedge of the graft, so that scar tissue in the region ofthe suture is not amalgamated with an overlyingscar in the skin.

Learmonth6o has advised that when the mainvessel of a limb is injured at the same time as anerve to that limb, it is wise to perform a pregang-lionic sympathectomy before the downgrowth ofaxons reaches the digits, to prevent the reimpositionof a degree of vaso-constrictor tone on the digitalvessels, and progressive circulatory inadequacy.3. Non-operative McasuresManagement of Denervated Extremities before

Operation and during the Post-operative Period.-This treatment involves the use of splints and othermethods of immobilisation, massage, exercises,baths, electrical treatment, and occupationaltherapy.

It was remarkable, during the recent war, howmuch less were contractures and other deformities

in denervated limbs which had not been immo-bilised than in those which had been carefullysplinted. Nevertheless, certain splints are desir-able. Splints should prevent the over-stretchingof paralysed muscles, but should permit the freemovement of the unaffected muscles. The aim ofa splint should be to approximate the point oforigin of the paralysed muscle to the point of itsinsertion,49 and to avoid stretching by gravity orby unparalysed antagonists. Splinting should notbe regarded as synonymous with immobilisation;the limb should not be immobilised and all jointsmust be allowed as full movement as is consistentwith the relaxation of paralysed muscle. Splintsmust not press on anaesthetic skin or interfere withthe circulation of an extremity whose circulation isin any case impeded, and they should be capable ofbeing worn during occupational therapy and evenat work. Useful splints of many varieties havebeen devised for the prevention of flexion deformityof the ring and little fingers in ulnar palsy, and forthe prevention of over-stretching of the extensorsby gravity in radial palsy, but in other situationssplints are rarely necessary. Even the splints forthese two deformities should be of a spring type,permitting full movement of all the antagonistmuscles and of all the joints of the digit. Volun-tary exercises have a high value, and should aim atexercising antagonist muscles and preventing thedevelopment of trick movements. They should beused too, but not to the point of fatigue, forrecovering muscles.

Re-education of muscle should begin as soon asvoluntary movement is recovered; a faradicresponse should not be awaited, for voluntarymovement may precede it by weeks or months.The movement of a joint through its whole normalrange will not usually over-stretch any paralysedmuscles. Before exercises are begun, the limbshould always be warmed for at least an hour; theexercise of the muscles of a cold limb has littleadvantage.

Massage is valuable, as also are baths in wax andin water, for the maintenance of nutrition in thelimbs and for the correction of deformities if theseare already present, but flaccid muscle is veryvulnerable, and must not, in massage, be com-pressed against bone. As Mennell7o has observed,the ability of a masseur to treat nerve injuries canbe measured by the change in technique when hepasses from the normal to the flaccid part of thelimb.

Electrical treatment is probably beneficial also,8A galvanic current should be employed until thefaradic response returns.4' Muscle atrophies eventhough electrical treatment be carried out, but theweight lost by a paralysed muscle is less if electricaltreatment is maintained than if it is omitted.38

Se,btembey, I946 239P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

POST-GRADUATE MEDICAL JOURNALFischer28 showed that muscle stimulated electri-cally for twenty minutes a day undergoes a strikingincrease of weight compared with its resting fellowof the opposite side, and Gutmann and Guttmann36demonstrated that while electrotherapy does notprevent atrophy during the first two weeks, itreduces the degree of later atrophy, and acceleratesthe return of function. Jackson and Seddon57showed also that electrical treatment does notprevent early wasting, but reduces the amount ofthat wasting, and they claimed, too, that galvanismarrests atrophy at about one hundred days afterinjury, whereas if galvanism is omitted atrophycontinues for four hundred days. Galvanism doesnot restore the volume of muscle already lost, andit should therefore be started early.4. Operative techniqueThe patient's position should be carefully ar-

ranged for operation, and a suitable anaestheticchosen. In most cases, a general anaesthetic isemployed. The only advantage conferred by localanaesthesia is the permission it affords for elec-trical stimulation of sensory fibres, but there isrelatively little value in information so obtained(see above).A long incision should be employed, placed

usually in the line of the nerve. If a plastic opera-tion has been done previously to provide skinreplacement of a scar, the incision should circum-scribe the transplanted skin.63 Care should betaken in approaching the nerve to effect completehaemostasis, but to avoid injury to any collateralchannels, particularly if the main vessel of thelimb has been simultaneously damaged. Thenerve trunk or the ends of the divided nerve areidentified and exposed, and the nerve is followedproximally and distally for a considerable distance,care being taken to preserve the origins of branches;the order in which branches leave the parent trunk,and even the actual level at which they arise, arefairly constant. The nerve can now be inspected,and, if it is undivided, surface petechiae, scarring,and congestion are noted. Careful palpation isperformed for the detection of any local hardnessor thickening, of a lateral neuroma, of alteration intexture, and of the presence of intraneural foreignbodies. If undivided or partly divided, the nerve isthen tested directly by faradic current. Both theproximal and distal segments are tested, and thevoltage employed must be graduated to avoidspread of current to neighbouring nerves, parti-cularly when the nerve studied lies close to the pointof separation of two main divisions, as in the caseof the popliteal nerves below the bifurcation of thesciatic. During electrical testing, the nerve mayconveniently be suspended on narrow tapes. Thedecision is then made whether to be content with

the neurolysis which has already been performed,or whether to proceed to excision and suture.There is no difficulty in the choice of operationwhen the nerve is completely divided. When thenerve is partly divided, it may be possible toseparate the divided from the undivided part andto perform a partial suture. This is particularlyconvenient when one or other of the poplitealdivisions is divided within the sciatic sheath, butthe same technique may be applied to almost anynerve which is the seat of partial neurotmesis.Should it be decided to perform partial suture, thepresence of undivided fibres in the scar of theapparently divided part of the nerve may bedetected directly by faradism.When the continuity of a nerve is not completely

interrupted, the decision whether to resect or notmay be exceedingly difficult. If loss of function iscomplete, if a reaction of degeneration is present,and if there is a palpable thickening and hardnessat the site of the lesion to suggest substantial intra-neural fibrosis, resection is usually desirable. Ifthe nerve, on the other hand, has a normal appear-ance and feel, it should be left undisturbed eventhough loss of function is complete and R.D.present. Lipchina66 and other Russian authorsadvise resection for localised intrafascicular haema-toma-formation and oedema, even though theperineurium is intact and soft, but in most Britishcentres the practice in early exploration has beento avoid resection of a nerve unless it presentssome local thickening and hardness. At earlyoperation, if a faradic response cannot be elicited,hardness to touch is the most reliable criterion ofthe need for resection.

In the case of a complete division, the neuromaof the proximal segment and glioma of the distalare excised by division of normal nerve. Eachbulb in turn is incised in successive incompleteslices, from its tip toward normal nerve, and thefirst slice which shows normal nerve bundles in thetransected trunk is made complete, and that levelis the point of excision of the bulb. It is particu-larly important to have well-defined funiculiobvious in the transected distal segment; thegrowing axons penetrate scarring in the proximalstump more readily than they enter a scarred distalstump. During the process of successive slicing ofthe bulb and final section of the nerve, the nerveand bulb may conveniently be laid on a straightmetalspatula, around which moist gauze is wrapped.The best instrument for section is a Gillette razorblade carried on an artery forceps.2o The finalsection of the nerve should be directly transverseacross it. In this way, two clean nerve ends areobtained for suture. The neuroma and glioma arepreserved for histological examination; the experi-enced neuro-pathologist, estimating the state of




j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/


?·: ;:irti'?.::

-

iRi

jGiuliani.........

.... .. ...

FIG. 9.-Repair of ulnar nerve at elbow. Neuroma drawn through tunnel deep to flexors near origin.

...1i

!1:1flil!

I-7M"..." ............:.

.:.:-...

.ii ......i..

FIG. io.--Repair of ulnar nerve at elbow. Suture completed. (A) Three tension sutures in place;insertion of anterior row of sutures. (B) The same; two rows completed.



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/


! .:..........

~-~' :i;~-· ··

*' ..".....

.I'" .....I .:.....::..:.:.:.

l 2...:.....

;.....'..... ..'.......

....,..;'....::' ..' ''!.....

........ .. .... ......... ........ .... ::- ·81...................................

FIG. I I.-Repair of high sciatic lesion.Question-mark incision.

(Reproduced by kind permission of "Surgery, Gynecologyand Obstetrics.")

(·J

rr

soR

·.!:::-.

·....,.:: .

-- :·;::;:::·:.·...

:e :·: ·I I·,,er IIII··llAla

FIG. 12.-Repair of high sciatic lesion. Division of insertion ofgluteus maximus.

(Reproduced by kind permission of "Surgery, Gynecology and Obstetrics."



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/

SURGERY OF PERIPHERAL NERVE INJURYthe divided ends of the nerve from the histologicalappearance of the extremities of the interveningsegment, and observing the degree of intraneuralfibrosis and the state of axons and Schwann ele-ments, can hazard a valuable prognosis of thedegree of recovery to be expected. When the endshave finally been prepared for suture the gapbetween them is carefully measured and recorded.Two or three stay sutures are inserted axially

through the clean nerve-ends, and are drawn tighttogether, and tied (Fig. IOA). They may con-veniently be of silk. As they are tightened,neighbouring joints are placed in that extremeposition which gives greatest relaxation to thenerve. The anastomosis then proceeds. Gutt-mann43 has shown experimentally that of theavailable material for suture, human hair and whitesilk produce least reaction, but it is probable thatthe type of suture material used is not of verygreat moment, provided the suture is fine, unab-sorbable, and carried on an eyeless needle of thesame diameter as the thread. In nerves such asthe radial which seem to lend themselves to suc-cessful suture, the suture is successful whatevermaterial is used; it seems likely, therefore, that inother nerves, want of success is due more to thenature of the nerve than to the qualities of thesuture material. The standard black silk atrau-matic arterial suture now obtainable is perhapsmost commonly used. Tantalum wire (0.003 inch)carried on an atraumatic needle is equally suitablefor suture of nerve, but its only advantage is thatdisruption at the suture line can be recognisedradiologically after operation.xo6 Young and Meda-warI30 introduced a form of sutureless anastomosisin experimental animals, and this was later usedclinically by Seddon and Medawar95 in man, andMichael and Abbott7i have reported favourably onhuman fibrinogen glue. In the method of Youngand Medawar, the nerve-ends, having been approxi-mated, are surrounded by a mixture of plasma andtissue extract which clots to a firm jelly and bindsthe nerve ends together. The plasma is freelypermeable to growing axons, and remains longenough to give firm union between the divided endsbefore its final dissolution, which is complete. Thisform of anastomosis, fibrin suture, cannot be em-ployed if there is any tension between the nerveends, but it is useful in certain situations, such asthe hand and fingers, where suture is not appli-cable.I32 Tarlov and his fellow-workersIIo, I havedevised a rubber mould which is used to surroundthe anastomosis with plasma clot, and to compressthe clot while it sets.

In most operative situations, anastomosis ismade by suture. The stay sutures having beentied the divided ends are approximated by inter-rupted stitches, which pierce perineurium only.7

The stitches are inserted at intervals of approxi-mately 2 mm. The stay sutures can be employedto rotate the anastomosis until it is circumferentiallycomplete (Fig. IOB). On completion of the sutures,the anastomosis is gently rolled between thumband forefinger, to return to the nerve its normalcylindrical contour.

If difficulty is found in approximating thedivided ends of the nerve, various manoeuvres areavailable to give relaxation. The nerve may befollowed and dissected free proximally and distallyfor many inches. When approximation is at-tempted, branches arising from the proximal ordistal segment may be found to be so taut that theyprevent full relaxation.. These branches usuallyretain their separate identity within the nervesheath for a substantial distance above theirapparent point of origin, and if they are stretchedgently away from the nerve they may be freed bydivision of the sheath proximally with scissors;the funiculus to coraco-brachialis in the musculo-cutaneous nerve, that to flexor carpi ulnaris in theulnar, and that to gastrocnemius and the posteriorpart of soleus in the medial popliteal retain theiridentity for particularly long distances within theirparent trunks.

In some situations, additional length may beobtained by transplantation. This is particularlyso in the case of the ulnar nerve, whose proximalsegment may be freed above a division of the nervein the forearm, or whose distal segment may befreed after division of the nerve in the upper armby anterior transposition of the nerve at the elbow.In most cases, however, the operator is dependentupon a combination of the position of the limb andmobilisation of the nerve.

In performing anastomosis, a mild tension at theline of suture is said to be an advantage, tending toproduce correct orientation of axon tips oppositeappropriate Schwann tubes.x29 Considerable ten-sion, however, may prevent regeneration,5' and thestretched nerve may, after operation, be trans-formed to a ribbon of connective tissue.52 Thebiological limit of stretching is less than theanatomical limit52; a stretch of 10 per cent of themobilised length of a nerve may be applied withoutdamage, but a 30 per cent stretch may beharmful even when done in stages. 9° The operatoris constantly faced with the temptation to over-stretch. Spurling0os reported that although aninsurmountable gap prevented suture in only Iper cent of approximately I,500 divided nervesexplored in American centres in this country be-tween D Day and VE Day, 4 per cent of theanastomoses were shown later to have separated.When the gap between the nerves is too great to

permit of repair by anastomosis without tension,the question arises whether a graft is likely to be

September, I946 243P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

POST-GRADUATE MEDICAL JOURNALsuccessful. It may be said, initially but conclu-sively, that preserved grafts, hetero-grafts (fromanother species); and homo-grafts (from otherindividuals of the same species) have in no singleinstance been found to allow regeneration inman.4, 132 Fresh hetero-grafts induce a severe re-action and undergo none of the degenerativechanges necessary for axon penetration.I33 Thealcohol-preserved and formol-fixed rat spinal cordof Gosset and Bertrand33 is attacked by giant-cellsand destroyed. Fresh homo-graft (in rabbit) mayoffer a basis for recovery,,9 33 but frequently pro-vokes a lymphocytic reaction, which is onlyslightly lessened by previous saline storage.39 Thealcohol-fixed transplant of Nageotte73 is removedby microphages. Klemme59 has employed cadaverhomo-grafts in man, but his results are not fullydocumented. The only peripheral nerve graftwhich has been used successfully in man is theauto-graft, the same individual serving as donorand recipient, and it has been shownI26 that newfibres grow through an auto-graft hardly moreslowly than through a peripheral stump. In somesituations, auto-grafts have proved highly success-ful, notably in the repair of the facial nerve withinthe facial canal; in developing this operation Dueland Ballance,5 24 adopted for transplant the pre-degenerated nerve recommended on theoreticalgrounds by CajalI7 in 1928; Sanders and Young9Ihave shown, however, that the only measurableadvantage conferred by predegeneration is an in-creased toughness of the perineurium to the tensionof sutures.

For a graft to be successful, it must not only beobtained from the same patient, it must also be ofthe same diameter or a little larger than the nervefor whose repair it is employed. The technique ofgrafting entails the interposition of the graft and adouble suture proximally and distally. No prac-tical advantage' has been reported from the secon-dary excision and suture of the distal anastomosisafter an interval as suggested by Davis and Cleve-landI9 and by Dogliotti.23The ideal type of nerve graft is found in double

lesions of the ulnar nerve when that nerve has beentransected both above and below the elbow; itsintervening segment, being from the same patientand of precisely the same size, may be transplantedto the front of the elbow as a perfect free graft.Nerve grafts may be used, in combination with aplasma technique, to replace wide gaps in digitalnerves. In nerve injury of both plantar nerves,the lateral plantar may be employed to replace adefect of the medial plantar, and in a doubleinjury of the popliteal nerves it is legitimate tosacrifice a part of a widely divided lateral popliteal,where paralysis can be compensated by ortho-paedic operation, to replace a defect in the medial

popliteal, whose regeneration is so nuch moreimportant. To repair a long defect in a large nervea "cable graft" may be employed-several pieces ofa small nerve inlaid in parallel, so that their sum-mated cross-sections exceed the sectional area ofthe injured nerve.Another expedient in the repair of large gaps is

the method of bulb suture. If it is decided, beforetrimming of the divided nerve ends, that the gap istoo large to close immediately by suture, the bulbsmay be left in place and sutured to each other as atemporary measure, excision and formal anasto-mosis being postponed until a later date. Hereagain, it is doubtful what success may be expected,for it seems probable that nerve tissue is notcapable even of slow stretching with safety. Whena gap is too large for anastomosis, at a site whichdoes not lend itself to nerve graft, all idea of sutureis usually abandoned, and reliance is placed on laterorthopaedic measures for the recovery of motorfunction in the limb.When anastomosis is complete, its distance from

the nearest bony landmark is measured and re-corded, for precision in calculating the distancewhich the growing axons must cover, and assistancein predicting the date of functional recovery. Thenerve should be placed in a comfortable bed, whereit is unlikely to be strangulated in scar tissue. Thesafest bed for an anastomosed nerve is in the planebetween two muscles, and, failing that, within thesubstance of a muscle. If possible, the nerveshould not be left in a subcutaneous position; thisapplies particularly to the ulnar nerve at theelbow, to the median nerve at the wrist, and to thesciatic nerve in the thigh. A nerve lying betweenmuscles or within a muscle need not be surrounded,as has been suggested, by tantalum foilos5 or by anarterial tubeI28 or by cargile membrane. In nocircumstances should a nerve be swathed in fascialata after anastomosis.'6While suture of the wound is completed, the

limb is continuously held in that position whichaffords maximum relaxation to the repaired nerve,and a well-padded plaster cast is applied to retainthat position after operation. The plaster castremains, usually, for 4-6 weeks after operation,and thereafter the limb may be gradually broughtinto the position ofcomfort andof optimumfunction.

After operation, splinting, massage, electricaltreatment, baths, occupational therapy, and psy-chological encouragement are continued; a closewatch is kept for the signs of recovery, which arecarefully documented, and gentle exercises arebegun as soon as voluntary power is noticed ineach successive recovering muscle.

PrognosisThe prognosis in a case of nerve injury is never




j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

SURGERY OF PERIPHERAL NERVE INJURYfinal throughout the whole course of treatment;prognosis is a dynamic development approaching,but never quite attaining, perfect accuracy as newfacts become available with the passage of time.An accurate early prognosis is possible if a nerve isseen intact at the initial debridement of an openwound; a less accurate prognosis is possible if it isseen divided at that time. Return of functionafter a week or two, too early to be explained byregeneration, is characteristic of neurapraxia andrelatively rapid and full recovery can be predicted.The reaction of degeneration, the onset of fibrilla-tion in the electro-myograph, and increasingsteepness of the strength-duration curve excludeneurapraxia and promise disability for somemonths.At the time of definitive operation, prognosis is

narrowed within more exact limits. If the nerve,though showing signs of degeneration, is foundintact and free from thickening (axonotmesis) itcan confidently be predicted that full function willbe recovered, and a fairly accurate date for thereturn of function in each denervated muscle maybe calculated from the known rules which governthe rate of regeneration. If, on the other hand, itis found divided, and is repaired by suture, a fairlyexact prognosis can be estimated from: (I) theperiod from time of injury to time of operation,(2) the degree of atrophy and the presence orabsence of stiffness and contracture, (3) the magni-tude of the gap, (4) the degree of tension aftersuture, (5) the neuropathologist's report upon thestate of intraneural architecture at the lines ofsection of proximal and distal stumps, (6) the levelof the anastomosis, (7) the habitual characteristicbehaviour of the particular nerve injured, (8) thecirculatory efficiency of the denervated area, and(9) the age, re-educability, and willingness of thepatient.

Later still, the strength-duration curve, themyograph, and sometimes muscle biopsy may beemployed to check, correct, or even alter, a previousprognosis.

Results of TreatmentA final evaluation of recovery should be delayed

until at least 24 or 3 years after injury.83 Re-covery is never complete after suture, but afteraxonotmesis and neurapraxia a complete return offunction may be hoped for. Foerster3o reported1,320 complete and 660 partial motor recoveries,all spontaneous, in a series of 3,099 peripheralnerve injuries sustained in the first world war.

Individual NervesNo complete consideration of the individual

nerves is attempted; notes are given which may behelpfully additional to a working knowledge of the

applied anatomy of those nerves which are morecommonly damaged.

I. Cranial Nerves. Facial nerve injuries havebeen extensively studied by Duel and Ballance,5, 24and their established technique for the replace-ment of lost facial nerve by degenerated lateralcutaneous nerve of the thigh is well established andmost successful. Starr has recorded a successfultwo-stage suture of the facial nerve trunk withinthe parotid gland after a gunshot injury. The onlyother cerebral nerves whose extracranial injurieshave been reported are the last four cranial nerves,which may be injured together in wounds of theneck with associated damage to the carotid vesselsand sometimes to the sympathetic chain. Thesymptoms such an injury produces are groupedtogether in the "traumatic syndrome of Villaretand Vernet."'I8, "9 The facial nerve is sometimesconcomitantly damaged. Injuries of these nerveshave not been successfully treated by operation,but all the symptoms of nerve damage tend toimprove except the vagus and accessory paralysis.There is usually a persistent abductor paralysis ofthe larynx, and some dysphagia.The accessory nerve may be injured alone in the

neck, with paralysis of the trapezius, and with orwithout sternomastoid palsy. This nerve does notlend itself to repair.

2. Brachial plexus injuries comprise a subject toovast and complicated for full description here. Itmay be said simply that injury of the cords and oftheir branches may be undertaken with high hopesof success, injury of the trunks should be attemptedbut success is unlikely, and in injury of the rootsoperation is hopeless.6, 44, 58

3. The long thoracic nerve is affected in less thanone half of i per cent of war wounds; serratusanterior paralysis has followed excision of breastand incisions for axillary abscess. Spontaneousparalysis of the serratus anterior muscle maydevelop from traumatic long thoracic neuritis as anoccupational lesion of such workers as plasterers,who hold the arms above the head for long periods,and in porters who carry on their shoulders weightswith sharp edges; in these the lower fibres oftrapezius sometimes suffer too for an unexplainedreason. A similar palsy in swimmers may becaused from friction by the scalenus medius as thearm is carried forwards in the breast-stroke, andperhaps this explains the occupational form of thepalsy.The long thoracic nerve cannot well be repaired

unless it is seen divided accidentally in the courseof an operation. The winging of the scapulawhich results from the nerve palsy can be correctedonly by the orthopaedic surgeon.

4. The axillary nerve is seldom injured alonein gunshot wounds, though it may be damaged

September, I946 245P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

POST-GRADUATE MEDICAL JOURNALin "crutch palsy"; deltoid paralysis (with lossof external rotations too) is the most prominentsign in upper brachial plexus forms of birthpalsy. Treatment lies in the field of the orthopaedicsurgeon.

5. Paralysis of the musculocutaneous nerve gives.relatively little disability. Flexion of the elbowcan be performed efficiently if part of brachialishas a radial supply; even if this is lacking, goodflexion of the pronated forearm is possible bybrachio-radialis; the supinated forearm can beflexed, if biceps and brachialis are out, only if theflexors of the forearm are exceptionally ablesubstitutes.

6. The medial cutaneous nerve of the forearmsupplies an area of skin whose anaesthesia is not ofgreat importance, but if it is divided together withmedian or ulnar its repair may be attempted whenthe larger nerve is explored. This nerve is tempt-ingly convenient for service as a graft for medianor ulnar, but its small calibre makes it quiteineffective for this purpose.

7. The radial nerve. This nerve may be injuredin the arm by a fracture of the humerus, by tourni-quet or plaster pressure, or by pressure on the edgeof an operating table. It may be damaged at theelbow in fracture of the neck or head of the radiusor in dislocation of the radius. It may be damagedat any level in an open wound. The posteriorinterosseous nerve may suffer an irritation neuritisjust at its entry into the supinator muscle, perhapsfrom repeated pronation and supination, as inGuillain and Courtellemont's orchestra conductor,35perhaps because of its close relation here to thebicipito-radial and interosseous bursae;I25 thethumb extensors suffer oftenest, but all wrist andfinger extensors may be affected.34, 6I If the nerveis injured below the middle of the arm, not onlytriceps but brachio-radialis escapes, and sometimesthe branches to extensor carpi radialis and eventhose to the finger extensors leave the parent trunkat a surprisingly high level. In injuries above theelbow, supination of the flexed forearm is lost, butbiceps continues to supinate the fully-extended fore-arm. The sensory paralysis varies with the levelof the injury. In injuries high in the arm, thelateral cutaneous nerve of the arm is involved, andthere is sensory loss also from the radial side of thethumb over the dorsum of the second and thirdmetacarpals, but even in high injuries free anasto-mosis with musculo-cutaneous and medial cutaneousnerves may prevent any sensory loss at all. Inlower injuries there is never much sensory loss, forthe musculo-cutaneous and lateral cutaneous nervesof the forearm virtually replace the sensory func-tion of the radial.When testing for extensor paralysis, each indi-

vidual extensor tendon should be tested, with the

metacarpo-phalangeal joints in the flexed position,so that the interossei, which can extend the extendedfingers, are thrown out of action. Extension of thethumb may be mimicked by abductor pollicisbrevis, and sometimes, especially if the extensor ten-dons have been shortened by splinting, the patientcan extend the fingers by flexing the wrist, justas he can dorsiflex the wrist by flexing the fingers(Fig. i). The radial nerve recovers after suturebetter than any other nerve in the body, perhapsbecause it is so purely motor that crossing of axonsis of no importance. After a suture in the arm,recovery may be expected in nine to twelve months.

8. The median nerve may be injured, not onlyby open wounds, but by fracture of the lowerhumerus, or even of both bones of the forearm; inthe latter fracture its volar interosseous branchmay suffer alone. The median nerve may beinjured also in dislocation of the lunate bone, andit may follow a Colles's fracture. It may be com-pressed by bandages or splints, or by a tourniquet,and an occupational paralysis of the median mayoccur in the users of trowels, scrubbing brushesand planes. An ischaemic paralysis may occurfrom injury to the volar interosseous artery, anddamage to this vessel should be avoided in repairinga median lesion. 99 In paralysis of the mediannerve, the only digits which lose all flexion are thethumb and index. Pronation is feeble, and com-pleted only by the weight of the forearm. In highlesions, the wrist flexes to the ulnar side. Theparalyses of abductor, opponens, and half of flexorbrevis pollicis are difficult to test individually,Io8but the first metacarpal is usually prominent, andit is difficult to raise the thumb forwards at rightangles to the plane of the palm (abductor), or tobring the thumb against resistance to touch the tipof the little finger (opponens). The former of thesetests is the most specific test for a median motorlesion,89 for the short abductor is the only thenarmuscle invariably innervated by the median,so butin its performance the muscle belly of abductorbrevis should be palpated to see whether it con-tracts, for if opponens has a substantial ulnarinnervation, it may, acting in concert with abduc-tor longus, carry the thumb well forward from theplane of the palm.1o9 Tricks which may be ac-quired to cover a median paralysis are: flexion ofthe wrist by abductor pollicis longus (continuingeven when median and ulnar nerves are bothout), and unusually good flexion of the fingers bythe ulnar part of profundus.89The sensory area of the median nerve is also

subject to considerable variation (Figs. 2, 3), butsensation to the tip of the index finger is exclusivelymedian. The median nerve is said to be subjectmore commonly than other nerves to neuromaformation after suture, and a case is recorded in




j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/


FIG. 13.-Repair of high sciatic lesion. Gluteus maximus elevated; line ofincision of piriformis tendon.

(Reproduced by kind permission of "Surgery, Gynecology and Obstetrics.")

IAN AIRD, CH.M., F.R.C.S.(ED.)

..4

C.-he^e^ .lkE, Y



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/


C Slhc f ,

FIG. I4.-Repair of high sciatic lesion. Gluteus maximus elevated andpiriformis divided. Note relaxation of branches of inferior glutealvessels, and full exposure of proximal segment of nerves.


IAN AIRD, CH.M., F.R.C.S.(ED.)P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

SURGERY OF PERIPHERAL NERVE INJURY .

which such a neuroma developed spontaneouslytwelve years after operation.26 The maximum gapwhich can be closed successfully by suture is 7 cm.5IThe chief disability in median paralysis is sensoryand this is of importance in deciding operativetreatment. Motor paralysis may be disregarded,provided some sensory recovery is obtained. Themost serious part of the motor disability is the lossof opponens.67 Recovery is rather more rapid,length for length, in the median nerve than in theulnar.67 The median nerve, like the posteriortibial, is subject to hyper-excitability and causalgiaafter injury.

9. The ulnar nerve may be injured by woundsand by fractures in the region of the elbow, and itsuffers, too, from tardy palsy if a valgus deformity,the result of a supracondylar fracture, producesirritation of the nerve behind the elbow. In con-tradistinction to division of the median nerve,division of the ulnar is imiportant by reason of 'themuscular rather than of the sensory paralysiswhich it produces. The fine movements of thehand are lost, and hand function is seriouslydamaged. The loss of the inner half of flexordigitorum profundis is not of great moment; theimportant paralysis is the loss of the small musclesin the hand. The typical main-en-griffe producedby loss of the interossei and lumbricals (affectingthe inner two fingers more than the outer, sincethe latter conserve a greater lumbrical action),gives extension at the metacarpal-phalangeal jointsand flexion of the distal finger joints. The pre-dominance of this deformity in annularis andminimus is lost to some extent if the lesion is abovethe ulnar contribution to profundus; hence, thehigher the lesion the less the deformity. There ishollowing on the dorsum between the extensortendons (from loss of the interosseous muscles) andthe hypothenar eminence is flattened. There maybe a noticeable flattening, too, on the dorsal aspectof the first metacarpal space, from loss of the firstdorsal interosseous muscle. The belly of thismuscle may be palpated when the index finger isabducted against resistance. The little finger, inulnar paralysis, is held slightly abducted, and can-not be adducted against the ring finger. Weaknessof the interossei can be determined by interlockingthe patient's fingers with the examiner's, and theloss of grip between the fingers is obvious. Thepatient, too, is unable to grip a sheet of paperbetween the extended digits. Trick movementshere are:

(I) Unusually effective flexion of the little fingerby a wide median innervation of profundus,

(2) Abduction of the fingers by the long extensortendons, and,

(3) Abduction of the fingers by the flexors of thefingers. 89

There is surprisingly little disability often after anulnar lesion if its muscular territory is small inextent.95 As in the case of median lesions, theprocaine block method of Highet is invaluable fordetermining anomalies of distribution.For the repair of injuries, the ulnar nerve offers

a greater prospect of the relief of wide gaps thandoes any other nerve in the body, for it can readilybe transplanted to the front of the elbow, and aftersuch transplantation a gap of II cm. can some-times be closed without tension. In exposure of aninjured ulnar nerve, a suitable incision (Lear-month62) is made along the line of the nerve (Fig.6), which may be exposed first in the groove behindthe internal condyle (Fig. 7). The nerve may thenbe followed upwards into the arm, and downwardsinto the forearm between the heads of flexor carpiulnaris. It usually gives off two articular branchesas it lies in the groove, and these may be sacrificedif transposition is required; indeed, transposition ofa sutured ulnar nerve is always desirable, since it isless exposed to the risk of strangulation by scartissue in the intermuscular plane in front of theelbow than in the bony canal behind it. In thedistal part of the groove, usually, the upper branchto flexor carpi ulnaris leaves the main trunk, andit may sometimes have to be sacrificed to mobilisethe nerve adequately upwards, though usually itmay be liberated from the parent trunk for severalcentimetres above its apparent point of exit. Thenerve can, however, be mobilised downwards as faras its contribution to flexor digitorum profundus.If the nerve has been divided above or below thelevel of the bony canal, it is unnecessary to detachcompletely the flexor origin from the medial epi-condyle. A forceps is merely passed deep to it,and a cotton swab is withdrawn deep to the flexororigin (Fig. 8), to make a tunnel through which theulnar segment of the divided nerve may be drawnby a stitch transfixing its bulb (Fig. 9). If thenerve division is precisely at the centre of the bonycanal, the flexors must be detached, as they aredetached in a transposition of the intact nerve. Inits final anterior position, the nerve should lie inthe same intermuscular plane as the median. Par-ticular care is required to make sure that themedial intermuscular septum is divided down tothe epicondylic ridge, and that the fascia of flexorcarpi ulnaris is divided down to the ulna, as eitherof these two fascial bands may produce a post-operative irritation of the nerve trunk. Trans-position of the ulnar nerve may permit the repairof a gap of the order of II cm.5I

IO. The sciatic nerve. The sciatic nerve maybe damaged by open wounds, by prolonged pressureon the back of the thigh, by tourniquet constric-tion, or by the pressure of an aneurysm of thecompanion artery of sciatic nerve3'; injury has

Se.ptember, r9466 249P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

250 POST-GRADUATE MEDICAL JOURNAL September, 1946occurred also as a result of dislocation of the hip orrepair of a congenital dislocation, pressure of thefoetal head on the lumbo-sacral plexus, in parturi-tion, and even an awkward jump or fall. In thesciatic palsy which has rarely followed a simplespringing jump, intraneural bleeding from the com-panion artery of the sciatic nerve has been blamed.

For repair of the divided sciatic nerve, the patientlies prone with a pillow under the knee to extendthe hip, and with the knees flexed. In the back ofthe thigh the nerve may be conveniently exposedby a median posterior incision, with avoidance ofinjury to the posterior cutaneous nerve. In theapproach to a high lesion, a question-mark in-cision47, 72 around the borders of the gluteusmaximus muscle, with its vertical limb down themiddle of the back of the thigh (Fig. II) allows thefascia lata and the insertion of gluteus maximus tobe divided (Fig. 12) and permits that muscle to bethrown medially towards its origin, like a lid, toexpose the sciatic nerve at the lower border ofpiriformis (Fig. I3). If, before gluteus maximusis so elevated, the tendon of the piriformis isdivided, piriformis can be elevated together withgluteus maximus.3 This relaxes the tension on thebranches of the inferior gluteal veins, as they runforwards under the edge of piriformis, and lessensthe risk that one of these may be torn from itsparent trunk (Fig. I4). If a torn gluteal vesselretracts into the pelvis, it may be difficult to applya haemostat safely to it. This form of exposurealso permits a quite high suture of the sciatic nerveto be performed at the very point of its emergencefrom the pelvis. When the wound is closed, caremust be taken to avoid including the posteriorcutaneous nerve in a stitch; this can be ensured byoverlapping the edges of the deep fascia (Fig. I5).The sciatic nerve, more perhaps than any other,

lends itself to a partial repair, for the medial andlateral poplital branches are separate within themain trunk usually for a considerable distance, andcan be treated as separate nerves. The maximumgap in a sciatic nerve which can be repaired withouttension is in the neighbourhood of 8 cm., but amuch smaller gap may be impossible to repair ifthe lesion is a high one. If the gap in the sciaticnerve is too wide to suture without dangeroustension, it is legitimate to use a segment of thelateral popliteal nerve as a graft to bridge thedefect in the medial popliteal. The muscular palsywhich is the most serious disability of a lateralpopliteal lesion can be corrected by orthopaedicmeasures, while the sensory paralysis of a medialpopliteal lesion is a permanent and serious dis-ability if not repaired. The position of the patientduring the operation must be maintained for threeweeks at least after suture of the sciatic nerve.This can conveniently be done by a complete hip

spica, or the flexed knee, in a plaster extendingfrom upper thigh to ankle, may be slung from aBalkan beam to maintain extension of the hip.The exaggerated position is tiresome for the patient,and is not always well borne.

II. The lateral popliteal nerve may be injured byopen wounds, by fractures in the region of the neckof the fibula, by direct blows on the neck of thefibula, and by traction injuries. Platt79 describedoriginally the closed injury of the lateral poplitealnerve which occurs from a severe adduction strainon the knee with rupture of the fibular collateralligament; sometimes the tendon of biceps too isruptured, or the head of the fibula avulsed; theiliotibial tract and the cruciate ligaments may alsosuffer damage. In some cases, the nerve is actuallyruptured, while in some the continuity is preserved.The injury was further studied by Highet andHolmess5; Leclerq64 had previously described casesof the injury, but had not recognised the mechan-ism by which it was produced. In one of Platt'spatients, the nerve suffered neurapraxia, which re-covered completely in four weeks, but in all othercases there was a complete lesion with extensivedamage, and spontaneous recovery did not occur inthem even when nerve fibres were subsequentlyshown to have penetrated the scar. The lateralpopliteal nerve may proceed to satisfactory recoveryafter repair in the thigh or in the popliteal fossa, andin this area gaps of as much as 9 cm. have beensuccessfully sutured, but adequate recovery isexceptional if the nerve is sutured beyond the pointof its departure from the fossa; it is usually a savingof time to the patient to proceed immediately toorthopaedic correction of "dropfoot" deformity.

12. The medial popliteal and posterior tibialnerves should be repaired by suture if possible, or,if not, by a nerve graft, for any sacrifice is justifiedto return sensation in the sole of the foot. The.motor disability after a medial popliteal lesion isnot great; paralysis of the short muscles of thefoot results in a clawing deformity which can be, tosome extent, corrected by a pad under the sole.The maximum gap which can be corrected bysuture in the case of this nerve is of the order of6 cm.5I

BIBLIOGRAPHY

(For illustrations of the methods of testing the voluntary power ofvarious muscles, reference should be made to the pamphlet "Aids to theInvestigation of Peripheral Nerve Injuries" (69), and to Haymaker andWoodhall's "Peripheral Nerve Injuries" (46). To avoid unmanageablelength, reference has been made mainly to papers published during thepast seven years.)I. ABERCROMBIE, M., and JOHNSON, M. L.,"Outwandering of Cells

in Tissue Cultures of Nerves undergoing Wallerian Degeneration,"J. Exper. Biol., I942, 19, 266-283.

2. ADRIAN, E. D., "Interpretation of the Electro-myogram," Lancet,1925, 1, I229-I233 and I282-I286.

3. AIRD, I., "The Surgical Approach to High Lesions of the SciaticNerve," Surg. Gynec. Obst. (In Press).

4. ANGLESIO, B., and BRUNI, A., "Sugli esiti di Lesioni Traumatichedei Nervi Periferici," Berichte iiber den VIII Kongress fur Unfall-medizin und Berufskrankheiten, Frankfurt, I938, 2, 440-448.



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/


FIG. 15.--Repair of high sciatic lesion. Closure of musculo-fascial layer.Note overlapping of fascia lata in thigh so that posterior cutaneousnerve. in one edge of fascia, escapes injury.


IAN AIRD, CH.M., F.R.C.S.(ED).P



j.bmj.com

/P


j.22.251.225 on 1 Septem

ber 1946. Dow

nloaded from

http://pmj.bmj.com/

September. 1946 SURGERY OF PERIPHERAL NERVE INJURY 2535. BALLANCE, C., and DUEL, A. B., "Operative Treatment of Facial

Palsy by Introduction of Nerve Grafts into Fallopian Canal and byother Intra-temporal Methods," Arch. Otolaryngol., 1932, 15, 1-70.6. BANKART, B., "Discussion on Injuries to the Brachial Plexus,"

Proc. Roy. Soc. Med., 1930, 23, 1286.7. BAUDENS, J. B. L., "Clinique des Plaies d'Armes a Feu," 8vo, pp. 6ioand xv; Paris; 1836, J. B. Bailliere.8. BAUWENS, P., "Electro-Diagnosis and Electro-therapy in Peri-

pheral Nerve Lesions," Proc. Roy. Soc. Med., 194I, 34, 459-468.9. BENTLEY, F. H., and HILL, M., "The Possibilities of Nerve Graft-

ing," Brit. Med. J., 1940, 2, 252-353.Io. BISHOP, G. H., "Responses to Electrical Stimulation of Single

Sensory Units of Skin," J. Neurophysiol., I943, 6, 361-382.I . BLACKWOOD, W., "A Pathologist Looks at Ischaemia," Edin. Med.

J., 1944, 51, I31-143.12. BOWDEN, R. E. M., "Changes in Human Voluntary Muscle in

Denervation and Re-innervation," Brit. Med. J., 1945, 2, 487-488.13. BOWDEN, R. E. M., and GUTMANN, E., "Denervation and Re-

innervation of Human Voluntary Muscle," Brain, 1944, 67, 273-313.

14. BOWDEN, R. E. M., and GUTMANN, E., "Clinical Value of MuscleBiopsies," Lancet, 1945, 2, 768-77I.

15. BUNNELL, S., and BOYES, J. H., "Nerve Grafts," Amer. J. Surg.,1939, 44, 64-75.

x6. CAIRNS, H., and YOUNG, J. Z., "Treatment of Gunshot Wounds ofPeripheral Nerves," Lancet, 1940, 2, 123-126.

I7. y CAJAL, R., "Degeneration and Regeneration of Nervous System,"Translated and Edited by Raoul M. May, 2 vol., pp 769 and xx, viii.8vo, London, 1928, Humphrey Milford.

18. COLEMAN, C. C., "Surgical Treatment of Peripheral Nerve Injuries,"Surg. Gynec. Obstet., I944, 78, 113-124.

19. DAVIS, L., and CLEVELAND, D. A., "Experimental Studies inNerve Transplants," Ann. Surg., 1934, 99, 271-283.

20. DEJARDIN, L., and JONCKHEERE, F., "La Chirurgie reparatricedes nerfs peripheriques." Berichte uber den VIII InternationalenKongress fur Unfallmedizin und Berufskrankheiten, Frankfurt,1938, 2, 361-401.

21. DENNY-BROWN, D. E., and BRENNER, C., "Lesion in Peri-pheral Nerve Resulting from Compression by Spring Clip," Arch.Neurol. Psvchiat., 1944, 52, I-I9.

22. DENNY-BROWN, D. E., and PENNYBACKER, J. B., "Fibrilla-tion and Fasciculation in Voluntary Muscle." Brain, 1938, 61,311-334.

23. DOGLIOTTI, A. M., "I Processi riparativi delle Lesioni dei NerviPeriferici (excluso il simpatico)," Berichte iber den VIII Kongressfur Unfallmedizin und Berufskrankheiten, Frankfurt, 1938, 2, 402-418.

24. DUEL, A. B., "History and Development of Surgical Treatment ofFacial Palsy," Surg. Gynec. Obstet., 1933, 56, 382-390.

25. ECCLES, J. C.,"Changesin Muscle Produced by Nerve Degeneration,"Med. J., Austral., 1941, 1, 573-575.

26. EHALT, W., "Behandlungsergebnisse von Verletzungen peripherenNerven," Berichte uber den VIII Kongress fur Unfallmedizin undBerufskrankheiten, Frankfurt, 1938, 2, 419-440.

27. ELKIN, D. C., and WOODHALL, B., "Combined Vascular andNerve Injuries of Warfare," Ann. Surg., 1944, 119, 411-431.

28. FISCHER, E., "The Effect of Faradic and Galvanic Stimulation uponthe Course of Atrophy in Denervated Skeletal Muscles," Amer. J.Physiol., 1939, 127, 605-619.

29. FOERSTER, 0., "The Dermatomes in Man," Brain, 1933, 56, I-39.30. FOERSTER, O., "Degeneration und Regeneration des Peripheren

Nervensystems," Deutsch. Zeitschr. f. Nervenheilk., 1930, 115,248-314.

3I. FOERSTER, 0., "Die Verletzungen der Peripheren Nerven," Berichteiiber den VIII International Kongressfiir Unfallmedizin und Berufs-krankheiten, Frankfurt, 1938, 2, 325-354.

32. FULTON, J. F., "Physiology of the Nervous System." 2nd. Ed. Ox-ford Univ. Press, London, 1943, pp. 614 with I12 illustrations.

33. GOSSET, A., and BERTRAND, I., "La moelle 6piniere, utiliseecomme greffon heteroplastique dans les blessures des nerfs peri-pheriques," J. Chirurg. (Par.), 1938, 5, 481-505.

34. GRIGORESCO, D., and IORDANESCO, G., "Un cas rare de Para-lysie partielle du Nerf radial," Rev. Neurol., 1931, 2, 102-104.

35. GUILLAIN, G., and COURTELLEMONT, "L'Action du musclecourt supinateur dans la paralysie du nerf radial," Presse Mdd.,I905, 1, 50-52.

36. GUTMANN, E., and GUTTMANN, L., "Effect of Electrotherapy ondenervated Muscles in Rabbits," Lancet, 1942, 1, I69-I70.

37. GUTMANN, E., GUTTMANN, L., MEDAWAR, P. B., and YOUNG,J. Z., "The Rate of Regeneration of Nerve," J. Exper. Biol., 1942,19, 14-44.

38. GUTMANN, E., "Factors affecting Recovery of Motor Functionafter Nerve Lesions," J. Neurol. Psychiat., 1942, 5, 81-95.

39. GUTMANN, E., and SANDERS, F. K., "Functional Recovery fol-lowing Nerve Grafts and other Types of Nerve Bridge," Brain,1942, 65, 373-408.

40. GUTMANN, E., and SANDERS, F. K., "Recovery of Fibre Numbersand Diameters in Regeneration of Peripheral Nerves," J. Physiol.,I943, 101, 489-518.

4I. GUTMANN, E., and GUTTMANN, L., "Effect of Galvanic Exerciseon Denervated and Re-innervated Muscles of Rabbit," Journ.Neurol. and Neurosurg. Psych., I944, 7, 7-17.

42. GUTTMANN, L.; "Topographic Studies of Disturbances of SweatSecretion after complete Lesions of Peripheral Nerves," J. Neurol.Psychiat., I940, 3, 197-210.

43. GUTTMANN, L., "Experimental Study on Nerve Suture with VariousSuture Materials," Brit. J. Surg., 1943, 30, 370-375.

44. HARRIS, W., "Discussion on Injuries to the Brachial Plexus," Proc.Roy. Soc. Med., 1930, 23, I282-1285.

45. HARVEY, A. M., and KUFFLER, S. W., "Motor Nerve Functionwith Lesions of the Peripheral Nerves. A Quantitative Study,"Arch. Neurol. Psychiat., i944, 52, 317-322.

46. HAYMAKER, W., and WOODHALL, B., "Peripheral Nerve In-juries," W. B. Saunders and Coy., Philadelphia, 1945, 8vo, 227 &xiv, 225 Illustrations.

47. HENRY, A. K., "Extensile Exposure applied to Limb Surgery," 8vo,180 & viii, E. & S. Livingstone, Edinburgh, I945.48. HIGHET, W. B., "Procaine Nerve Block in Investigation of Peri-

pheral Nerve Injuries," J. Neurol. Psychiat., 1942, 5, IOi-ii6.49. HIGHET, W. B., "Splintage of Peripheral Nerve Injuries," Lancet,

I942, 1, 555-558.50. HIGHET, W. B., "Innervation and Function of Thenar Muscles,"

Lancet, 1943, 1, 227-230.51. HIGHET, W. B., and HOLMES, W., "Traction Injuries to Lateral

Popliteal Nerve and Traction Injuries to Peripheral Nerves afterSuture," Brit. J. Surg., 1943, 30, 2I2-233.

52. HIGHET, W. B., and SANDERS, F. K., "Effects of StretchingNerves after Suture," Ibid., 1943, 30, 355-369.

53. HOFMANN, E., "Zur Indikationsstellung operativer Eingriffe nachSchussverletzungen peripherer Nerven," Deutsch. Militdrartf,1943, 8, 217-223.

54. HOLMES, W., and MEDAWAR, P. B., "Local Application ofSulphonamide to Peripheral Nerves," Lancet, 1942, 2, 334-335.

55. HOLMES, W., and YOUNG, J. Z., "Nerve Regeneration after Im-mediate and Delayed Suture," J. Anat., 1942, 77, 63-96.

56. HYMAN, I., and BESWICK, W. F., "Measurement of Skin Resis-tance in Peripheral Nerve Injuries," War Medicine, Chicago, 1945,8, 258-260.

57. JACKSON, E. C. S., and SEDDON, H. J., "Influence on GalvanicStimulation on Muscle Atrophy resulting from Denervation," Brit.Med. J., 1945, 2, 485-486.

58. JEFFERSON, G., "Discussion on Injuries to the Brachial Plexus,"Proc. Roy. Soc. Med., 1930, 23, 1282-1285.

59. KLEMME, R. M., WOOLSEY, R. D., and de REZENDE, N. T.,"Autopsy Nerve Grafts in Peripheral Nerve Surgery. ClinicalApplication," J. Amer. Med. Assoc., 1943, 123, 393-396.

60. LEARMONTH, J. R., "Personal Experience of Exploration andRe-exploration of Injured Nerves," Proc. Roy. Soc. Med., 1944, 37,553-555.

6i. LEARMONTH, J. R., and WOLTMANN, H. W., "ProgressiveParalysis of the Nervus Interosseus Dorsalis," Brain, 1934, 57,25-31.

62. LEARMONTH, J. R., "A Technique for Transplanting the UlnarNerve," Surg. Gynec. Obstet., 1942, 75, 792-793.

63. LEARMONTH, J. R., and WALLACE, A. B., "Certain PlasticProblems in the Surgery of Peripheral Nerves," Ibid., 1943, 76,106-O19.

64. LECLERCQ, J., and MULLER, M., "Lesions du nerf sciatiquepoplite externe par traumatismes fermis," Berichte uber den VIIIKongress fur Unfallmedizin und Berusfkrankheiten, Frankfurt,1938, 2, 402-418.

65. LEWIN, W., "Pressure Palsy in the Paralysed Limb," Lancet, 1943,2, 756-758.

66. LIPCHINA, L. P., "Intra-neural Processes after Trauma to Peri-pheral Nerves without Damage to the Epineurium," ProblemyNyerokhirurgiy, Moscow, 1943, pp. 19-28.

67. MARBLE, H. C., HAMLIN, E., and WATKINS, A. L., "Regenera-tion in the Ulnar, Median, and Radial Nerves," Amer. J. Surg.,I942, 55, 274-294.

68. M.R.C. Nerve Injuries Committee, "Note on the Assessment of Re-covery in Peripheral Nerve Injuries" (W. B. HIGHET).

69. M.R.C. Nerve Injuries Committee, "Aids to the Investigation of Peri-pheral Nerve Injuries," 1943. (Prepared by Staff of Departmentof Surgery, University of Edinburgh.) pp. 48, with 78 illustrations.

70. MENNELL, J., "Massage, Movements, and Exercises in the Treat-ment of Nerve Suture and Repair," Brit. J. Phys. Med., 1942, 5,40-47.

71. MICHAEL, P., and ABBOTT, W., "The Use of Human Fibrinogenin Reconstructive Surgery," J. Amer. Med. Assoc., 1943, 123, 279.

72. NAFFZIGER, H. C., and NORCROSS, N. C., "Surgical Approach toLesions of Upper Sciatic Nerve and Posterior Aspect of Hip Joint,"Surgery, 1942, 12, 929-932.

73. NAGEOTTE, J., "Sur la greffe des tissus morts," Compt. rend. Soc.de Biol., Par., 1917, 53, 459-470.

74. NATHAN, P. W., and RENNIE, A. M., "Value of Tinel's Sign,"Lancet, 1946, 1, 6io-611.

75. NORCROSS, N. C., "Early Repair of Neural Wounds with PenicillinTherapy," Arch. Surg., 1945, 50, 67-68.

76. PARKES, A. R., "Traumatic Ischaemia of Peripheral Nerves withsome Observations on Volkmann's Ischaemic Contracture," Brit.J. Surg., I945, 32, 403-414.

77. PHILIPPIDES, D., "Die Priifung der Vasomotor funktion beiPeripheren Nerven Lasionen," Der Chirurg., I942, 14, 385-389.

78. PLATT, H., "War Injuries of Peripheral Nerves," Post-Graduate Med.J., 1940, 16, 256-259.

79. PLATT, H., "Traction Lesions of the External Popliteal Nerve,"Lancet, 1940, 2, 612-614.

80. POLLOCK, L. J., "Overlap of So-called Protopathic Sensibility asSeen in Peripheral Nerve Lesions," Arch. Neurol. Psychiat., 1919,2, 667-700.



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/

254 POST-GRADUATE MEDICAL JOURNAL September, i9468i. POLLOCK, L. J., "Nerve Overlap as related to the relatively early

Rebirth of Pain Sense following Injury to the Peripheral Nerves,"J. Compar. Neurol., 1920, 32, 357-378.

82. POLLOCK, L. J., and DAVIS, L., "Peripheral Nerve Injuries,"N.Y., 1933, Hoeber, pp. 678 (312 illustrations).

83. POLLOCK, L. J., "Evaluation of Incapacity Produced by Injuriesof the Peripheral Nerves," Surg. Gynec. Obstet., 1941, 73, 462-47I.

84. RICHARDS, R. L., "Vasomotor Disturbances in the Hand afterInjuries of Peripheral Nerves," Edin. Med. J., 1943, 50, 449-468.

85. RICHARDS, R. L., "The Assessment of Sensory Denervation afterInjuries of Peripheral Nerves," Ibid., I946, 53, Iio.

86. RICHTER, C, P., and KATZ, D. T., "Peripheral Nerve Injuriesdetermined by Electrical Skin Resistance Method; Ulnar Nerve,"J. Amer. Med. Assoc., I943; 122, 648-651.

87. RITCHIE, A. E., "The Electrical Diagnosis of Peripheral NerveInjury," Brain, 1944, 67, 314-430.

88. ROTTGEN, "Zur Frage der Kontrastdarstellung der VerletzungenPeripherer Nerven," Der Nervenarzt, I943, 16, 244-54.

89. RUSSELL, W. R., and HARRINGTON, A. B., "Early Diagnosis ofPeripheral Nerve Injuries in Battle Casualties," Brit. Med. J.,I944, 2, 4-8.

9o. SANDERS, E. K., "The Repair of Large Gaps in the PeripheralNerves," Brain, 1942, 65, 281-337.

91. SANDERS, F. K., and YOUNG, J. Z., "The Degeneration and Re-innervation of Grafted Nerves," J. Anat., 1942, 76, 143-166.

92. SCHAFER, E. SHARPEY, "Effects of Denervation of CutaneousArea," Quart. J. Exp. Physiol., 1928, 19, 85-107.

93. SCHAFER, E. S., "Recovery after Severance of Cutaneous Nerves,"Brain, 1927, 50, 538-547.

94. SEDDON, H. J., "Classification of Nerve Injuries," Brit. Med. J.,I942, 2, 237-239.

95. SEDDON, H. J., and MEDAWAR, P. B., "Fibrin Suture of HumanNerves," Lancet, 1942, 2, 87-88.

96. SEDDON, H. J., YOUNG, J. Z., and HOLMES, W., "HistologicalCondition of Nerve Autograft in Man," Brit. J. Surg., 1942, 29,378-384.

97. SEDDON, H. J., "Three Types of Nerve Injury," Brain, 1943, 66,237-288.

98. SEDDON, H. J., MEDAWAR, P. B., and SMITH, H., "Rate ofRegeneration of Peripheral Nerves in Man," J. Physiol., I943,102, 191-215.

99. SEDDON, H. J., and HOLMES, W., "Ischaemic Damage in thePeripheral Stump of a Divided Nerve," Brit. J. Surg., I945, 32,389-391.

0oo. SHAPIRO, K. M., "The Significance of Chronaximetry in theDiagnosis of Gunshot Wounds of Peripheral Nerves," ProblemyNyerokhirurgiy, Moscow, 1943, pp. 28-37.

Ioi. SILVERMAN, J. J., and POWELL, V. E., "A Simple Technic forOutlining the Sweat Pattern," War Medicine, Chicago, 1945, 7,178-180.

102. SOLANDT, D. Y., and MAGLADERY, J. W., "Relation of Atrophyto Fibrillation in Denervated Muscle," Brain, 1940, 63, 255-263.

103. SPEIDEL, C. C., "Studies of Living Nerves, III. Phenomena ofNerve Irritation and Recovery, Degeneration and Repair," J.Comp. Neurol., 1935, 61, i-8o.

104. SPERRY, R. W., "Functional Results of Crossing Sensory Nervesin the Rat," Ibid., I943, 78, 59-90.

205. SPURLING, R. G., "Peripheral Nerve Injuries in European Theatreof Operations. Management with special Reference to EarlyNerve Surgery," J. Amer. Med. Assoc., I945, 129, 011-1014.

io6. SPURLING, R. G., and WOODHALL, B., "Experiences with EarlyNerve Surgery in Peripheral Nerve Injuries," Ann. Surg., 1946,123, 731-748.

107. SULLIVAN, W. E., and MORTENSEN, O. A., "Visualisation ofMovement of Brominised Oil along Peripheral Nerves," Anat.Rec., I934, 39, 493-50I.

io8. SUNDERLAND, S., "The Significance of Hypothenar Elevation inMovement of Opposition of the Thumb," Austral. & New Zeal. J.Surg., 1944, 13, I57-I59.

o09. SUNDERLAND, S., "Voluntary Movements and the DeceptiveAction of Muscles in Peripheral Nerve Lesions," Ibid., I944, 13,I60-184.

i1o. TARLOV, I. M., and BENJAMIN, B., "Plasma Clot and Silk Sutureof Nerves; Experimental Study of Comparative Tissue Reaction,"Surg. Gynec. Obstet., 1943, 76, 366-374.III. TARLOV, I. M., DENSLOW, C., SWARZ, S., and PINELES, D.,"Plasma Clot Suture of Nerves; Experimental Technique," 1943,47, 44-58.

112. TINEL, J., "Le Signe du fourmillement dans les lesions des nerfsperipheriques," Presse Mdd., 1915, 23, 385-386.

113. TINEL, J., "Nerve Wounds; Symptomatology of Peripheral NerveLesions Caused by War Wounds." Translated by F. Rothwell.Rev. and edited by C. A. Joll. London, Bailliere, Tindall & Cox,1917.

I14. TONNIS, W., and GOTZE, W., "Zur operativen Behandlung derSchussverletzungen der peripheren Nerven und ihre Erfolgsaussichten," Deutsch. Militarartz, 1942, 7, 245-253.

115. TONNIS, W., "Wann sollen die Schussverletzungen der peripherenNerven operiert werden?" Ibid., 1944, 9, 175-I77.116. TROTTER, W., and DAVIES, H. M., "Experimental Studies in theInnervation of the Skin," J. Physiol., I909, 38, 134-246.117. TROTTER, W., and DAVIES, H. M., "The Peculiarities of Sensi-bility found in Cutaneous Areas supplied byRegenerating Nerves,"Journ. fir Psychol. und Neurol., 1913, 20, 102-150.118. VERNET, M., "Syndrome du Tron dechir6 post6rieur (Paralysie desnerfs glosso-pharyngien-pneumogastrique-spinal), Rev. Neurol.Paris, 1918, 25, I17-148.

19. VILLARET, M., "Le syndrome de l'espace retro-parotidien pos-terieur," Presse Mdd., I917, 23, 430-431.

120. WALSHE, E. M. R., "Anatomy and Physiology of CutaneousSensibility," Brain, 1942, 65, 48-112.

12I. WEDDELL, G., and GLEES, P., "The Early Stages in the De-generation of Cutaneous Nerve Fibres," J. Anat., 1941, 76, 65-93.

122. WEDDELL, G., "Axon Regeneration in Cutaneous Nerve Plexuses,"Ibid., I942, 77, 49-62.

123. WEDDELL, G., GUTTMANN, L., and GUTMANN, E., "TheLocal Extension of Nerve Fibres into Denervated Areas of Skin,"J. Neurol. Psychiat., I941, 4, 206-225.

124. WEDDELL, G., FEINSTEIN, B., PATTLE, R. C., "The ClinicalApplication of Electromyography," Lancet, 1943, 1, 236-239.

125. WEINBERGER, L. M., "Non-Traumatic Paralysis of Dorsal Inter-osseous Nerve," Surg. Gynec. Obstet., 1939, 69, 358-363.

126. WEISS, P., "Experimental Innervation of Muscles by Central Endsof Afferent Nerves with functional Tests," J. Comp. Neurol.,1935, 61, I35-I74.

127. WEISS, P., "Reunion of Stumps of Small Nerves by Tubulationinstead of Suture," Science, 1941, 93, 67-68.

128. WEISS, P., "Nerve Regeneration in the Rat following TubularSplicing of Severed Nerves," Arch. Surg.,'I943, 26, 525-547.

I29. WEISS, P., and TAYLOR, A. C., "Histomechanical Analysis ofNerve Reunion in Rat after Tubular Splicing," Ibid., I943, 47,4I9-447.

130. YOUNG, J. Z., and MEDAWAR, P. B., "Fibrin Suture of Peri-pheral Nerves; Measurement of the Rate of Regeneration,"Lancet, 1940, 2, 126-128.

I3I. YOUNG, J. Z., HOLMES, W., and SANDERS, F. K., "NerveRegeneration; Importance of the Peripheral Stump and theValue of Nerve Grafts," Lancet, 1940, 2, 128-130.

132. YOUNG, J. Z., "The Functional Repair of Nervous Tissue," Physiol.Rev., I942, 22, 3I8-374. (See this for physiology of repair.)

I33. YOUNG, J. Z., "The Basic Sciences in Surgery," Edin. Med. J.,1946, 52, 262-276.



http://pmj.bm

j.com/

Postgrad M


eptember 1946. D

ownloaded from

http://pmj.bmj.com/

surgery of peripheral nerve injury · knee or tearing of the roots of the brachial plexus in a...

Documents