Palatal myoclonus in a child: herald of acute encephalitis.

To the Editor: In adults, palatal myoclonus occurs months after the onset of a lesion involving the triangle of Mollaret (inferior olive, red nucleus, and contralateral cerebellar nuclei)’; it may be caused by an infarct, demyelination, or a neoplasm.’ There have been few cases in children and none in which palatal myoclonus was the presenting feature of an evolving lesion, as in the following case. C.M., a 5Y2-year-old boy, had been treated for acute lymphocytic leukemia with prednisone, intrathecal methotrexate, and cranial irradiation. He had been in bone marrow and CNS remission for 2 years, with no clinical or C T evidence of latent adverse effects. Except for a mild respiratory infection, he had been well; following a dose of intrathecal methotrexate. he was admitted for the onset of “twitching” eyebrows, dysphagia, and dysarthria. Examination revealed a cushingoid boy with no fever, rash, or striae. He was alert and followed three-step commands, but did not speak and uttered only broken syllables. Cranial nerve functions were normal except for continuous, rhythmic, “blinking” contractions of the orbicularis oculi muscles. The movements were bilateral, sometimes more pronounced on the right. Similar movements of the soft palate were synchronous with the orbicularis movements, at a rate of about 1.5 per second. The eyes were not affected pupils and fundi were normal. Facial strength was equal and full, and a gag reflex was present. Proximal limb wasting and weakness were attributed to steroid myopathy. During the next week, right facial contractions appeared and were asynchronous with the orbicularis and palatal movements. Following this, right hand and toe jerks appeared, also out of phase with the facial movements. The palatal myoclonus became intermittent. After the fifth hospital day, the child became lethargic. Brainstem evoked potentials and brain CT were normal. EEG revealed a left slow-wave focus that followed the facial contractions but preceded limb jerks (figure). Later, the EEG showed encephalopathic changes compatible with the child’s lethargy. On the second day, the CSF fluid was acellular with normal protein and glucose, but the peripheral blood white cell count was less than 1,ooO. Though smears and cultures for bacteria, tuberculosis, and fungi were negative, viral cultures yielded a picornavirus. On the ninth hospital day, CSF was still normal except for myelin basic protein content of 8.4 mg/ml. On the sixteenth day, the CSF showed 15 white cells and protein of 57 mg/ml. By that time, encephalitis was manifest by obtundation, slow EEG, and diffusely increased contrast in cortical areas on CT. By the end of the second week, there were asynchronous myoclonic jerks of the eyes, right side of the face, left shoulder, and right hand and foot. Phenytoin and phenobarbital had no effect; IV diazepam, administered under EEG monitoring, abolished the movements without affecting the EEG. The myoclonus, palatal and generalized, was completely controlled by a combination of clonazepam (0.09 mg/kg/d in three divided doses) and diazepam (1 mg PO tid). A brain biopsy on the seventeenth day revealed microglial nodules suggesting an acute viral infection, but cultures for herpes and picornavirus were negative. The white matter appeared normal with no leukemic infiltrates. Over the following 6 months, the patient remained free of myoclonus, with a normal mental status but with persistent dysarthria and dysphagia as well as a right hemiparesis. Matsuo and Ajax’ suggested denervation supersensitivity as the underlying mechanism for palatal myoclonus. They noted the “obligatory” lag time between onset of a lesion (hemorrhage, infarction, and so on) and the appearance of the palatal myoclonus. The onset of the movement disorder was always at least 2 months after the inciting event, with a mean interval of 10 to 11 months. Thus, in all reported cases, palatal myoclonus has been a new symptom related to an old. established lesion. Our patient had been in remission from acute lymphocytic leuke-

mild respiratory infection, he had been well; following a dose of intrathecal methotrexate. he was admitted for the onset of "twitching" eyebrows, dysphagia, and dysarthria.
Examination revealed a cushingoid boy with no fever, rash, or striae. He was alert and followed three-step commands, but did not speak and uttered only broken syllables. Cranial nerve functions were normal except for continuous, rhythmic, "blinking" contractions of the orbicularis oculi muscles. The movements were bilateral, sometimes more pronounced on the right. Similar movements of the soft palate were synchronous with the orbicularis movements, at a rate of about 1.5 per second. The eyes were not affected pupils and fundi were normal. Facial strength was equal and full, and a gag reflex was present. Proximal limb wasting and weakness were attributed to steroid myopathy.
During the next week, right facial contractions appeared and were asynchronous with the orbicularis and palatal movements. Following this, right hand and toe jerks appeared, also out of phase with the facial movements. The palatal myoclonus became intermittent. After the fifth hospital day, the child became lethargic.
Brainstem evoked potentials and brain CT were normal. EEG revealed a left slow-wave focus that followed the facial contractions but preceded limb jerks (figure). Later, the EEG showed encephalopathic changes compatible with the child's lethargy. On the second day, the CSF fluid was acellular with normal protein and glucose, but the peripheral blood white cell count was less than 1,ooO. Though smears and cultures for bacteria, tuberculosis, and fungi were negative, viral cultures yielded a picornavirus. On the ninth hospital day, CSF was still normal except for myelin basic protein content of 8.4 mg/ml. On the sixteenth day, the CSF showed 15 white cells and protein of 57 mg/ml. By that time, encephalitis was manifest by obtundation, slow EEG, and diffusely increased contrast in cortical areas on CT. By the end of the second week, there were asynchronous myoclonic jerks of the eyes, right side of the face, left shoulder, and right hand and foot. Phenytoin and phenobarbital had no effect; IV diazepam, administered under EEG monitoring, abolished the movements without affecting the EEG. The myoclonus, palatal and generalized, was completely controlled by a combination of clonazepam (0.09 mg/kg/d in three divided doses) and diazepam (1 mg PO tid). A brain biopsy on the seventeenth day revealed microglial nodules suggesting an acute viral infection, but cultures for herpes and picornavirus were negative. The white matter appeared normal with no leukemic infiltrates. Over the following 6 months, the patient remained free of myoclonus, with a normal mental status but with persistent dysarthria and dysphagia as well as a right hemiparesis.
Matsuo and Ajax' suggested denervation supersensitivity as the underlying mechanism for palatal myoclonus. They noted the "obligatory" lag time between onset of a lesion (hemorrhage, infarction, and so on) and the appearance of the palatal myoclonus. The onset of the movement disorder was always at least 2 months after the inciting event, with a mean interval of 10 to 11 months. Thus, in all reported cases, palatal myoclonus has been a new symptom related to an old. established lesion.
Our patient had been in remission from acute lymphocytic leuke- Figure. EEG depicting contraction of the facial muscles preceding the onset of a slow cortical wave, maximal at the left temporal leads. mia for over 2 yeare. He had received intrathecal methotrexate and cranial radiotherapy. There was no CT evidence of leukoencephalopathy, and all the clinical and pathologic evidence pointed to an acute encephalitis.
On electrophysiologic studies, EMG spikes of facial muscles were followed sequentially by a cortical discharge on EEG and the limb contraction (figure). This sequence was consistent with brainstem or subcortical origin of the discharge, spreading sequentially to adjacent facial muscles, cortex, and finally to the limbs. ' The onset of palatal myoclonus in this patient correlated with the acute onset of encephalitis, presumably due to a picomavirus, and was the first sign of brainstem dysfunction. CSF, brainstem evokedpotentials, and CT were normal a t that time. EEC confirmed the subcortical origin of the electrical discharge associated with the myoclonus.
Within 2 weeks, changes in mental status, EEC, and CT implied progression to cortical structures with the appearance of CSF pleocytosis.
Only one other child with palatal myoclonus has been reported: a 7-year-old girl had an ill-defined viral illness 18 months before the onset of myoclonus." In this child, unlike adults, the palatal myoclonus was transient.

Alcohol improves dystonia: Opioid system involvement?
To the Editor: The finding that ethanol infusions1 improve dystonic scores in patients with spasmodic torticollis may provide some insight about neurotransmitter dysfunction in this disorder. Analgesic doses of nitrous oxide also ameliorate the ~ymptome;~ and nitroue oxide interacts with the endogenous opioid system (EOS) in animals:',' and humans5.6 as well as in vitro receptor-binding asmys.7.8 Some of the effects of ethanol may be mediated by the opioid system? and nitroue oxide ameliorates the alcohol withdrawal state.Iu Therefore, the beneficial effects reported by Biary and Koller! may have been produced by stimulation of the EOS.
If so, their data would support our hypothesis that there is an underactivity of the EOS in torticollis."

Johannesburg, South Africa
Reply from the Authors: Dr. Gillman cites data indicating that nitrous oxide may improve spasmodic torticollis, and he suggests that this effect may be mediated by the endogenous opioid system. Furthermore, he hypothesizes that some of the effects of alcohol may be mediated by the spinal system, and that this may be the underlying mechanism for the effects of alcohol in dystonia. However, alcohol has diverse actions and affects many physiologic functions and many neurotransmitter systems. Depending upon the dose, alcohol may stimulate or depress neuronal function in peripheral nerve, spinal cord, Purkinje cell, reticular activating system, and cortical neurom.1 Alcohol alters dopaminergic, noradrenergic, serotonergic, and CABAergic functions.l.2 Attempts to aesociate a neurotransmitter system with a disease are best achieved by using agonists and antagonists that are specific for a particular neurotransmitter system.

Respiratory dyskinesia in Parkinson's disease
To the Editor: De Keyeer and Vinckenl described a patient with Parkinson's disease who had levodopa-induced respiratory disturbance that was suppressed by tiapride, a dopamine receptor blocker. We recently studied a similar patient with levodopa-induced respiratory dyskinesia that was managed by adjusting the dosage of levodopa. which we think is a better approach. A 77-year-old man had Parkinson's disease for 8 yeare, with good control of his symptom by Sinemet 10/100, one-half tablet three timesa day. For 6 months, he had episodes of tachypnea,dyspnea, and perspiration. The epieodee occurred 30 to 40 minutes after each dose of Sinemet and lasted 15 to 45 minutea. Cardiopulmonary evaluation was normal. In the hospital, several respiratory episodes were timerelated to taking Sinemet. There were also orofacial dyskinetic movements during the attacks. Three to 4 h o w after Sinemet, the antiparkinson effects wore off, and he had hypomimia, tremor at rest in the hands and 1egs;moderate cogwheel rigidity, stooped posture, and shuffling gait without armswing.
On the third hospital day, the Sinemet was discontinued, and the respiratory attacks disappeared within 24 hours. After baseline spirometry (6gure 1) the patient was given one tablet of Sinemet 10/100, and within 1 hour, hie reapiratory rate rapidly increased from a baaehne rate of 30 per minute to a rate of 80 per minute (figure 2). In addition, the minute volume increased from 15 l/min to 40 l/min, but arterial blood gases did not change. The respiratory distreea was associated with omfacial dyekineaia and with profbe diaphoresis. By a4justing the Sinemet dosage to one-fourth tablet of 10/100 six times a day, the Parkinson symptoms have been well controlled, and he has had no further respiratory symptoms.
Respiratory dyskinesia is an uncommon complication of levodopa therapy.2 It may be aeeociated with laryngeal dystonia or other dys- kinesias at the time of the levodopapeak-dose effect."The mechanism of levodopa-induced dyskinesias is unknown, but dopamine may activate postsynaptic striatal receptors. Dyskinesia of respiratory muscles may be due to stimulation of the catecholaminergic brainstem nuclei that regulate respiration. The nucleus tractus solitarius contains neurons responsible for rhythmic breathing, and the thyrotropin-releasing hormone (TRH) induces rhythmic bursting of these neurons.4 Levodopa may interfere with the TRH-regulating action on the respiratory neurons, resulting in incoordination of upper airway muscles and expiratory-inspiratory muscles:5,fi The authors1 contend that "not unlike other dopamine antagonists such as haloperidol and pimozide," tiapride reduces levodopa-304 NEUROLOGY 36 F e b r u a r y 1986 induced dyskinesias "without increasing parkinsonian disability." However, tiapride was presumably selected as an antidyskinesia agent because of its preferential effect on the mesolimbic system.' In contrast, the other neuroleptic drugs block the nigrostriatal pathway and thus may have a more adverse effect on the Parkinson symptoms. Despite this theoretical differentiation, tiapride may increase parkinsonian disability.' Our patient illustrates that levodopa-induced dyskinesias can be controlled by merely adjusting levodopa dosage, rather than by blocking the dopamine receptors with tiapride which, despite some reports, might exacerbate parkinsonism. Finally, this dopamine-antagonist is not readily available to clinical neurologists in the United States.

Joseph Jankovic, MD Farid Nour, MD Howton, TX
Reply from the Authors: The respiratory disturbance described by Jankovic and Nour differs in several ways from the one we reported. In their patient, it was a side effect of chronic levodopa therapy, was associated with orofacial dyskinesia, and consisted of a marked increase in respiratory rate. From figure 1. their patient was already tachypneic at baseline spirometry (respiratory rate, 36 per minute), but the reason is not clear from their letter. In our patient, the respiratory disturbance occurred after starting levodopa therapy, was not associated with orofacial or limb dyskinesia, and was characterized by a breathing pattern that was irregular in rate and depth.
Although the underlying mechanism may be completely different, in both cases dyspnea was related to intake of levodopa. We entirely agree that if the side effect can be avoided by simple adjustment of levodopa dosage, that is the most logical approach. In our patient, parkinsonian disability was severe, and doses of levodopa 200 mg three times a day were required to obtain favorable improvement. A t doses of 100 mg of levodopa, she already experienced dyspnea, and a simple adjustment of levodopa dosage would have been not as effective as administration of tiapride. Tiapride may indeed increase parkinsonian symptoms, but much less than claaaic neuroleptic drugs and other substituted benzamides, such as metoclopramide or su1piride.l However, it doee not significantly increase parkinsonian disability if the levodopa dosage is adjusted concurrently.* Careful adjustment of tiapride/levodopa therapy can suppress dyskinesias and allow maximal antiparkinsonian benefit from levodopa.

Jacques De Keyser, MD Walter Vincken, MD Brussels, Belgium
An additional pilomotor seizure To the Editor: In view of the recent interest in pilomotor seizures,l.5 we describe one more case.
A 56-year-old man was admitted because of diffuse headaches for 1 month. For 2 weeks he noted an unusual smell, which was followed by the feeling that his hair was standing on end. This sensation began in the legs and spread to the arms, trunk, and head; it was stronger on the left. He felt anxious, sensed flushing in the face, and was "in a cold sweat" during the 1-minute episodes. Afterwards he felt tired. He sometimes had nausea but no epigastric symptoms. There were no attacks for 1 week, but then they recurred. Examination showed only mildly impaired recent memory. When he was obaerved during an event, there was symmetric horripilation. The vital signs, skin color and temperature, consciousness, speech, and behavior remained normal. Contrast-enhanced CT was normal. EEG was normal, but none was obtained during an attack. The diagnosis was uncinate seizures with autonomic features, and he was given phenytoin, 300 mg daily.
The attacks did not recur.
Three months later he was readmitted after a complex partial seizure that involved the right arm. He had aphasia, left papilledema, right homonymous field cut, and right hemiparesis. CT showed a large left temporoparietal, white-matter, ring-enhancing lesion with marked mass effect. Left parietal craniotomy disclosed a glioblastoma multiforme. He died 2 weeks later. This is the sixth report of pilomotor seizure^.^.^ In four casee, a seizure disorder began with this form of seizurel.".4; in two, other typea of seizure preceded the onset of pilomotor seinuee.:'.s Piloerection was asymmetric in four ca8e8: ipeilatedy predominant in o m , starting ipsilateral to the lesion and generalizing in two,:I,4 and unilateral in 0ne.l Anticonvulsants controlled symptoms in three They had no effect in one, although craniotomy had a transiently beneficial effect.:' Seizures appeared after temporal lobectomy in one patient5 and continued after that procedure in another.' Glioblastoma multiforme was responsible for four of six ~a s e s .~-~ A possible antecedent infection' and prior epilepsy with temporal resec-tion5 were noted in the others. The infiltrative nature of astroglial tumors with preservation of normal structures accounts for the selectivity of manifestations. The high frequency of astrocytomas mean8 they will be the cause of most cases of pilomotor seizures. Pilomotor seizures are more common than is apparent; they may be transient or obscured by more severe symptoms. The symptoms may also not be mentioned by patienta unless specifically sought.'

Cysticercosis acquired in the United States, without compatible travel history
To the Editor: There have been few reporta of acquired cysticercoeis in the United States since 19791-4 and apparently no previous reports of cerebral cysticercosis in an urban-dwelling American who never traveled and whose family members never traveled outaide this country. This 6-year-old girl, well until the evening of admission, suddenly had ditficulty speaking and fell to the right. Minutes later, her eyes deviated to the right, and there were right-sided tonic-clonic move- menta that became generalized. In the emergency mom, the seizure stopped. On examination, she was afebrile, with an extensor plantar response on the right. Thereafter, the hospital course wae unremarkable. She was treated with phenytoin. CT showed an increased density in the periphery of the left parietal lobe, with no enhancement (figure). Skull films were negative. CSF, routine blood studies, and EEG were negative. The serum titer of antibodies for cysticercosis wae 1 : 128; 6 weeks later, it was 1 :64 (normal, less than 1 : 32s). No antibodies were found in CSF. The diagnosis was suspected because of the calcified lesion on CT, the afebrile seizure, and the history of frequent pork ingestion. In one seried of children with cerebral cysticercosis, lesions were seen by CT in all caees, but plain ~kull film were abnormal in only 10%. When there were completely calcified lesions, they were usually found in the peripheral areas of the cerebral hemisphere, as in our case. Our patient had seizures, as in the vast majority of children with ~ysticercosis.~~~ This ~a s e raises several public health concerns. Some pork in the United States may be infected with the larval form, Cysticercus ceklosae. and is not being prepared properly for human comumption. Also, egg-harboring feces of infected individuals may contaminate the food or drink of individuals who have never traveled to endemic regions,I0 which better explains the etiology in our patient since the family denied eating any undercooked meat.
Cerebral cyeticercosis may be diagnosed serologically in more patienta with afebrile seizures and appropriate CT abnormalities, even when there is no compatible travel history.

Dystonia and calcification of the basal ganglia: another case
To the Editor: Larsenl described a family with autosomal dominant dystonia, calcification of the basal ganglia, and no abnormality of calcium metabolism. We studied a child with a similar disorder. He was born in 1973 from healthy parents; there was no familial history or consanguinity of similar disorder. After normal psychomotor development, a t age 6 years, he began to show impaired speech a t age 8, pigmentary degeneration of retina was recognized, ERG was extinguished, and brain CT was normal; a t age 9, there was symmetric progressive axial and segmental dystonia. At age 10, symmetric calcification of basal ganglia was evident on CT (figure). Now, a t age 12 years, his speech is almost incomprehensible because of dysarthria; the limbs are fixed in abnormal dystonic postures.
Dementia is not overt, but it is difficult to evaluate IQ. Plasma and urine amino acids, lipoprotein electrophoresis, oligosacchariduria, leukocyte arylsulfatase A, P-hexosaminidase and B-galactosidase, CSF proteins, EEG. ECG, and ultrastructural skin biopsy examination were normal. Studies of calcium, iron, and copper metabolism revealed no abnormality.
The association of pigmentary degeneration of retina and progressive extrapyramidal symptoms may suggest Hallervorden-Spatz disease, but no diagnostic laboratory test is available, and that diagnosis could be confirmed only by postmortem examination. CT changes reported in Hallervorden-Spatz disease differ from those of our patient, and resemble the patterns of Huntington's disease. ' We do not believe that calcification of the basal ganglia in this case is a coincidence, as it may be in the elderly..'