6
Journal of Neurology, Neurosurgery, and Psychiatry 1991;54:505-510 5 Central motor pathways in patients with mirror movements T C Britton, B-U Meyer, R Benecke Abstract Central motor pathways were inves- tigated in three patients with congenital mirror movements using magnetic motor cortex stimulation. Response thresholds, amplitudes and latencies were normal. The projection of the cor- ticomotoneuronal pathways was assessed by placing the coil over the vertex and comparing the size of responses in the first dorsal interosseous (FDI) muscles evoked by clockwise and anticlockwise coil currents. In normal subjects, right FDI responses are larger with clockwise currents than with anticlockwise currents at the same stimulation strength and vice versa. In two out of three patients with congenital mirror movements, this sensitivity of response amplitude to coil current direction was reversed. The third patient with con- genital mirror movements and a fourth patient with acquired mirror move- ments had responses which were normally sensitive to current direction. These findings support the hypothesis that some cases of congenital mirror movements may be due to abnormal projection of corticomotoneuronal path- ways. Department of Neurology, University of Dusseldorf, Germany T C Britton* B-U Meyer R Benecke Correspondence to: Professor Benecke, Neurologische Klinik, Universitat Dusseldorf, Moorenstr. 5, 4000 Dusseldorf, Germany *MRC Human Movement and Balance Unit, Institute of Neurology, Queen Square, London, UK Received 8 December 1989 and in final revised form 2 August 1990. Accepted 10 August 1990 Mirror movements are a special type of associated movement in which voluntary movements performed by one part of the body, particularly the arm or hand, are involuntarily and symmetrically performed by the other side as a result of bilateral activation of homologous muscles.'2 Mirror movements occur in children during development but normally disappear with maturation. Persis- tence of mirror movements into adult life, termed congenital mirror movements, is con- sidered pathological and may be associated with other developmental abnormalities including the Klippel-Feil syndrome'4 and Kallmann's syndrome.5 Mirror movements can also appear with a variety of acquired neurological conditions.' 2 Anatomically, congenital mirror movements are often associated with midline fusion dis- orders affecting the motor pathways. Incom- plete pyramidal tract decussation has been reported in the necropsy examinations of two cases that had congenital mirror movements associated with the Klippel-Feil syndrome."4 Dysraphic defects at different levels within the central nervous system are thought to underlie congenital mirror movements associated with the syndrome of anosmia and hypophyseal dysfunction.5 Neurophysiological evidence for abnormal ipsilateral corticomotoneuronal projections has recently been reported in brief abstract form in two patients with congenital mirror movements' and in one patient with mirror movements associated with Klippel-Feil syn- drome7 using the technique of trans-cranial electrical motor cortex stimulation. This study on three patients with congenital mirrror movements reports that abnormal corticomotoneuronal projections can be demonstrated with the technique of magnetic motor cortex stimulation' in two of the three cases. However, the third patient with con- genital mirror movements and a patient with acquired mirror movements as a result of internal capsular infarction had normally projecting corticomotoneuronal pathways. These findings have implications for the pathophysiological significance of non- decussated pyramidal pathways in congenital mirror movements. Methods PATIENTS AND SUBJECTS Four patients with mirror movements were investigated. Three had had mirror movements since childhood. The fourth developed mirror movements one week following a left internal capsular infarction. Case 1 A 52 year old right handed man had noticed mirror movements of his hands since childhood. He was moderately disabled in fine motor tasks of the hands. Pregnancy and early development were otherwise normal. On examination there were non-suppressible mirror movements on finger or wrist movements of either side. Simultaneous sym- metrical flexion-extension movements of the wrists could be made normally, but the patient experienced considerable difficulty in produc- ing simultaneous flexion of one wrist and exten- sion of the other. Sequential opposition of the thumb to each of the fingers was slow. No mirror movements were observed in the legs. Muscle strength, tone and reflexes in the upper and lower limbs were normal. The neck was of normal length and a cervical spine radiograph was normal. Sense of smell was normal. Case 2 A 26 year old man, with a mild right sided hemiplegia since birth, had noticed involuntary movements of his left hand when 505 on June 6, 2021 by guest. Protected by copyright. http://jnnp.bmj.com/ J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.54.6.505 on 1 June 1991. Downloaded from

Central motor pathways in patients withmirror movements · JournalofNeurology,Neurosurgery, andPsychiatry 1991;54:505-510 5 Centralmotorpathwaysinpatientswithmirror movements TCBritton,

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

  • Journal of Neurology, Neurosurgery, and Psychiatry 1991;54:505-510 5

    Central motor pathways in patients with mirrormovements

    T C Britton, B-U Meyer, R Benecke

    AbstractCentral motor pathways were inves-tigated in three patients with congenitalmirror movements using magneticmotor cortex stimulation. Responsethresholds, amplitudes and latencieswere normal. The projection of the cor-ticomotoneuronal pathways was assessedby placing the coil over the vertex andcomparing the size of responses in thefirst dorsal interosseous (FDI) musclesevoked by clockwise and anticlockwisecoil currents. In normal subjects, rightFDI responses are larger with clockwisecurrents than with anticlockwisecurrents at the same stimulationstrength and vice versa. In two out ofthree patients with congenital mirrormovements, this sensitivity of responseamplitude to coil current direction wasreversed. The third patient with con-genital mirror movements and a fourthpatient with acquired mirror move-ments had responses which werenormally sensitive to current direction.These findings support the hypothesisthat some cases of congenital mirrormovements may be due to abnormalprojection of corticomotoneuronal path-ways.

    Department ofNeurology, Universityof Dusseldorf,GermanyT C Britton*B-U MeyerR BeneckeCorrespondence to:Professor Benecke,Neurologische Klinik,Universitat Dusseldorf,Moorenstr. 5, 4000Dusseldorf, Germany*MRC Human Movementand Balance Unit, Instituteof Neurology, Queen Square,London, UKReceived 8 December 1989and in final revised form2 August 1990.Accepted 10 August 1990

    Mirror movements are a special type ofassociated movement in which voluntarymovements performed by one part of thebody, particularly the arm or hand, areinvoluntarily and symmetrically performed bythe other side as a result of bilateral activationof homologous muscles.'2 Mirror movementsoccur in children during development butnormally disappear with maturation. Persis-tence of mirror movements into adult life,termed congenital mirror movements, is con-sidered pathological and may be associatedwith other developmental abnormalitiesincluding the Klippel-Feil syndrome'4 andKallmann's syndrome.5 Mirror movementscan also appear with a variety of acquiredneurological conditions.' 2

    Anatomically, congenital mirror movementsare often associated with midline fusion dis-orders affecting the motor pathways. Incom-plete pyramidal tract decussation has beenreported in the necropsy examinations of twocases that had congenital mirror movementsassociated with the Klippel-Feil syndrome."4Dysraphic defects at different levels within thecentral nervous system are thought to underlie

    congenital mirror movements associated withthe syndrome of anosmia and hypophysealdysfunction.5

    Neurophysiological evidence for abnormalipsilateral corticomotoneuronal projectionshas recently been reported in brief abstractform in two patients with congenital mirrormovements' and in one patient with mirrormovements associated with Klippel-Feil syn-drome7 using the technique of trans-cranialelectrical motor cortex stimulation. Thisstudy on three patients with congenitalmirrror movements reports that abnormalcorticomotoneuronal projections can bedemonstrated with the technique of magneticmotor cortex stimulation' in two of the threecases. However, the third patient with con-genital mirror movements and a patient withacquired mirror movements as a result ofinternal capsular infarction had normallyprojecting corticomotoneuronal pathways.These findings have implications for thepathophysiological significance of non-decussated pyramidal pathways in congenitalmirror movements.

    MethodsPATIENTS AND SUBJECTSFour patients with mirror movements wereinvestigated. Three had had mirror movementssince childhood. The fourth developed mirrormovements one week following a left internalcapsular infarction.

    Case 1 A 52 year old right handed man hadnoticed mirror movements of his hands sincechildhood. He was moderately disabled in finemotor tasks of the hands. Pregnancy and earlydevelopment were otherwise normal. Onexamination there were non-suppressiblemirror movements on finger or wristmovements of either side. Simultaneous sym-metrical flexion-extension movements of thewrists could be made normally, but the patientexperienced considerable difficulty in produc-ing simultaneous flexion ofone wrist and exten-sion of the other. Sequential opposition of thethumb to each of the fingers was slow. Nomirror movements were observed in the legs.Muscle strength, tone and reflexes in the upperand lower limbs were normal. The neck was ofnormal length and a cervical spine radiographwas normal. Sense of smell was normal.

    Case 2 A 26 year old man, with a mild rightsided hemiplegia since birth, had noticedinvoluntary movements of his left hand when

    505

    on June 6, 2021 by guest. Protected by copyright.

    http://jnnp.bmj.com

    /J N

    eurol Neurosurg P

    sychiatry: first published as 10.1136/jnnp.54.6.505 on 1 June 1991. Dow

    nloaded from

    http://jnnp.bmj.com/

  • Britton, Meyer, Benecke

    moving with his right hand and, additionally,involuntary movements of his right hand whenattempting to move only his left hand. Onexamination he had a right hemiparetic gait anda mild right hemiparesis with increased muscletone and brisk reflexes on the right. Mirrormovements of the left hand were observedwhen he attempted to move only the right handand mirror movements of the right hand wereseen when he attempted to move only the lefthand. CT of the head showed hemi-atrophy ofthe left cerebral cortex. There was no abnor-mality of the neck and sense of smell wasnormal.

    Case 3 A 38 year old right handed woman hadbeen aware of mirror movements in her handsall her life. She was the second of identicaltwins. Pregnancy and development were other-wise normal. On examination there were non-suppressible mirror movements of both hands,which were present with self initiatedmovements as well as with externally triggered(reaction time paradigm) movements. Finefinger movements were mildly impaired. Shehad considerable difficulty in producing alter-nating flexion-extension movements of onewrist and extension-flexion movements of theother. Muscle strength, tone and tendonreflexes were normal. There was no abnor-mality of the neck and sense of smell wasnormal.

    Case 4 A 59 year old right handed hyperten-sive man suddenly developed a severe righthemiparesis. CT revealed a hyperdense lesionin the region of the left internal capsule com-patible with cerebral haemorrhage. Threeweeks later, after considerable functionalrecovery, he noticed that attempts to move hisright hand resulted in associated mirrormovements in the left hand; movements of hisleft hand, however, were not accompanied bymovements of the right.A fifth patient, aged 48, with a left hemiplegia

    as a result of embolic occlusion of the rightmiddle cerebral artery was also studied.

    Five male normal subjects (age range 27 to 31years, mean 28 years) were studied. Ethicalcommittee approval for these investigationshad been obtained.

    RECORDING AND ANALYSISElectromyographic (EMG) recordings weretaken from both first dorsal interosseous (FDI)musces -using. Ag/AgCl -surface electrodesplaced over the muscles. Signals were amplifiedby a Tonnies Myograph II (TonniesMedizinische Elecktronik, Freiburg, Ger-many) with bandpass filtering between 20 Hzand 3000 Hz. Data was then collected andstored on floppy disks using a Tandon personalcomputer (IBM PC compatible) andAUTOLAB data collection programmes(sampling frequency 8000 Hz/channel;AUTOLAB, Kunze Software, Dusseldorf,Germany). Onset latencies of responses weremeasured from the first deviation from thebaseline by visual inspection of the signals on

    the computer screen. Response amplitudes(peak to peak) were measured by computer.The motor cortex was stimulated using the

    commercially available MAGSTIM 200(Novametrix Medical Systems, Connecticut,USA). The coil (consisting of 19 turns ofcopper wire; inner diameter 5 5 cm, outerdiameter 12 cm) was laid flat against the scalpand was centred carefully over the vertex.

    It has previously been noted that, with thismake of stimulator (not necessarily with oth-ers9) and with the coil centred over the vertex,currents flowing clockwise around the coil (asviewed from above) normally produce largerresponses in the right FDI muscle than the leftFDI, whereas anticlockwise currents evokelarger responses in the left FDI than rightFDI.'° This was confirmed in our normalsubjects. Given the known contralateralemphasis (particularly for hand muscles) of thecorticomotoneuronal projection, the mostlikely explanation for this result is that clock-wise currents stimulate the left cerebral hemis-phere more effectively (at least in terms ofevoking hand muscle responses) than the rightcerebral hemisphere and that anticlockwise coilcurrents stimulate the right cerebral hemis-phere more than the left.Thus one method of studying the cortico-

    motoneuronal projections in normal subjectsand in patients with mirror movements wouldbe to compare directly the size of responsesevoked in the two FDI muscles with clockwisecurrents (when, in normal subjects, right FDIresponses are larger than left FDI responses) orwith anticlockwise currents (when, in normalsubjects, left FDI responses are larger thanright FDI responses). The occurrence of rever-sed responses (for example, clockwise currentsproducing larger responses in the left FDI thanthe right FDI, or vice versa) could then beinterpreted as indicating a pathological,ipsilaterally projecting corticomotoneuronaltract. There is, however, a problem with suchreasoning in that the size of an individual FDIresponse may be significantly influenced notonly by the relative proportion of contra- andipsilaterally projecting corticomotoneuronalaxons but also by other factors (electrodeplacement, peripheral nerve lesions, lesionsof central motor pathways such as lacunarinfarction).To avoid these confounding factors, the size

    of responses evoked by clockwise coil currentswas compared with the size of responses evokedby anticlockwise currents at the same stimula-tion strength and in the same FDI muscle. Innormal subjects, right FDI responses are largerwith clockwise coil currents than anticlockwisecoil currents at the same stimulation strength(within the range of stimulation strengths fromthreshold to 1 5 times threshold): for left FDImuscles, anticlockwise currents produce largerresponses than clockwise currents. Theimplications of these normal findings with thecorticomotoneuronal projections and the inter-pretation of reversed response patterns areraised in the discussion.With the coil position carefully centred over

    the vertex, responses were recorded in FDI

    506

    on June 6, 2021 by guest. Protected by copyright.

    http://jnnp.bmj.com

    /J N

    eurol Neurosurg P

    sychiatry: first published as 10.1136/jnnp.54.6.505 on 1 June 1991. Dow

    nloaded from

    http://jnnp.bmj.com/

  • Central motor pathways in patients with mirror movements

    Clockwise Anticlockwise

    L FDIl

    R FDI

    Figure I Electromyographic recordingsfrom the left (upper traces) and right (traces) first dorsal interosseous (FDI) muscles in a normal subjectfollowing magbrain stimulation with clockwise coil currents (left side offigure) and anticlockwicurrents (right side offigure). The coil was centred over the vertex and the stimudstrength was 75% of the maximum stimulator output. Note how in the left FDI aresponse is produced by an anticlockwise coil current, while in the right FDI a larresponse is produced by a clockwise coil current. Calibration bars 20 ms and I mTV

    Left

    I ToTI

    1

    30% -

    T

    0

    T.

    _

    50

    Right

    T

    T

    100 0Stimulator output (%)

    50

    muscles with clockwise and anticlockwisecurrents (direction of coil current was changedby simply turning over the coil) using a series ofstimulation strengths between 30 and 100% ofmaximum output ofthe stimulator. At least fiveresponses were obtained for each stimulationstrength. During stimulation, subjects wereasked to relax completely and this was checkedby constant display of the surface EMG on anoscilloscope.To quantify the difference in response size

    with different coil currents, the mean peak tolower pleak amplitude of the evoked responses was

    neticoi calculated for each stimulation strength andlation then plotted on a graph (fig 2). A line was drawnlarger across the graph at a height equal to halfthe size

    ,yer of the largest response obtained at rest. Thestimulation strength (expressed as percent ofmaximal stimulator output) needed to producesuch a half maximal response when the currentflowed anticlockwise around the coil was sub-tracted from the stimulation strength neededwhen the current flowed clockwise. If responseof half maximal size were not evoked when thecoil current flowed in the non-preferentialdirection, then the highest stimulation strengththat was used in the particular subject was

    T employed in this calculation. For right FDImuscles of normal subjects this results in a

    l negative value, and for left FDI muscles apositive value.

    ~5% For the purposes of obtaining corticallyevoked responses with the shortest latency,magnetic stimulation of the motor cortex was

    T repeated with pre-activation of the muscle and° with the stimulator output set at 1-5 times

    ° . threshold for responses in the relaxed100 muscle." 12 Peak to peak size of responses

    evoked with pre-activation of the muscle wereaf half also measured.'rgure 6 LyrruFr UJ a'XVerUgeSyM UpUK 31zt; UJ {-e3pflu*& etmit"iU tri; tt tJ Lty truJ

    offigure) and right (right half offigure) first dorsal interosseous (FDI) muscles of anormal subject following magnetic brain stimulation with clockwise (filled circles) andanticlockwise (open circles) coil currents plotted against stimulation strength expressedas percent of maximal output of the stimulator. Note how clockwise coil currents producelarger responses in the right FDI than anticlockwise currents at the same stimulationstrength and vice versa. This difference in the size of responses with currents of differentorientation has been quantified by subtracting the stimulation strength required to producea half maximal response with anticlockwise currentsfrom that required with clockwisecurrents.

    Clockwise

    LLFDI

    R FDI

    Anticlockwise

    Figure 3 Electromyographic recordings from the left (upper traces) and right (lowertraces) first dorsal interosseous (FDI) muscles in a patient with congenital mirrormovements (case 1) following magnetic brain stimulation with clockwise coil currents(left side offigure) and anticlockwise coil currents (right side offigure). The coil wascentred over the vertex and the stimulation strength was 90% of the maximumstimulator output. Note how in the left FDI a larger response is produced by a clockwisecoil current, while in the right FDI a larger response is produced by an anticlockwise coilcurrent. This is the opposite pattern from normal (see fig 1). Calibration bars 20 ms and05 mV.

    ResultsNormal subjectsResponses in relaxed FDI muscles could beobtained in all normal subjects with the coilcurrent in either direction. Threshold for res-ponses in right FDI muscles (mean 46% ofmaximum stimulator output; range 40%-60%)was always lower with clockwise coil currents(when viewed from above) than with anticlock-wise currents, whereas threshold for responsesin left FDI muscles (mean 48%; range 40-65%) was always lower with anticlockwisecurrents. Amplitude of the responses in an FDImuscle was dependent on stimulation strengthand current direction (figs 1, 2). For the samestimulation strength, the peak to peakamplitude of responses in right FDI muscleswere always greater with clockwise coilcurrents than with anticlockwise currents. Theopposite was true for the left FDI muscles. Toproduce a response of half maximal size in theright FDI, the stimulation strength requiredwith clockwise coil currents was 22% (range18% to 30%) less than that for anticlockwisecurrents, while for half maximal responses inthe left FDI the stimulation strength requiredwith clockwise currents was 18% (range 10%to 32%) more than that for anticlockwisecurrents.

    5

    4

    E-f3Na)

    CD

    °o2am

    1

    0

    507

    Fiaijr,p 2 Ownhhe nf nvornvo honk tn honk ci,-,p nf rochnmocoliritod i" tho IPft (Ip

    on June 6, 2021 by guest. Protected by copyright.

    http://jnnp.bmj.com

    /J N

    eurol Neurosurg P

    sychiatry: first published as 10.1136/jnnp.54.6.505 on 1 June 1991. Dow

    nloaded from

    http://jnnp.bmj.com/

  • Britton, Meyer, Benecke

    0

    Left

    I

    I

    1 z

    11

    IIi

    1,0* 06*

    3

    2

    1

    Rigi

    a

    6f

    4

    II

    . n I

    ar

    6

    II

    I

    4

    2

    0 50

    I

    I

    50

    ____ O

    100 0Stimulator output (%)

    Figure 4 Graphs of average peak to peak size of responses elicited in theoffigure) and right (right half offigure) first dorsal interosseous (FDI)patients with congenital mirror movements,following magnetic brain stimclockwise (filled circles) and anticlockwise (open circles) coil currents plcstimulation strength expressed as percent of maximal output of the stimulin A (case 1) and B (case 2) responses in the right FDI were larger or oJwhen the coil current was anticlockwise compared to when the coil currentthis is the opposite of the normalfindings. Conversely responses in the leftlarger or of equal size when the coil current was clockwise compared to wicurrent was anticlockwise, which is again an abnormalfinding. These restthe pyramidal pathways in these two patients are not normally decussatecthird patient with congenital mirror movements (C; case 3) had response.dependent on coil current direction in a normalfashion.

    With pre-activation of the mu,tion strength at 1-5 times threshopreferential coil current directio:ponse latencies were 21-2 ms (rangms) for the right FDI and 21-3 mto 23-0 ms) for the left FDI.

    ht Patients with congenital mirror movementsResponses could be obtained in both FDImuscles in each patient following magneticbrain stimulation. Threshold for responses(range 45-60%) lay within the normal range,but in three out of six sides the threshold forclockwise and anticlockwise coil currents was

    I equal, in contrast to the normal findings (fig 3).t The amplitude of responses is plotted against0l stimulation strength in figure 4. The graphs for

    , case 1 and case 2 show qualitative and quan-titative differences from the findings in normal

    r a subjects. Qualitatively, the size of responses in*04 an FDI muscle was not dependent on coil

    current direction in the same way that it was fornormals. Thus for right FDI muscles anti-clockwise coil currents produced equally sizedor larger responses than clockwise currents at

    the same stimulation strength (which is theopposite of that found in normals). Further-

    1'T 1 more, in these two patients, left FDI responseswere preferentially evoked by clockwise

    currents (in normal subjects anticlockwisecurrents produced larger responses). Quan-titatively, in order to produce responses of halfmaximal size in the right FDI muscles, thestimulation strength required with clockwisecoil currents was 0% (case 1) and 3% more

    4jl (case 2; mean for normal subjects 22% less)61

    than that for anticlockwise currents, while forhalf maximal responses in the left FDI thestimulation strength required with clockwisecurrents was 16% less (case 1) and 0% (case 2;mean for normals 18% more) than that foranticlockwise currents.

    In cases 1 and 2, we made the additionalobservation that the size of responses in theright FDI muscles could be increased bymoving the coil over the right (ipsilateral)cerebral hemisphere, while the size of left FDI

    t 1 responses could be increased by moving thecoil towards the left cerebral hemisphere.

    In case 3, .response size varied with coilcurrent direction in a normal manner. That is,clockwise currents evoked larger responses in

    6the right FDI than anticlockwise currents at

    00o ~ the same stimulation strength and vice versa.'*100 With pre-activation of the muscle, stimula-

    tion strength at 1 5 times threshold and use ofpreferential coil current direction, corticallyevoked response latencies and amplitudes in all

    e left (left half three patients with congenital mirrorulaston woithree movements were found to be within our normalotted against range. Response configurations also appearedator. Note how normal although this was not studied(equal size sytmicl.t was clockwise: systematically.FDI wereien the coil Patient with acquired mirror movementstHowevr that Responses could be obtained in the relaxed

    s which were FDI muscles of both sides, although theresponses on the hemiparetic side were muchsmaller than those on the non-paretic side.Threshold for responses were within our nor-mal range (right 65%; left 50%). Amplitudes of

    scle, stimula- responses were dependent on stimulationold and use of strength and current direction in a normaln, mean res- manner (fig 5). Responses in the right FDIre 18-8 to 22-4 muscle were larger with clockwise coil currentsks (range 20 7 than with anticlockwise currents for the same

    stimulation strength. With muscle pre-activa-

    3 -

    2

    1 _

    6 -

    > 4 -ENU)ecn _

    C.w0cr

    0-

    8 _

    6 -

    4

    508

    I 0%I

    I In

    on June 6, 2021 by guest. Protected by copyright.

    http://jnnp.bmj.com

    /J N

    eurol Neurosurg P

    sychiatry: first published as 10.1136/jnnp.54.6.505 on 1 June 1991. Dow

    nloaded from

    http://jnnp.bmj.com/

  • Central motor pathways in patients with mirror movements

    5

    4

    31-

    2

    >ECDN

    CoCD

    1

    0

    Left 5'I0

    I

    01~I

    1'T0

    .T

    0J

    "it

    0 50

    4

    3

    2

    T

    100

    1

    r509

    Right

    51

    4

    3

    No responses2

    1

    0

    Stimulator output (%)

    T

    I

    ;0

    I

    011I I

    IIi~Ti0

    fI T,1

    o

    50

    Figure S Graphs of average peak to peak size of cortically evoked responses in the lej(left half offigure) and right (right halfoffigure) first dorsal interosseous (FDI) mustof one patient with acquired mirror movements (upper graphs) and of one patient withcomplete right hemiplegia (lower graphs) elicited with clockwise (filled circles) andanticlockwise (open circles) coil currents plotted against stimulation strength expressesas percent of maximal output of the stimulator. Note how the responses are dependent icoil current direction in a normal manner: that is clockwise coil currents produced largresponses than anticlockwise currents at the same stimulation strength in the right FDTwhile anticlockwise currents were better at stimulating the left FDI. No responses coulbe obtained in the completely paralysed right FDI.

    tion, cortically evoked response latencies w(within normal limits, although somew}slower in the right FDI (22-4 ms) comparwith the left (21-7 ms). Response amplituwas also reduced on the right (1 -6 ncompared to 6 7 mV).

    Patient with right CVAResponses following magnetic brain stimution could be obtained in the right FDI musiwith both clockwise and anticlockwise ccurrents, but no responses could be evokedthe left FDI. Responses in the right FDI w(normal for threshold (45%) and for 1direction of current that produced the larnresponses for a given stimulation strength. Tstrength of stimulation required to produchalf maximal response was 60% with a clocwise coil current and 70% with an anticlocwise current. The presence of responses in tright FDI muscle with both clockwise aanticlockwise coil currents in this paticindicates that bilaterally projecting c(ticomotoneuronal pathways are not necessi

    to explain the finding of bilateral responses innormal subjects when the coil is centred overthe vertex.

    DiscussionWe have demonstrated that hand muscle res-ponses evoked by magnetic motor cortexstimulation in two out of three patients withcongenital mirror movements were preferen-tially elicited by using the opposite direction ofcoil current to that which produces the largestresponses in normal subjects. This suggests

    * that either: 1) their primary motor corticeswere different from normal and were more

    .-i sensitive to stimulation with coil currents in theopposite direction to normal (that is, clockwisecoil currents stimulated the right cerebralhemispheremore effectively than the left hemis-phere) or 2) the corticomotoneuronal com-ponent of the pyramidal pathways projected tospinal alpha motor neurons abnormally. Wefavour the latter hypothesis in view of ourobservation that moving the coil over theipsilateral hemisphere in these cases oftenfurther increased the size of the response. Inaddition, given that the threshold, latency andsize of responses evoked by magnetic stimula-tion were normal, a major abnormality in thearchitecture of the motor cortices seems un-likely. Furthermore, such a hypothesis wouldbe in keeping with the results of electricalmotor cortex stimulation.67

    100 This study has also shown that congenitalmirror movements are not invariablyassociated with abnormal corticomotoneuronal

    fles projections. The third patient, whose congen-i a ital mirror movements were clinically indistin-

    guishable from the mirror movements of thed first and second patients, had entirely normalon,er motor responses to magnetic stimulation,l, including responses which were appropriatelyd sensitive to the direction of current. The fast

    corticomotoneuronal component of thepyramidal tracts therefore projected normallyto spinal alpha motor neurons, although this

    ere does not exclude the possibility that the projec-hat tion of slower components of the pyramidalred pathway might have been abnormal.ide What can be said of the pathophysiologicalnV basis of congenital mirror movements? The

    appearance of mirror movements on attempt-ing to move only one side of the body dependsupon the original command from high centres

    ila- within the brain to move only one side of theIcle body inappropriately reaching the spinal alpha-oil motor neurons of both sides. Clearly, severalin different routes, via pyramidal or extra-ere pyramidal pathways, could be suggested tothe account for this abnormal transmission.'3 Theger initial command might be incorrectly sent to'he motor centres on both sides of the brain (fore a example, to the motor cortices of both cerebralck- hemispheres), or signals in the descendingck- motor pathways from motor centres ofone sidethe of the brain (for example, pyramidal pathways)nd might activate spinal structures bilaterally as aent result of the abnormal functional/anatomicalor- projection of such pathways, or activity on oneary side of the spinal cord might cross to the other

    on June 6, 2021 by guest. Protected by copyright.

    http://jnnp.bmj.com

    /J N

    eurol Neurosurg P

    sychiatry: first published as 10.1136/jnnp.54.6.505 on 1 June 1991. Dow

    nloaded from

    http://jnnp.bmj.com/

  • Britton, Meyer, Benecke

    side as a result of abnormal interneuronalconnections.Could the abnormal ipsilateral cortico-

    motoneuronal projections as revealed in twoout ofour three patients explain the appearanceof congenital mirror movements? In normalsubjects making unilateral hand movementsonly the contralateral motor cortex isactivated.4 15 A totally non-decussated(ipsilaterally projecting) corticomotoneuronalprojection, by itself, could not thereforeaccount for mirror movements. Assuming thatmotor cortex activation occurs normally, whatis required is that activity in the motor neuronsof one cerebral motor cortex reaches the alphamotor neurons on both sides of the spinal cord.This could be effected either by individualpyramidal axons branching to synapse directlywith alpha motor neurons on both sides of thespinal cord or by a proportion of activatedcortical motor neurons projecting ipsilaterallyand a proportion projecting contralaterally.Farmer et al,7 on the basis of a narrow peak oncross-correlograms between motor units in thehand muscles of a patient with Klippel-Feilsyndrome, showed that the spinal motorneurons subserving homologous hand muscleson each side receive synaptic input from acommon neuron and they suggested that thecommon neuron was a cortical motor neuronwith an abnormally branched axon.The finding that one of our patients with

    congenital mirror movements had normalcorticomotoneuronal pathways as assessed bymagnetic motor cortex stimulation is importantsince it implies either that there is more thanone pathophysiological cause of congenitalmirror movements or that the abnormalcorticomotoneuronal projections seen in theother patients is not the pathological cause ofthe mirror movements, merely an associatedfeature. That mirror movements are also seenwith a variety of acquired disorders,' in whomthere would be no reason to suspect an abnor-mal pyramidal pathway projection does suggestthat an abnormal corticomotoneuronal projec-tion is not a necessary requirement for theappearance of mirror movements. This is

    supported by the normal projection found inour patient with acquired mirror movements asa result of internal capsular infarction. Wherethe abnormality lies in these patients isuncertain.

    ErratumThroughout the text (and in the table) the word "anticlockwise"should be replaced by "clockwise" and vice versa. Previouspublished studies using the Novametric magnetic stimulatormay also contain this error (see JNNP 1990;53:707).

    This study was supported by the DFG and EC. TCB is an ECResearch Fellow.

    I Zulch KJ, Muller N. Associated movements in man. In:Vinken PJ, Bruyn GW, eds. Handbook of clinicalneurology, Vol 1. Amsterdam: North Holland, 1969:404-26.

    2 Myrianthopoulos NC. Mirror movements. In: Vinken PJ,Bruyn GW, eds. Handbook of clinical neurology, Vol 42.Amsterdam: North Holland, 1982:233-234.

    3 Avery LN, Rentfro CC. The Klippel-Feil syndrome: apathologic report. Arch Neurol Psychiatr 1936;36:1068-76.

    4 Gunderson CH, Solitare GB. Mirror movements in patientswith the Klippel-Feil syndrome. Arch Neurol 1968;18:675-9.

    5 Conrad B, Kriebel J, Hetzel WD. Hereditary bimanualsynkinesis combined with hypogonadotropic hypo-gonadism and anosmia in four brothers. J Neurol1978;218:263-74.

    6 Cohen LG, Bandinelli S, Lelli S, Hallet M. Non invasivemapping of hand motor somatotopic representation usingmagnetic stimulation. J Clin Neurophysiol 1988;5:371.

    7 Farmer SF, Ingram DA, Roche SW, Stephens JA. Evidencethat mirror movements in the Klippel-Feil syndrome inman are produced by activity in branched-stem pre-synaptic inputs from corticospinal neurones. J Physiol1989;418: 102.

    8 Barker AT, Jalinous R, Freeston IL. Non-invasive magneticstimulation ofhuman motor cortex. Lancet 1985;i: 1 106-7.

    9 Claus D, Flugel D, Brenner PM, Gropp N. Zum Einfluss derParameter magneto-elektrischer Reize aud deren Wirk-ung im Nervensystem. Z EEG EMG 1990;21:187.

    10 Hess CW, Mills KR, Murray NMF. Responses in smallhand muscles from magnetic stimulation of the humanbrain. J Physiol 1987;338:397-419.

    11 Hess CW, Mills KR, Murray NMF, Schriefer TN. Mag-netic brain stimulation: central motor conduction studiesin multiple sclerosis. Ann Neurol 1987;22:744-52.

    12 Ingram DA, Thompson AJ, Swash M. Central motorconduction in multiple sclerosis: evaluation of abnor-malities revealed by transcutaneous magnetic stimulationof the brain. J Neurol Neurosurg Psychiatry 1988;51:487-94.

    13 Forget R, Boghen D, Attig E, Lamarre Y. Electromyo-graphic studies of congenital mirror movements.Neurology 1986;36:1316-22.

    14 Kristeva R, Keller E, Deecke L, Kornhuber HH. Cerebralpotentials preceding unilateral and simultaneous bilateralfinger movements. Electroencephalogr Clin Neurophysiol1979;47:229-38.

    15 Roland PE. Organisation of motor control by the normalhuman brain. Human Neurobiol 1984;2:205-16.

    510 on June 6, 2021 by guest. P

    rotected by copyright.http://jnnp.bm

    j.com/

    J Neurol N

    eurosurg Psychiatry: first published as 10.1136/jnnp.54.6.505 on 1 June 1991. D

    ownloaded from

    http://jnnp.bmj.com/