three symptom change tourette's and disorderall structures are...

J7ournal ofNeurology, Neurosurgery, and Psychiatry 1996;61:497-505

Three cases of symptom change in Tourette'ssyndrome and obsessive-compulsive disorderassociated with paediatric cerebral malignancies

Bradley S Peterson, Richard A Bronen, Charles C Duncan

AbstractObjective-To correlate behaviour mani-festations with tumour location in threechildren who had Gilles de la Tourette'ssyndrome (GTS), obsessive-compulsivedisorder (OCD), and primary cerebralmalignancies.Method-Cases were ascertained from achart review in a GTS and OCD specialtyclinic. For each case the temporal pro-gression of change in neuropsychiatricsymptoms was qualitatively correlatedwith radiographic documentation oftumour progression.Results-The change in symptom severi-ties during tumour progression and treat-ment, together with prior neurobiologicalstudies of GTS, suggest that the ventralstriatum, corpus callosum, thalamus, andmidbrain are potentially importantneural substrates in the formation ormodulation of tic symptoms. The limbicsystem, including the hypothalamus andcingulate, and the caudate nucleus, seemto be important in the neurobiology ofOCD. All structures are neuroanatomi-cally and functionally related to the corti-costriato-thalamocortical circuitry that isthought to subserve symptom generationin both GTS and OCD.Conclusion-Although the malignancieswere not likely to have caused the tic andOCD symptoms in these children, thelocations ofthese intracranial lesions pro-vide important clues in identifying brainregions that may contribute to the deter-mination of tic and OCD severities.

(J Neurol Neurosurg Psychiatry 1996;61:497-505)

Yale Child StudyCenter, New Haven,CT, USAB S PetersonYale University SchoolofMedicine in NewHaven, CT, USAB S PetersonR A BronenC C DuncanCorrespondence to:Dr B S Peterson, Yale ChildStudy Center, 230 SouthFrontage Road, New Haven,CT 06520, USA.Received 1 March 1996and in final revised form27 June 1996Accepted 1 July 1996

Keywords: Tourette's syndrome; obsessive-compulsivedisorder; neoplasms

Gilles de la Tourette's syndrome (GTS) is a

childhood onset movement disorder consistingof multiple motor and phonic tics that fluctuatein severity. It is thought to be aetiologicallyrelated to at least one form of obsessive-com-pulsive disorder (OCD),' a neuropsychiatricillness characterised by the experience ofrecurrent, intrusive, and distressing thoughts,images, and urges to action. In clinical settingsGTS is also often associated with attentiondeficit hyperactivity disorder (ADHD).We describe three cases of GTS and OCD

associated with paediatric cerebral malignan-cies. In each case, GTS and OCD symptomsseemed to change during specific phases oftumour development and treatment. For eachcase we attempt to correlate lesion locationwith changes in symptomatology. These corre-lations may aid in identifying the neurobiolog-ical substrate of GTS related disorders. Thecases were ascertained from among more than800 that presented to our GTS/OCD specialtyclinic over a period of 10 years.

Case 1A right handed girl at the age of 8 developedunilateral facial tics, vocal tics, and compulsivestereotypies during progression of her opticchiasm astrocytoma.

EARLY INTRAMORBID COURSEHer tumour was first diagnosed at the age of18 months when she presented with acceler-ated linear growth. Because her growth slowedsubstantially during a trial of bromocriptine,the tumour was erroneously thought to bebenign.

TUMOUR DESCRIPTIONYearly MRI showed a large tumour thatseemed stable in size (fig 1A and B). Thetumour involved the entire rostrocaudal extentof the hypothalamus, and extended bilaterally(right greater than left) into the thalami, infe-rior genu, the posterior limb of the internalcapsule, ventral striatum, nucleus accumbens,septal nuclei, and the columns of the fornix.The tumour extended minimally into the pos-terosuperior aspect of the amygdala and didnot seem to involve the hippocampus.Inferiorly it invaded the medial substantianigra bilaterally and surrounded both rednuclei. The caudate nuclei appeared thinneddue to dilatation of both lateral ventricles.

FAMILY AND DEVELOPMENTAL HISTORYThe girl's maternal grandfather had obsessivepreoccupations with orderliness, cleanliness,germs, and disease, and corresponding order-ing, arranging, and cleaning compulsions thatmet DSMIII-R criteria for OCD,' but thatnever required treatment. No family memberhad a history of motor or vocal tics. Otherthan her large stature, this girl's early growthand development were normal until the age of7, when she began to exhibit brief periods ofunexplained aggression. Although she gained

497

on January 21, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.61.5.497 on 1 Novem

ber 1996. Dow

nloaded from

http://jnnp.bmj.com/

Peterson, Bronen, Duncan

Figure 1 Case 1: opticchiasm astrocytoma. AxialT2 weighted MRI at thelevel of the midbrain(A) and thalamus (B)at the age of 8 shows ahyperintense tumour withits epicenter at the opticchiasm. The tumourinvolves the entire extent ofthe hypothalamus,extending bilaterally intothe thalamus, inferioraspect of the genu andposterior limb of theinternal capsule, ventralstriatum, columns of thefornix, nucleus accumbens,septal nuclei, amygdala,and midbrain withinvolvement of the medialsubstantia nigra andanterior red nuclei. Theright sided structures aremore affected than the left.(C) Postoperative sagittalTl weighted contrastimaging at the age of 8years and 9 months showsthe surgical defectinvolving the midbody ofthe corpus callosum andthe cingulate (arrows), thecolumns of thefornix, anda partially resectedenhancing tumour (curvedarrow).

nocturnal continence by the age of 7, nightlyenuresis returned at the age of 8.

GTS AND OCD SYMPTOMSShortly after she was 8 she developed frequentthroat clearing and lyrical vocal tics that weresoon followed by a left facial tic, a left eye-blinking tic, and non-purposeful, left sidedmouth opening. She also often readjusted theglasses on her nose until they felt "just right".

PROGRESSIONTics increased in frequency and intensity overthe next five months. At the age of 8 years and5 months she experienced acute loss of vision,headache, vomiting, and episodic loss of con-sciousness. Brain CT showed obstructivehydrocephalus. She underwent a non-curativetumour resection that involved a bifrontalcraniotomy, retraction of both cerebral hemi-spheres, section of the anterior corpus callo-sum, and partial tumour excision.

POSTOPERATIVE TUMOUR DESCRIPTIONPostoperative MRI showed surgical resectionand retractor injury involving the fornicesbilaterally, left wall of the third ventricle, lefthypothalamus, genu of left internal capsule,left caudate nucleus head, right cingulategyrus, and mid-body of the corpus callosum(fig 1 C). Histopathological analysis of thetumour disclosed a highly malignant grade IIastrocytoma. Two months later she underwent

palliative radiation treatment, receiving5400 cGY in the region of her optic chiasm.

SYMPTOM PROGRESSIONImmediately after surgery her simple motorand vocal tics abruptly disappeared. She soondeveloped, however, numerous repetitive, selfstimulating, and self manipulating stereotypiesand compulsions. She constantly rocked backand forth and picked her nose or ears untilthey bled. She rubbed sand into her scalpwithout explanation, and she compulsivelyfondled her genitals in public. Other compul-sions were more bizarre. She would pick hernose and then insert the mucus into her vaginaor rectum. She could not offer explanationsfor these behaviours, other than to say shecould not stop herself. She developed anobsessive preoccupation with tactile stimuli,especially with the feeling of clothes on herbody and the texture of objects. She alsodeveloped a hyperphagia that produced aweight gain of more than 40 pounds in asmany days. She rummaged through garbagecans for food and locks had to be put on refrig-erator doors. Neuropsychological deficits doc-umented, after surgery and before radiationtherapy, a verbal IQ of 123, a performance IQof 77, and impaired immediate and delayedrecall.

OUTCOMEFour months after surgery the girl again had arapid decline in visual acuity and MRI docu-mented rapid tumour enlargement. She begana course of carboplatinum and vincristine.Serial MRI documented no further growth ofher tumour nearly five years after her partialresection. She has outlived all medical predic-tions for her life expectancy.

Case 2A right handed boy had had obsessive-com-pulsive symptoms and tics since the age of 7.He developed a left parietal anaplastic mixedglial cell malignancy.

EARLY INTRAMORBID COURSEThis boy was intelligent and well adjusted

498


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1996. Dow

nloaded from


Three cases ofsymptom change in Tourette's syndrome and obsessive-compulsive disorder associated with paediatric cerebral malignancies

Figure 2 Case 2: parietalanaplastic glial celltumour. Sagittal Tlweighted (A), axialprotondensity weighted (B), andcoronal proton densityweighted images show thetumour to be hypointenseto grey matter on Tlweighted images andhyperintense on protondensity scans, with regionsofsignal hypointensityconsistent with calcification(arrowhead). This largetumour involves the bodyof the corpus callosum (b),the left cingulate gyrus (c),fornix, and body and tailof the left caudate nucleus.

L

despite the presence of infrequent eye blinkingand throat clearing tics since the age of 7,which his family regarded as a habit and forwhich they did not seek treatment. Althoughhe did not meet formal criteria for OCD, hekept all of his belongings spotless and hisclothing and toys neatly arranged.

FAMILY AND DEVELOPMENTAL HISTORYA maternal uncle and cousin had chronicfacial tics, but no family member had GTS or

OCD. Occasional bedwetting episodes beganat the age of 8, after a year of nocturnal conti-nence. At the age of 9 he developed a left sidedfocal sensory seizure disorder that began withleft sided facial numbness before generalising.Neurological examination was normal, but anEEG showed "Rolandic" epileptiform dis-charge from the right central mid-temporalregion. A non-contrast head CT was normal,and a diagnosis of benign focal epilepsy wasmade. Phenytoin was started after severalbouts of enuresis raised the possibility of noc-turnal seizures.

GTS AND OCD SYMPTOMSAt the age of 11 spontaneous and darting left-ward conjugate eye movements began. Thesewere soon accompanied by a dystonic leftwardhead turning tic. Multiple neurological investi-gations were unrevealing, and the movementswere diagnosed as simple motor tics. A routinefollow up EEG disclosed a single right anteriormid-temporal sharp wave transient that wassuggestive of an irritative process, although theanomaly was deemed to be an artifact.

Tics were infrequent initially, but three

months later increased acutely, producingscores of eye movements each minute. Coverttongue thrusting that badly eroded his leftsided molars was also attributed to GTS.Clonidine (up to 0 3 mg/day) produced sub-stantial improvement in tic symptoms over thenext year, although pronounced exacerbationswere seen during times of stress or excitement,or after missing a medication dose.

PROGRESSIONAt the age of 12 years and 6 months he devel-oped frequent and forceful leg and hand tap-ping (more on the left), overt tonguethrusting, leftward head jerks, and a compul-sion to hit nearby people. He also developedinfrequent bitemporal throbbing headaches,thought to be migraines. Two months later anEEG showed independent sharp wave tran-sients bilaterally, greater in the left anteriorquadrant, which were attributed to migraines.His grades deteriorated from As and Bs tomostly Cs. His worsening tics, aggression, andpoor school performance were attributed toGTS. Two months later, at the age of 13 yearsand 6 months, recurrence of seizures was char-acterised by left arm stiffening and ataxicspeech, followed by left arm clonus. MRI thendisclosed his tumour.

TUMOUR DESCRIPTIONHis MRI (fig 2A and B) showed a 6 x 6 x 10cm left parietal tumour involving the corpuscallosum (isthmus, posterior, and midbody,seeming to cross the midline posteriorly), thebody and isthmus of the left cingulate gyrus,the fornix, and the body and tail of the leftcaudate nucleus. The temporal and occipitalhorns of the lateral ventricles were moderatelydilated due to obstructive hydrocephalus atthe level of the atrium. The tumour waspartially resected through a left parietal cran-iotomy. Histology showed a mixed oligo-dendroglioma and astrocytoma with anaplasticfeatures. Radiotherapy at the age of 14 wasunhelpful. Seizures continued, and the tumourquickly infiltrated through the cerebrum.

OUTCOMEIn the year after discovery of the tumour, hedeveloped hyperacusis and cold intolerance.Several strokes contributed to rapid clinicaldecline, and he died at the age of 15.

Case 3A right handed boy with GTS, OCD, andADHD developed a midbrain glioma and arecurrent major depressive illness.

EARLY INTRAMORBID COURSEAt the age of 4 this boy was evaluated withpsychological testing for frequent nightmaresand depression. He exhibited frequent eye-blinking, facial grimacing, and arm jerks. Novocal tics were yet evident. His IQ was normal,but severe visual-motor and fine motor controldeficits were seen. He was also impulsive, per-severative, inattentive, and motorically hyper-active, meeting DSMIII-R criteria for ADHD.

499


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1996. Dow

nloaded from



Figure 3 Case 3:midbrain glioma. Axialproton density MRI beforechemotherapy (A) at theage of 15 shows ahyperintense midbrainlesion (arrow) affecting theright tectum, extending tothe ipsilateral red nucleus.The tumour also involvesthe pulvinar of the rightthalamus (not shown).(B) At the age of 20, afterchemotherapy andradiotherapy, the tumourhas regressed in size.

Neurological evaluation at the age of 7 forADHD confirmed these findings. A head CTand an EEG were both normal. Stimulantswere not prescribed because of the presence oftics. At the age of 10 the neurologist againevaluated him for increasingly frequentmigrainous headaches, associated with nauseaand vomiting, that had been present since theage of 4. A repeat EEG was normal.

FAMILY AND DEVELOPMENTAL HISTORY

Family history included GTS in the patient'sfather and possible motor and phonic tics inhis paternal grandmother. No family memberhad ADHD or OCD. Early childhood paedi-atric evaluations suggested mild motor devel-opmental delays. He evidenced separationanxiety during most of his childhood.

GTS AND OCD SYMPTOMSAt the age of 1 1, as the frequency of migrainousheadaches increased, tics also increased in fre-quency and forcefulness. By the end of thatyear, tics were incapacitating and nearly con-stant. He had numerous frequent motor andvocal tics that included eyeblinking, eyebrowraising, facial grimacing, lip pouting, headjerks, finger, arm, and leg jerking, abdominaltensing, nose rubbing, arm extension, sniffing,throat clearing, high pitched humming, andecholalia. Symptoms ofOCD were severe, andincluded frequent checking, ordering, andevening up compulsions, list making, and ritu-alised behaviours. Clonidine (0 25 mg/day)was unhelpful for his tic and ADHD symp-toms.

PROGRESSIONBy the age of 12, the boy was complaining ofworsening headaches, frequent nausea andvomiting, and acoustic hypersensitivity.Symptoms of OCD and GTS remainedsevere. Because of the persistence of frequentheadaches, a brain MRI was obtained at theage of 12 years and 4 months.

TUMOUR DESCRIPTION

Brain MRI showed that a right tectal lesion

was obstructing the cerebral aqueduct, pro-ducing hydrocephalus of the lateral and thirdventricles. The lesion involved the periaque-ductal grey, right red nucleus, and right pos-teroinferior thalamus. A ventriculoperitonealshunt was placed through the right parietooc-cipital lobe junction, through the white matterof the parietal lobule into the centrum semio-vale and the right lateral ventrical. After shuntplacement, headaches and vomiting persistedbut were attributed to migraines.

SYMPTOM PROGRESSIONThe boy's tics worsened in frequency andintensity over the next six months, and hisOCD symptoms became intolerable. Pimozide(3 mg/day) was unhelpful for his tics. Serialscans in the next six months showed no evi-dence of recurrent hydrocephalus, although agradually increasing tumour size was seen latein his 12th year, extending from the rightquadrigeminal plate into the right posteroinfe-rior hemithalamus, with calcification of thepulvinar. Despite enlargement, the tumourwas believed to be benign. Phenytoin pro-duced remarkable headache relief. Tic andOCD symptoms continued to be severe. Atthe age of 13 he developed an affective illnessthat met DSMIII-R criteria for major depres-sion. He made three unsuccessful suicideattempts. Antidepressant trials were unhelpful.

At the age of 14, recurrent headaches, vomit-ing, and repeat CT indicated recurrent hydro-cephalus, prompting three shunt revisions.Headaches, nausea, and vomiting returned atthe age of 15. An MRI (fig 3A) at the age of 15years and 5 months suggested tumour growthand perhaps low grade malignancy, withinvolvement of the posterior portion of theright red nucleus and possible anterior exten-sion into the right subthalamic nucleus.Carboplatinum and vincristine chemotherapywas initiated but tumour growth continuedover the next six months. A course of radiationtherapy then arrested tumour progression bythe age of 16 years and 2 months. He was stillincapacitated by OCD symptoms despite treat-ment with 40 mg/day fluoxetine.

500


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1996. Dow

nloaded from


Three cases ofsymptom chanige in Tourette's syndrome and obsessive-compulsive disorder associated with paediatric cerebral malignancies

Patient 1

I *

1 0~~1

Patient 2

10

0,01) .'%ole/ <.-,. /(P,,. 0

00 *^- /01) lp%l,P+ ";.

¶1 4 1 5

¶7G15s¶

Patient 3

A

9 10 1 1 12 ¶L ` ¶ ¶ 13 14 15 16 ¶ 17

Figure 4 Time line of tumour and symptom progressiotn for each of the three cases. Details appear in the text. OCD sx = obsessive compulsive disordersymptoms; GTS sx = Gilles de la Tourette's symptoms.

OUTCOME

Tumour size remained stable, with continuedinvolvement of the inferior and superior colli-culi, the right red nucleus, and a slightly largerportion of the right pulvinar nucleus (fig 3B).Tic and OCD symptoms gradually but pro-

gressively improved between the ages of 16and 17. Currently, at 21 years of age, motortics are mild, and phonic symptoms are

absent. Mild OCD symptoms persist. He is offall medication and attends a communitycollege.

DiscussionThis is the first report of changes in symptomsof GTS and OCD associated with cerebralmalignancies. Only one other report has asso-ciated GTS with an intracranial tumour3 andwe know of no cases of cerebral malignancyassociated with OCD. In each of the presentcases, symptoms changed with tumour pro-

gression and with either effective treatment or

clinical demise (fig 4). All three children mayhave had a genetic diathesis to GTS or OCD,and their neuropsychiatric symptoms (particu-

larly in cases 2 and 3) antedated by severalyears the discovery of the tumours; it is there-fore unlikely that the malignancies caused theneuropsychiatric symptoms. The symptoms ofeach child nevertheless increased during peri-ods of rapid tumour growth or decreased soonafter successful treatment, providing an oppor-tunity to correlate symptom change withchanges in tumour size and location that occur

within a context of GTS and OCD illness ordiathesis. These lesion correlates of symptomchange can therefore help confirm and extendcurrent theories of GTS and OCD neurobiol-ogy.

GTS NEUROBIOLOGY

Although the cause of GTS is still unknown, itis thought to have strong genetic determi-nants.45 Family studies suggest that the vul-nerability to GTS is transmitted as a singlemajor gene that can be variably expressed aseither GTS or as OCD.' The specific geneticloci conferring this vulnerability have yet to beelucidated. It is currently unclear whetherADHD is also an alternative expression of thissame vulnerability gene." I The effects ofgenetic and non-genetic determinants in GTS

501

0

1 1

T(ffi@°>7

0/I,l10009"

<*, XI'

0%P0%

16

4 5 6 7 8

4V,

1 9


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1996. Dow

nloaded from



and GTS related OCD are thought to bemediated through the basal ganglia and associ-ated corticostriato-thalamocortical (CSTC)circuitry.9 11 Evidence for basal gangliainvolvement in GTS derives primarily frompostmortem case reports and in vivo imagingstudies showing hypoplasia and hypometabo-lism of GTS basal ganglia.12 18

NORMAL CSTC CIRCUITRYCircuits in the CSTC are composed of multipleloops that direct information from the cerebralcortex to the subcortex, and then back again to

specific regions of the cortex. Within each ofthese loops, information seems to be directedfrom multiple but functionally related corticalareas into specific portions of the subcortex-firstly, the striatum, then the globus pallidus(or substantia nigra pars reticulata), then to

specific thalamic nuclei-before returning tothe cortex.'9 21 These circuits are of centralimportance in the normal control of motoric,cognitive, and affective functioning. Thetumours in each of the children described herelikely altered GTS and OCD symptom sever-

ity by affecting the functioning of one or more

portions of this circuitry.

CLINICOPATHOLOGICAL CORRELATIONS FORGTS SYMPTOMSDespite the differing tumour locations in eachof these children, it is noteworthy that they allhad prominent midline lesions, and that intwo cases a change in symptom level was asso-

ciated with corpus callosum and cingulatelesions, whether from tumour or surgery. Inaddition to these common anatomical fea-tures, the temporal sequences of changes inanatomical involvement seem to correlatewith the development of specific behaviours inways that are consistent with previous clinicaland preclinical studies of GTS and OCD(table).

CASE 1This girl's tic symptoms permanently disap-peared after excision of her tumour and surgicallesion of the body of her right cingulate gyrus,

her anterior corpus callosum, and the neigh-bouring interhemispheric sensorimotor fibres.

Ventral striatumIt is possible that her left sided facial tics were

produced by tumour involvement of either herright sided ventral striatum or right amygdala.Unilateral electrical or chemical stimulation ofthe basal ganglia or the amygdala have beenreported to produce tic-like stereotypies andvocalisations in animals and humans.22 23

Amygdala stimulation produces primarily ipsi-lateral facial movements, however, and thegreater extent of this girl's tumour was con-

tralateral to her tics, which would suggest thather tics were not due to involvement of theamygdala. Anterior corpus callosum lesions inanimals, moreover, fail to abolish the facialmovements induced by amygdala stimula-tion,22 so that the disappearance of this girl'stics after callosotomy strongly suggests thather hemifacial tics were produced by tumour

involvement of her contralateral ventral stria-tum, and not her amygdala.

Corpus callosum and cingulateIt is improbable that partial evacuation of thisgirl's tumour would have completely relievedher tic symptoms. It seems more likely that thecallosal and cingulate lesions eliminated hertics, for two reasons: (1) one imaging studyhas reported decreased corpus callosum size inpatients with GTS, with a smaller corpus callo-sum correlating with decreasing severity ofsymptoms, suggesting that reduction in size ofthe corpus callosum may be compensatory inGTS and not necessarily pathogenic24; (2) pre-

clinical studies of basal ganglia function in ani-mals suggest that an intact corpus callosum isnecessary for normal basal ganglia function-ing.25 It is possible, therefore, that the corpus

callosum surgical lesion disrupted the knowninterhemispheric connectivity of the basal gan-

glia bilaterally,202' 26 and that this connectivitywas necessary for the expression of the leftsided tics due to the irritative lesion of herright ventral striatum.The cingulate lesion may also have pro-

Hypothesised regional neurobiological substrate of GTS and OCD, and the regions that are thought to modulate them. The lesions of each child and thesymptoms that are postulated to be either affected or caused by them are cross tabulated with these regions. Further details appear in the text

PATIENT 1 PA TIENT 2 PA TIENT 3Neurobiologicalsubstrate Regions Lesions Symptoms Lesions Symptoms Lesions Symptoms

Hypothesised CSTC circuits Ventral striatum L-sided facial Corpus callosum Motor and phonic R posterior GTScentral to Basal ganglia R> L tics L caudate tic symptom hemithalamus and OCDGTS Subthalamic Thalamus Vocal tics? exacerbation R subthalamic symptom

Thalamus nucleus Conjugate eye nucleus exacerbation?Supplementary and movements and

premotor cortices neck dystoniaCorpus callosum Tongue thrusting

Hypothesised Limbic circuits R cingulate Hyperphagia L cingulate Complex tics Checkingcentral to Cingulate gyrus (retractor Bizarre Fornix Aggressive ordering,OCD induced) compulsions motor tics evening up

Hypothalamus Hypothalamus Circadian and ritualsAmygdala Amygdala dysrhythmiaFornix Fomix

Modulating Brainstem Ascending Vocalisations? Hyperacusis Periaqueductal GTS andregions Periaqueductal grey monoaminergic grey OCD symptom

Locus coeruleus projections Locus coeruleus exacerbation?Ventral tegmentum Ventral tegmentum HyperacusisSubstantia nigra Substantia nigraDorsal raphe Dorsal raphe

R red nucleusSuperior colliculi

502


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1996. Dow

nloaded from



duced improvement in her tic symptoms.Electrical stimulation of of the anterior cingu-late in humans produces semi-involuntarymovements similar to complex motor tics,2728and anterior cingulotomies have beenreported, in uncontrolled studies, to alleviatetic symptoms in some patients.29 30 The portionof the cingulate that was damaged by theretractor in this girl, moreover, is the portionwhich receives afferent projections from themotor cortex."3

CASE 2Corpus callosum and cingulateThis boy's changes in symptom severity alsoseem to have been related to tumour involve-ment of the corpus callosum and cingulate,but this time primarily to the left of midline.Interhemispheric fibres topographically associ-ated with the sensorimotor strip, and thoseassociated with the parietal cortex, were mostprominently involved. Whereas surgicallesions in the corpus callosum in the firstpatient apparently reduced tic symptoms byinterrupting the interhemispheric transfer ofirritative stimuli, the presence of tumour inthis boy may have produced symptom exacer-

bation by producing an irritative focus withininterhemispheric sensorimotor fibres. Al-though involvement of the the left caudatenucleus may also have produced tic exacerba-tion in this patient, it seems more likely thatthe caudate's greater connectivity to frontaland association cortices produced his new,more complex tics (his repetitive tapping andhitting).9

CASE 3This boy's tumour involved primarily the tec-tum of the midbrain at the level of the superiorcolliculus, and it extended dorsally into theventeroposterior thalamus. Midbrain lesionsinduced by encephalitis lethargica have previ-ously been associated with GTS," and havebeen hypothesised to disturb dopaminergicpathways ascending from the substantia nigrapars compacta (SNJ) and the ventral tegmen-tal area (VTA). If the presence of tumourexacerbated tic symptoms in this patient, how-ever, then its tectal location (in the dorsum ofthe midbrain, away from the SNc and VTA)would suggest that the nigra and ventraltegmentum were not involved, and that themechanism of exacerbation was not due todisruption of ascending dopaminergic fibres.Rather, tic exacerbation due to involvement oftectal structures such as the superior colliculusand periaqueductal grey seems more plausible.

Superior colliculusThe superior colliculus receives heteroge-neous, multimodal, and topographicallyordered inputs from visual and auditory cen-

tres, motor cortex, and other brainstemregions that are hypothesised to subserve itsrole in the orienting response." Efferent path-ways from the colliculus receive and integratesensory stimuli, and then project to premotorcortical portions of CSTC circuits, whichcould explain an influence on tic symptoms

from this location. Situated just ventral to thesuperior colliculus, furthermore, is the mesen-cephalic nucleus of the fifth cranial nerve, anucleus that relays unconscious proprioceptiveinformation from the face. Disturbing thisrelay could affect the reflex reactivity of mus-cle spindles in the face, the region most com-monly affected by tics in this boy and in mostpatients with GTS.

Periaqueductal greySymptom exacerbation could also be due toinvolvement of the periaqueductal grey (PAG)matter, a vaguely defined cell mass that is con-tinuous rostrally with the ventromedial andlateral hypothalamus and caudally with thedorsal raphe nucleus. Stimulation of the PAGcan produce a wide array of behaviouralresponses, including sexual behaviours, vocali-sation of emotion, aggressive and defensiveposturing, and activation of the sympatheticnervous system,34 similar to many of the ticrelated behaviours seen in GTS. Disturbancesin the PAG previously have been hypothesisedto underlie the production of motor and vocaltics associated with encephalitis lethargica." Inaddition, noradrenergic fibres ascending fromthe locus coeruleus pass through the PAG andprobably were disrupted by this boy'stumour,35 which may have worsened his symp-toms. A similar mechanism of symptom exac-erbation was postulated in an adult malepatient with GTS whose PAG was calcified bya pineal tumour.3

ThalamusThe extension of this tumour, and that of thefirst patient, into the venteroposterior portionof the thalamus, complements a series of neu-rosurgical case reports in which thalamiclesions attenuated tic symptoms.3641 By con-trast with the destructive, non-irritative surgi-cal lesions, the thalamic tumour involvementin these children more likely produced focalirritation. Venteroposterior regions of the thal-amus and its subdivisions are an integralportion of the CSTC pathways carrying infor-mation for motor programmes that probablyparticipate in tic symptom formation. An irri-tative focus anywhere along this pathway isprone to produce pathological transmission ofinformation that it carries.

CLINICOPATHOLOGICAL CORRELATIONS FOROCD SYMPTOMSCommon to all three children, other than themidline location of their tumours, was theinvolvement of structures that are limbic sys-tem components: in patient 1, it was the hypo-thalamus, accumbens, and fornices from thetumour, and the body of the cingulate fromsurgery; in patient 2, it was the body of thecingulate and the fomices; and in patient 3, itwas the PAG, which has strong and integralconnections with the hypothalamus and cingu-late. Tumours involved the caudate nucleus intwo of the patients. These structures-thecaudate nucleus and other basal ganglia struc-tures, the cingulate and other limbic systemcomponents, and their connections with the

503


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1996. Dow

nloaded from



frontal cortex have been hypothesised tocomprise a disinhibited, reverberating neuralcircuit that gives rise to symptoms of OCD.'2

CingulateIt is noteworthy that whereas surgical lesionsof the anterior cingulate gyrus are thought toalleviate symptoms of OCD in a substantialnumber of subjects refractory to pharma-cotherapy,' the retractor lesion in the body ofthe right cingulate in patient 1 and the involve-ment of the body of the cingulate in patient 2were both associated with the onset or pro-gression of compulsive behaviours. Perhapsdiffering locations of lesions within the cingu-late produce differential effects on symptomswith OCD, with anterior lesions improving,and more posterior lesions exacerbating, OCDsymptoms.

HypothalamusThe cingulate gyrus and hypothalamus areintegrally interconnected, and both have beenimplicated in OCD and compulsive behav-iours. The hypothalamus, for instance, is thesite of oxytocin production, and oxytocin con-centrations have been shown to be increasedin OCD that is not genetically related toGTS."-' Oxytocin is one of the mediators ofnumerous behaviours that are subserved byhypothalamic functioning and that could beconsidered phylogenetically conserved precur-sors to the content of common obsessions andcompulsions. Oxytocin mediates behaviouralaspects of grooming, maternal care and affilia-tion, and sexual and aggressive behaviours, allof which are components of common obses-sive-compulsive symptoms.45

Caudate nucleusThe tumours of patients 1 and 2 also involvedthe caudate nucleus, which in neuroimagingstudies has been implicated in OCD patho-physiology. Caudate nucleus hypermetabolismthat has been seen in patients with OCDseems to normalise in response to effectivetreatment, which may suggest that the caudatenucleus is involved in either generation orsupression of symptoms of OCD.'5 -'8 Thedestruction of the caudate nuclei by tumour inthese two patients may have producedimpaired inhibitory responses to impulses thatare normally suppressed, thereby releasing, ordisinhibiting, obsessive thoughts and compul-sive behaviours.

ConclusionThe few patients in this series, the compro-mise of multiple brain regions by tumour, andthe likely compromise of additional brainregions by microscopic tumour infiltration,oedema, and hydrocephalus all limit the con-clusions that can be drawn concerning theanatomical localisation of the neural substrateof tic and OCD symptoms. The structuresinvaded by these tumours are neverthelessthose implicated in the pathogenesis of GTSand OCD, both in theory and in previous clin-ical and preclinical studies. Midline structures,

including the corpus callosum, thalamus,brainstem, and limbic system, particularly thehypothalamus and cingulate, are the structuresinvolved commonly in these three patients.The ventral basal ganglia, corpus callosum,thalamus, and brainstem seem to be primarilyinvolved in tic symptom formation and modu-lation, whereas the hypothalamus, cingulate,and possibly caudate nucleus are likely pre-dominantly involved in formation and modu-lation of symptoms of OCD. All thesestructures are either components of CSTC cir-cuitry or else functionally modulate it throughintimate neuroanatomical connections. Thiscommon neurobiological substrate may medi-ate the genetic relatedness of GTS and OCDand thereby explain their common clinicalcomorbidity. The portions of CSTC circuitrythat are affected, or the neuromodulatory cir-cuits that are compromised, may then deter-mine which genetically vulnerable patients willdevelop GTS, OCD, or both.

This work was supported in part by grants MH49351,MH18268, MH30929, NIH RR06022, HD03008, and1K20MH01232-Ol, and by an award from the TouretteSyndrome Association.

1 Pauls DL, Towbin KE, Leckman JF, Zahner GE, CohenDJ. Gilles de la Tourette's syndrome and obsessive-com-pulsive disorder. Evidence supporting a genetic relation-ship. A-ch Geti Psychiatr- 1986;43: 1180--2.

2 American Psychiatric Association. Diagniostic anid StatisticalManual of Menital Disorders. 3rd ed revised. Washington,DC: APA, 1987.

3 Lakke JP, Wilmink JT. A case of Gilles de la Tourette'ssyndrome with midbrain involvement. Y Neurol Neiuriosuri-gPsychiatry 1985;48:1293-6.

4 Price RA, Kidd KK, Cohen DJ, Pauls DL, Leckman JF. Atwin study of Tourette syndrome. Arich Geti Ps3'chiatrr1985;42:815-20.

5 Pauls DL, Raymond CL, Stevenson JM, Leckman JF. Afamily study of Gilles de la Tourette syndrome. Anti 7Hunt1 Genet 1991,48:154-63.

6 Comings DE, Comings BG. A controlled study of Tourettesyndrome. I. Attention-deficit disorder, learning disor-ders, and school problems. An, _7 Humi Genet 1987;41:701-41.

7 Pauls DL, Hurst CR, Kruger SD, Leckman JF, Kidd KK,Cohen DJ. Gilles de la Tourette's syndrome and atten-tion deficit disorder with hyperactivity. Evidence againsta genetic relationship. Archl Get, Psy,chiatry, 1986;43:1177-9.

8 Pauls DL, Leckman JF, Cohen DJ. Familial relationshipbetween Gilles de la Tourette's syndrome, attentiondeficit disorder, learning disabilities, speech disorders,and stuttering. .7 Anti Acad Child Adolesc P.sychiatry1993;32: 1044-50.

9 Leckman JF, Knorr AM, Rasmusson AM, Cohen DJ. Basalganglia research and Tourette's syndromes. TrenidsNeurosci 1991 ;14:94.

10 Leckman JF, Peterson BS. The pathogenesis of Tourette'ssyndrome: epigenetic factors active in early CNS devel-opment. BiOl Psychliatn-s 1993;34:425-7.

11 Wise S, Rapoport JL. Obsessive compulsive disorder: is itbasal ganglia dysfunction? In: Rapoport JL. Obsessi've-comtptulsive disorder i71 chlildreni a;td adolescme,ts. Washington,DC: American Psychiatric Press, 1989:327-46.

12 Balthazar K. Uber das anatomishe substrat der general-isierten tic-krankeit (maladie des tics, Gilles de laTourettte): Entwicklungshemmung des corpus striatum.Arch Psychiatr Nert ekr 1956;195:531-9.

13 Peterson B, Riddle MA, Cohen DJ, Katz t.D, Smith JC,Hardin MT, Leckman JF. Reduced basal ganglia vol-umes in Tourette's syndrome using three-dimensionalreconstruction techniques from magnetic resonance

images. Neurology^ 1993;43:941-9.14 Singer HS, Reiss AL, Brown JE, Aylward EH, Shih B,

Chee E, et al. Volumetric MRI changes in basal gangliaof children with Tourette's syndrome. Neurolsoga 1993;43:950-6.

15 Hyde TM, Aaronson BA, Randolph C, Weinberger DR.Cerebral morphometric abnormalities in Tourette's syn-drome: a quantitative MFI study of monozygotic twins.

Neurology 1995;45:1176- 82.16 Chase TN, Geoffrey V, Gillespie M, Burrows GH. Struc-

tural and functional studies of Gilles de la Tourette syn-drome. Revue Neurologique 1986;142:851-5.

17 Braun AR, Stoetter B, Randolph C, Hsiao JK, Vladar K,Gernert J, et al. The functional neuroanatomy ofTourette's syndrome: an FDG-PET study. I. Regionalchanges in cercbral glucose metabolism differentiating

504


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1996. Dow

nloaded from



patients and controls. Neuropsychopharmnacology 1993;9:277-91.

18 Riddle MA, Rasmusson AM, Woods SW, Hoffer PB.SPECT imaging of cerebral blood flow in Tourette syn-drome. Adv Neurol 1992;58:207-1 1.

19 Alexander G, Crutcher M, DeLong M. Basal ganglia-thala-mocortical circuits: Parallel substrates for motor, oculo-motor, "prefrontal", and "limbic" functions. Prog BrainRes 1990;85:119-46.

20 Hazrati L, Parent A. Contralateral pallidothalamic and pal-lidotegmental projections in primates: an anterogradeand retrograde labeling study. Brain Res 1991 ;567:212-23.

21 Parent A, Hazrati L. Functional anatomy of the basal gan-glia. I. The cortico-basal ganglia-thalamo-cortical loop.Brain Res Rev 1995;20:91-127.

22 Baldwin M, Frost LL, Wood CD. Investigation of the pri-mate amygdala. Movements of the face and jaws.Neuirology 1954;4:596-8.

23 MacLean PD, Delgado JMR. Electrical and chemical stim-ulation of frontotemporal portion of limbic system in thewaking animal. Electroenzcephalogr ClGn Neurophy,siol 1953;5:91-100.

24 Peterson BS, Leckman JF, Duncan JS, Wetzles R, RiddleMA, Hardin M, Cohen DJ. Corpus callosum morphol-ogy from magnetic resonance images in Tourette's syn-drome. Psychiatry Res: Neuroimaging 1994;55:85-99.

25 Sullivan RM, Parker BA, Szechtman H. Role of the corpuscallosum in expression or behavioral asymmetriesinduced by a unilateral dopamine lesion of the substantialnigra in man. Brain Res 1993;609:347-50.

26 McGuire PK, Bates JF, Goldman-Rakic PS. Inter-hemispheric integration: II. Symmetry and convergenceof the corticostriatal projections of the left and the rightprincipal sulcus (PS) and the left and the right supple-mentary motor area (SMA) of the rhesus monkey. CerebCortex 1991;1:408-17.

27 Talairach J, Bancaud J, Geier S, Bordas-Ferrer M, Bonis A,Szikla G, Rusu M. The cingulate gyrus and humanbehavior. Electroencephalogr Clin Neurophysiol 1973;34:45-52.

28 Bonnet KA. Neurobiological dissection of Tourette syn-drome: a neurochemical focus on a human neuroanatom-ical model. In: Friedhoff AJ, Chase TN, eds. Gilles de laTourette syndrome. New York: Raven Press, 1982: 77-82.

29 Kurlan R, Kersun J, Ballantine H Jr, Caine ED.Neurosurgical treatment of severe obsessive-compulsivedisorder associated with Tourette's syndrome. MovDisord 1990;5:152-5.

30 Robertson M, Doran M, Trimble M, Lees AJ. The treat-ment of Gilles de la Tourette syndrome by limbic leuco-tomy. _7 Neurol Neurosurg Psychiatry 1990;53:691-4.

31 Bates JF, Goldman-Rakic PS. Prefrontal connections ofmedial motor areas in the rhesus monkey. .7 Cosnp Neurol1993;336:211-28.

32 Devinsky 0. Neuroanatomy of Gilles de la Tourette's syn-drome. Possible midbrain involvement. Arch Neuirol1983;40:508-14.

33 Nieuwenhuys R, Voogd J, Van Huijzen C. Special sensory

systems. In: The human central nervous system. 3rd ed.New York: Springer Verlag, 1988:165-96.

34 Nieuwenhuys R, Voogd J, Van Huijzen C. Olfactory andlimbic systems. In: The human central nervous systemi. 3rded. New York: Springer Verlag, 1988:293-363.

35 Nieuwenhuys R. Survey of chemically defined cell groupsand pathways. Chemoarchitecture of the brain. New York:Springer Verlag, 1985:7-113.

36 Leckman JF, de Lotbiniere AJ, Marek K, Gracco C, Scahill L,Cohen DJ. Severe disturbances in speech, swallowing, andgait following stereotactic infrathalamic lesions in Gilles dela Tourette's syndrome. Neurology 1993;43:890-4.

37 Rauch SL, Baer L, Cosgrove GR, Jenike MA. Neuro-surgical treatment of Tourette's syndrome: a criticalreview. Comip Psychiatry 1995;36:141-56.

38 Cooper IS. Involuintary mzovement disorders. New York:Harper and Row, 1969:274-79.

39 de Divitiis E, D'Errico A, Cerillo A. Stereotactic surgery inGilles de la Tourette syndrome. Acta Neurochir 1977;24:73.

40 Hassler R, Dieckmann G. Stereotaxic treatment of tics andinarticulate cries or coprolalia considered as motor obses-sional phenomena in Gilles de la Tourette's disease. RevNeurol (Paris) 1970;123:89-100.

41 Korzen AV, Pushkov VV, Kharitonov RA, Shustin VA.Stereotaxic thalamotomy in the combined treatment ofGilles de la Tourette's disease. Zh Nevropatol Psikhiatri1991;91:100-1.

42 Insel TR. Toward a neuroanatomy of obsessive-compulsivedisorder. Arch Gen Psychiatry 1992;49:739-44.

43 Baer L, Rauch SL, Ballantine HT, Martuza R, Cosgrove R,Cassem E, et al. Cingulotomy for intractable obsessive-compulsive disorder. Arch Gen Psychiatry 1995;52:384-92.

44 Leckman JF, Goodman WK, North WG, Chappell PB,Price LH, Pauls DL, et al. Elevated cerebrospinal fluidlevels of oxytocin in obsessive-compulsive disorder.Comparison with Tourette's syndrome and healthy con-trols. Arch Gen Psychiatry 1994;51:782-92.

45 Leckman JF, Goodman WK, North WG, Chappell PB,Price LH, Pauls DL, et al. The role of central oxytocin inobsessive compulsive disorder and related normal behav-ior. Psychoneuroendocrinology 1994;19:742-58.

46 Baxter L Jr, Schwartz JM, Bergman KS, Szuba MP, GuzeBH, Mazziotta JC, et al. Caudate glucose metabolic ratechanges with both drug and behavior therapy for obses-sive-compulsive disorder. Arch Gen Psychiatry 1992;49:681-9.

47 Rauch SL, Jenike MA, Alpert NM, Baer L, Breiter HC,Savage CR, Fischman AJ. Regional cerebral blood flowmeasured during symptom provocation in obsessive-compulsive disorder using oxygen 15-labeled carbondioxide and positron emission tomography. Arch GenPsychiatry 1994;51:62-70.

48 Swedo SE, Pietrini P, Leonard HL, Schapiro MB, RettewDC, Goldberger EL, et al. Cerebral glucose metabolismin childhood-onset obsessive-compulsive disorder.Revisualization during pharmacotherapy. Arch GenPsychiatry 1992;49:690-4.

505


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1996. Dow

nloaded from


three symptom change tourette's and disorderall structures are...

Documents