Letters to the Editor

after admission, a week before the appear-ance of the cutaneous eruption.The pathogenesis of HSV-2 myelitis in

this case remains ill defined. Although IgMantibodies were detected in the first serumsample we are not sure that the patient'smyeloradiculitis reflected true HSV-2 pri-mary infection because reappearance ofIgM antibodies can be detected in herpesvirus reactivation. All but one preriouslyreported case of HSV-2 myelitis were fataland occurred in immunocompromisedhosts. This patient was neither diabetic norHIV-infected, but despite normal CTexaminations of the thorax and abdomen, ahidden malignancy cannot be excluded.Reactivation of a latent infection withindorsal root ganglia neurons with a contigu-ous spread via sacral ganglia to the spinalcord has already been proposed in HSV-2ascending myelitis.' It is of note that despiteantiviral treatment, our patient died withinsix weeks. Thus an immunologicallymediated injury triggered by the herpesinfection could also be involved here, as inanother patient with HSV-1 myelitis whosenecropsy examination showed patchydemyelination of the spinal cord.7

EMMANUEL ELLIEService de Neurologie,

Hopital du Haut-Livique,CHU Bordeaux, FranceFLORE ROZENBERG

Service de Bactiriologie-Virologie,H6pital Saint-Vincent de Paul,

Paris, FranceVINCENT DOUSSET

Service de Neuroradiologie,CHU Bordeaux, France

MARIE BEYLOT-BARRYService de Dermatologie,CHU Bordeaux, France

Correspondence to: Dr Emmanuel Ellie,INSERM U394, rue Camille Saint-Saens, 33077Bordeaux Cedex, France.

1 Britton CB, Mesa-Tejada R, Fenoglio CM,Hays AP, Garvey GJ, Miller JR. A newcomplication of AIDS: thoracic myelitiscaused by herpes simplex virus. Neurology1985;35:1071-4.

2 Tucker T, Dix RD, Katzen C, Davis RI,Schmidley JW. Cytomegalovirus and herpessimplex virus ascending myelitis in a patientwith acquired immune deficiency syndrome.Ann Neurol 1985;18:74-9.

3 Wiley CA, VanPatten PD, Carpenter PM,Powell HC, Thal U. Acute ascendingnecrotizing myelopathy caused by herpessimplex virus type 2. Neurolog 1987;37:1791-4.

4 Iwamasa T, Yoshitake H, Sakuda H, et al.Acute ascending myelitis in Okinawa causedby herpes simplex virus type 2. VirchowsArch A Pathol Anat Histopathol 1991;418:71-5.

5 Iwamasa T, Utsumi Y, Sakuda H, et al. Twocases of necrotizing myelopathy associatedwith malignancy caused by herpes simplexvirus type 2. Acta Neuropathol (Bert) 1989;78:252-7.

6 Ahmed I. Survival after herpes simplex type IImyelitis. Neurology 1988;38: 1500.

7 Klastersky J, Cappel R, Snoneck JM, FlamentJ, Thirty L. Ascending myelitis in associa-tion with herpes simplex virus. N Engl J7 Med1972;287: 182-4.

Peduncular hallucinosis and right hemi-parkinsonism caused by left mesen-cephalic infarction

Since the original description of Lhermittelseveral causes of peduncular hallucinosishave been reported, but always in relationto a bilateral mesencephalic lesion. Wedescribe here a patient with prolonged vis-ual hallucinations and right hemiparkinson-

ism. An MRI showed a unilateral infarctioninvolving the left cerebral peduncle.A previously healthy 70 year old right

handed woman presented with visual hallu-cinations. Two months previously, shebegan, one night, to see objects (motor-bikes), animals (dogs, horses), and people(Japanese) entering and driving silentlyround her room, across the entire visualfield. Although the images were of normalcolours and sizes, she was aware that theywere not real, and never described "dejavu" or "jamais vu" phenomena, tactile, orauditory hallucinosis. The hallucinatoryevents became progressively longer andmore frequent, lasting from minutes tohours, during both day and night. Hermedical history included mild hypertension,normofunctional multinodular goitre, andsurgery for left breast carcinoma six yearsearlier, without subsequent evidence ofrecurrence. There were no other remarkablepersonal or familial antecedents. On exami-nation, she was alert, oriented, and cooper-ative. She remembered four of five wordsafter five minutes, and the mini mental statetest was 32/35. She remained in a left tiltedposture, and showed severe impairment ofpostural reflexes, mild bradykinesia, cog-wheel rigidity, and intermittent restingtremor in the right extremities (mainly inthe lower limb). Tendon reflexes were briskand increased on the right side, but therewas no clonus and the plantar responseswere flexor. There were no other remark-able findings on general or neurologicalexamination. Laboratory investigations,including ESR, routine haematological, bio-chemical, and immunological studies, thy-roid function tests, serological tests forsyphilis, cerebrospinal fluid examination,EEG, and cranial CT were normal or nega-tive. An MRI showed an abnormal highintensity signal in the left cerebral peduncleon T2-weighted images (figure). Multiplefoci of T2-weighted high signal intensitywere also seen throughout the periventricu-lar white matter. Such findings were consis-tent with ischaemic damage. Despitetreatment with haloperidol and phenytointhe patient became more impaired withinthe next several weeks, showing continuousvisual hallucinations, with frequent episodesof agitation and disorientation.

Vascular lesion of the upper brainstem isthe most often cause of peduncular halluci-nosis. A case of peduncular hallucinosis dueto bilateral mesencephalic infarction diag-nosed by MRI has been recently reported.2The present one is the first report, however,to our knowledge, of peduncular halluci-

Axial T2-weightedMR image obtained with a

05-T unit shows an area of hyperintensity inthe left cerebral peduncle, consistent withinfarction.

nosis due to unilateral lesion. Although theprecise anatomical basis for peduncular hal-lucinosis remains unclear, it seems that thesubstantia nigra pars reticulata (SNpr) isdirectly implicated.3 Little is known aboutthe pathogenesis of peduncular hallucinosis.A relation with the rapid eye movement(REM) phase of sleep has been proposed.In this sense, it is known that the SNpr mayplay an important part in the regulation ofthe different phases of sleep through itsconnections with centromedian/parafascicu-lar nuclei of the thalamus, superior collicu-lus, and reticular formation.4

Finally, although we cannot rule out thepresence of brainstem Lewy bodies, it isprobable that the right hemiparkinsonismfound in our patient was related toischaemic damage of the left substantianigra pars compacta.5

RAUL DE LA FUENTE FERNANDEZJOSE MAREY LOPEZ

PABLO REY DEL CORRALService ofNeurology

FERNANDO DE LA IGLESIA MARTINEZDepartment ofInternal Medicine,

HospitalYuan Canalejo,La Coruna, Spain

Correspondence to: Dr Rail de la FuenteFernandez, Servicio de Neurologia, Hospital JuanCanalejo, Xubias de Arriba, 84, 15006 LaCoruina, Spain.

1 Lhermitte J. Syndrome de la calotte du pedon-cule cerebral. Les troubles psycho-sensorielsdans les lesions du mesocephale. Rev Neurol(Paris) 1922;38: 1359-65.

2 Geller TI, Bellur SN. Peduncular hallucinosis:magnetic resonance imaging confirmation ofmesencephalic infarction during life. AnnNeurol 1987;21:602-4.

3 McKee AC, Levine DN, Kowall NW,Richardson EP Jr. Peduncular hallucinosisassociated with isolated infarction of thesubstantia nigra pars reticulata. Ann Neurol1990;27:500-4.

4 Beckstead RM, Frankfurter A. The distribu-tion and some morphological features ofsubstantia nigra that project to the thala-mus, superior colliculus and pedunculopon-tine nucleus in the monkey. Neuroscience1982;7:2377-88.

5 Jellinger K. The pathology of parkinsonism.In: Marsden CD, Fahn S, eds. Movementdisorders. Vol 2. London: Butterworths,1987:124-65.

Intraneural ganglion of the sciaticnerve: detection by ultrasound

Intraneural ganglia are a rare cause ofperipheral nerve lesions most often affectingthe peroneal nerve. Their origin isunknown. Some 50 cases were reported upto 1979.' Sciatic nerve ganglia are very rare.We report a sciatic nerve lesion caused by agiant ganglion situated at the level of thedistal thigh and damaging the tibial portiononly.A 36 year old male right handed and

right footed state officer complained of painin his right calf for about six years especiallywhen jogging and walking for more than 30minutes. He was treated for a lumbar dischernia, but a lumbar CT was unremarkable.Simultaneously he noted discomfort in hisright toes when wearing shoes; this wasrelieved by wearing orthopaedic sandals.On first neurological examination he was

found to have Lasegue's sign with his rightleg at an angle of about 850. Reflexes andsensation in the lower extremities were nor-mal. There was moderate paresis of the toeflexors. Plain radiographs of the lower spineshowed six lumbar vertebrae and a fissuredvertebral arch of the first sacral segment.

870

Letters to the Editor

On electromyographic examination fibrilla-tion potentials and positive sharp waveswere recorded from the abductor hallucisand flexor digitorum brevis muscles. Motorconduction velocity of the right tibial nervewas slow on the right (38 m/s) and normalon the left (55 m/s) side; sensory nerve con-duction velocity of the right sural nerve andH-reflexes bilaterally were normal.

Eleven months later the pain waslocalised to the outer aspect of the right calfand foot, and the ankle jerk was diminishedon the right side. Weakness of the toe flex-ors was unchanged, Hoffmann-Tinel's signwas positive over the posterior tarsal tunneland palpation of the popliteal fossa yieldedneither a mass nor tenderness. Sensorynerve conduction velocity of the rightmedial plantar nerve could not be deter-mined. The distal motor latency of the righttibial nerve was prolonged (7-8 ms).

As the findings were not typical of atarsal tunnel syndrome, surgery was dis-couraged.

Thirteen months later MRI of the lowerleg up to the popliteal fossa showed noabnormalities other than atrophy of thepopliteal and posterior tibial muscles.Electromyography yielded equivocal evi-dence of neurogenic damage to the posteri-or tibial muscle prompting examination ofthe proximal tibial and sciatic nerve by B-scan ultrasound. Here a cystic formation of

about 7 x 3 x 3 cm was found at thebifurcation of the sciatic nerve at the level ofthe distal thigh (figure). A subsequenttransaxial CT confirmed this result.The patient was operated on one week

later. A spindle like multiloculated cystmeasuring roughly 15 x 3 x 2 cm wasfound between the biceps and semitendi-nosus/semimembranosus muscles within thesciatic nerve, producing a gelatinous masson incision. The cyst was excised by amicrosurgical technique leaving the fasciclesof both the tibial and peroneal compart-ment intact. Histologically, a benign gan-glion with pseudocystic walls was found.After the operation paresis of the short andlong toe flexors was more pronounced andthe ankle jerk and tibialis posterior reflexwere absent, accompanied by mild trophicchanges. One year later the pain had disap-peared and walking for two hours was possi-ble without complaint. The ankle jerk wasstill absent, flexion and distention of thetoes paretic, the tips of the toes hypesthetic,and the sole of the foot anhidrotic.

Having first sought the location of thelesion too far cephalic (lumbar disc hernia)it subsequently was suspected too far caudal(tarsal tunnel) because of clinical involve-ment of the toe flexors only. Sciatic nervelesions are more prone to involve the per-oneal than the tibial fascicles and the cystwas missed initially. It was the ultrasound

(A) Longitudinal (left) and transaxial (right)section of the distal thigh by B-scan ultrasound(Siemens Sonoline SL-2). Septated echofreespace occupying mass (black and white solidbold arrows) measuring 2 cmin diameter distending and displacing thesciatic nerve (+, - - and x), which measures7-0 x 4-9 x 9-2 mm. (B) Transaxial CTthrough the distal thigh (Siemens SomatomDR) confirmed a septated cystic spaceoccupying hypodense mass of 2-8 cm indiameter (slim black arrows) less dense (11 to12 Hounsfield units) than the surroundingmuscles (M semimembranosus, semitendinosus,and biceps femoris).

technique which gave the first clue as to thetrue level of the sciatic ganglion.

Although the ganglion in this patient wasseated proximal to the bifurcation of the sci-atic nerve into the tibial and peroneal por-tion, only tibial fibres were affected. Theorigin of the cyst is not clear. A mucoiddegeneration of collagen fibres, metaplasiaof irritated connective tissue, ektopia ofembryonal synovial tissue, or a mucoiddegeneration of a formerly solid tumour forexample, a Schwannoma-are possibilities.

After Mahaley's2 first report of a ganglionof the (posterior) tibial nerve at thepopliteal fossa other ganglia or (Baker's)cysts were also found at the tarsal tunnel.3Only two intraneural synovial cysts of thesciatic nerve were reported between 1980and 1991.4However rare, a ganglion such as this

must be considered in the differential diag-nosis of L5 and SI root, sciatic, tibial andperoneal nerve lesions, as it is amenable tosurgery and results are usually favourable.Like other authors5 we strongly recommendnerve ultrasonography in any case of pro-gressive peripheral nerve damage of unex-plained origin.The MRI was done by Dr V Buchholz and col-leagues, Erlangen.

C J G LANGThe Neurological Hospital

U NEUBAUERS QAIYUMI

R FAHLBUSCHThe Neurosurgical Hospital,

University of Erlangen-Nuremnbergat Erlangen, Germany

Correspondence to: PD Dr Christoph J G Lang,Neurologische Universitatsklinik, Schwabachan-lage 6, D-91054 Erlangen, Germany.

1 Eiras J, Garcia Cosamalon PJ. Intraneuralganglion of the common peroneal nerve.Neurochirurgia (Stuttg) 1979;22:145-50.

2 Mahaley MS Jr. Ganglion of the posterior tib-ial nerve. Case report. .7 Neurosurg 1974:40:120-4.

3 Poppi M, Giuliani G, Pozzati E, Acciarri N,Forti A. Tarsal tunnel syndrome secondaryto mtraneural ganglion. Y Neurol NeurosurgPsychiatry 1989;52:1014-5

4 Gazzeri G, Santucci N. Le cisti sinoviali intra-neurali dello sciatico popliteo estemo.Osservazioni su due casi trattati con tec-niche microchirurgiche. Minerva Chir 1984;39:1143-5.

5 Leijten FSS, Arts W-F, Puylaert JBCM.Ultrasound diagnosis of an intraneural gan-glion cyst of the peroneal nerve. 7 Neurosurg1 992;76: 538-40.

Familial Parkinson's disease and poly-morphism at the CYP2D6 locus

Because 10% of patients with Parkinson'sdisease have an affected relative' and kin-dred with a pattern of disease transmissioncompatible with autosomal dominant inher-itance have been described,2 genetic factorsmay play a part in the pathogenesis of thedisease. Defective 4-hydroxylation ofdebrisoquine by CYP2D6, a member of thecytochrome P-450 family, has been foundin more than 50% of patients with idio-pathic Parkinson's disease, but in less than20% of controls.' Four alleles of theCYP2D6 gene, containing point mutationsor deletions which inactivate the gene,result in a poor metaboliser phenotype.4The mutant allele CYP2D6B in particular,occurs twice as often in the patients than inthe controls, with a relative risk ratio of2 7.5 The risk of idiopathic Parkinson'sdisease in those with the CYP2D6 mutant

871