J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from Journal of Neurology, , and Psychiatry 1997;63:547–558 547

Before treatment tients in this study was too small to give 35 robust answers in this respect. LETTERS TO After treatment Tyramine restriction during combined 30 MAO inhibition is inconvenient. But if THE EDITOR further studies confirm the additional efficacy 25 of combined MAO-A and MAO-B inhibi- tion, then this hindrance is justifiable. 20 * ERNSTNHJANSEN STEUR Moclobemide and selegeline in the 15 Department of Neurology LEO A P BALLERING treatment of depression in Parkinson’s 10 disease Department of Clinical Chemistry, MST Hospital, 5 Enschede, The Netherlands

Monoamine oxidase (MAO) is the predomi- Hamilton depression score Correspondence to: Dr ENHJansenSteur, nant enzyme implied in the catabolism of the 0 Department of Neurology, Hospital Medisch Spec- Moclobemide Moclobemide trum Twente, PO Box 50.000, 7500 KA, Enschede, monoamines , noradrenaline, and and 1 The Netherlands. Telephone +31 53 4872000; fax serotonin. Dopamine is a substrate for both selegeline +31 53 4873100. isoforms of MAO: the extraneuronal MAO-B; and MAO-A which is found both extraneuro- Intensity of depression before and after treatment 2 with moclobemide or moclobemide and 1 DaPrada M, Zürcher G, Wüthrich I, Haefely nally and intraneuronally. Intraneuronal selegeline. MAO-A is active for dopamine, noradrena- WE. On tyramine, food, beverages and the reversible MAO inhibitor moclobemide. J line, and serotonin, which have a paramount Neural Transm 1988;8(suppl 26):31–56. influence in the pathogenesis of depression. 1.8 hours, and in the MOSES group with 5.2 2 Youdim MBH, Riederer P. Dopamine metabo- The combination of MAO-A and MAO-B hours. This diVerence was not significant lism and neurotransmission in primate in relationship to monoamine oxidase A and B inhibition results in a strong increase of both (P=0.671, two sample t test). Baseline clinicians’ rating of depression inhibition. JNeuralTransm1993;91:181–96. dopamine, noradrenaline, and serotonin. De- 3 Lees AJ. on behalf of the Parkinson’s Disease pression is the most frequent psychopatho- (HDRS) was not diVerent between the two Research Group of the United Kingdom. logical finding in Parkinson’s disease. groups. After six weeks, however, the im- Comparison of therapeutic eVects and provement of depression in the MO group mortality data of levodopa and levodopa com- In this study we compared moclobemide was significantly less than in the MOSES bined with selegeline in patients with early, monotherapy with combined therapy with mild Parkinson’s disease. BMJ 1995;311: group (P=0.0029; two sample test, figure). moclobemide and selegeline, under tyramine t 1602–7. The main phenomenology of the depres- 4 Sieradzan K, Channon S, Ramponi C, Stern restriction. sive illness was a depressed mood with GM, Lees AJ, Youdim MBH. The therapeutic Ten patients with idiopathic Parkinson’s potential of moclobemide, a selective monoam- psychomotor retardation. disease of two to 15 years in duration ine oxidase A inhibitor in Parkinson’s disease. J The scores of MMSE disclosed no diVer- 1995; (suppl 2):51S– contributed to the study. The actions of Clin Psychopharmacol 15 ences in both groups at entry in the study. At 59S. moclobemide and selegeline were described the end of the study the score was unchanged 5 Takats A, Tarczy N, Simo M, Szombathely E, in the informed consent. Inclusion criteria Bodrogi A, Karpati R. Moclobemide/aurix in the MO group, and there was a slight but were Parkinson’s disease disability Hoehn treatment in Parkinson’s disease with depres- significant increase of MMSE score in the sion. New Trends in Clinical Neuropharmacology and Yahr stages II-IV, major depression using MOSES group (P=0.0479; two sample t 1994;8:260. 6 Knegtering H, Eyck M, Huysman P. EVects of DSM III.R criteria, mini mental state test). examination (MMSE) score of 21 or above, antidepressants on cognitive functioning of There was no increase in blood pressure. In elderly patients. Drugs Aging 1994;5:192–9. and self recorded “on-oV” patient diaries. A both groups one patient reported sympto- 7 Roth M, Mountjoy CQ, Amrein R. Mo- cut-oV point of 21 on MMSE score was used matic hypotension. clobemide in elderly patients with cognitive to allow study of patients with Parkinson’s decline and depression. Br J Psychiatry 1996; The therapeutic potential of selective inhi- 168:145–57. disease without substantial cognitive impair- bition of MAO B with selegeline in Parkin- ment. All patients had been on optimal and son’s disease has been documented for many stable dosage of antiparkinson medication years, with a clear capacity to postpone the and patients with atypical during the previous three months. need for levodopa in early Parkinson’s presentations of sporadic Creutzfeldt- The present major depression had a dura- disease. The neuroprotective properties of Jakob disease http://jnnp.bmj.com/ tion of three to 15 months. Antidepressant selegeline are controversial or inconclusive.3 medical treatment had not been initiated in Selective MAO-A inhibition in Parkinson’s Brain biopsy, or , or both are any of these 10 patients before entering this disease has not been examined in detail. Sier- performed to exclude treatable neuropatho- study. adzan et al4 found a mild symptomatic eVect logical disorders in a range of cases in which Moclobemide (600 mg) with or without of moclobemide in non-depressed patients clinical, neuroradiological, and other investi- selegeline (10 mg) was administered during with Parkinson’s disease. Takats et al5 re- gative findings fail to define a diagnosis. The the six week study period. Regular antipar- ported an antidepressant eVect of mo- description of a new clinicopathological vari- kinson medication was kept unchanged. Fol- clobemide in depressed patients with Parkin- ant of Creutzfeldt-Jakob disease shows that stein’s mini mental state examination score, son’s disease. Moclobemide lacks some cases of this rare disease may present on September 26, 2021 by guest. Protected copyright. patients’ “on-oV” diaries during five days, anticholinergic and hepatotoxic eVects, and is atypically and the diagnosis in life may and the Hamilton depression rating scale even able to antagonise the cognitive impair- depend on brain biopsy.1 Although the (HDRS) for depression were assessed one ment resulting from cholinergic blockade.6 Department of Health guidelines on neuro- week before and after six weeks of treatment; The present study confirms the antidepres- surgical and ophthalmic operative procedures MMSE and HDRS were scored in the “on” sant eVects of moclobemide in depressed minimise any risk of subsequent iatrogenic state of motor response to levodopa. Patients patients with Parkinson’s disease. The combi- transmission 6% of cases of Creutzfeldt- were randomly assigned to moclobemide nation of moclobemide and selegeline, how- Jakob disease present with a -like disor- monotherapy or to the combination of ever, did have a more pronounced eYcacy on der with equivocal clinical, electrophysiologi- moclobemide and selegeline. These two mood and cognitive performances. Increased cal, or neuroradiological support for the treatment groups are further referred to as concentrations of the monoamines by com- diagnosis.2 There is, therefore, the potential the MO group and the MOSES group. bined MAO inhibition is a plausible reason danger of iatrogenic transmission of Bradykinesia and start hesitation improved for the superior eYcacy of moclobemide/ Creutzfeldt-Jakob disease after inadvertent in six patients, one patient in the MO group selegeline treatment over monotherapy with brain biopsy of such patients. The Depart- and all patients in the MOSES group. Brady- moclobemide in depressed patients with Par- ment of Clinical Neurosciences in Edinburgh kinesia deteriorated in one patient (MO kinson’s disease. therefore, created a dedicated set of neuro- group). increased in four patients, The eVect of combined MAO inhibition on surgical instruments designated purely for one in the MO group and three in the the MMSE scores was significant. The influ- open brain biopsy. These instruments were MOSES group. ence of MAO-A and MAO-B inhibition on repacked and sterilised after each biopsy. There were no significant diVerences for cognitive functions is becoming of increasing This precaution was vindicated by the the percentage “on” hours per day between importance in recent years. The cognitive following case. the MO group and the MOSES group at improvement with moclobemide is parallel A 65 year old man presented with a history entry and at the end of study. Waking hours to, but independent from, the establishing of of headache for some months followed by the spent “on” increased in the MO group with antidepressant eVects.7 The number of pa- acute onset of expressive dysphasia and right J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from 548 Letters, Correspondence, Correction, Book reviews . There was subsequently some 1 Will GR, Ironside JW, Zeidler M, et al.Anew Transient global amnesia provides a model progression in the weakness and he was varient of Creutzfeld-Jakob disease in the UK. of severe, yet transient, amnesia.6–8 Findings Lancet 1996;347:921–5. admitted to hospital for investigation. There 2 McNaughton H, Will R. Creutzfeldt-Jakob dis- during transient global amnesia have con- was a history of previous stroke in June 1989 ease presenting as stroke. Analysis of 30 cases. firmed that there is a profound loss of antero- presenting with expressive dysphasia and Ann Neurol 1994;36:313. grade episodic memory for both verbal and 3 Hayward PAR, Bell JE, Ironside JW. Prion pro- non-verbal material, together with a variable hemiplegia which recovered completely over tein immunocytochemistry. The development two months. The patient was a regular of reliable protocols for investigation of extent of temporally graded remote memory smoker and had previously had an operation Creutzfeldt-Jakob disease. Neuropathol Appl deficit. By contrast, working and semantic for excision of an arterial embolus. Clinical Neurobiol 1994;20:375–83. memory seem normal.9 These findings are of 4 Budka H, Aguzi A, Brown P, et al. Neuropatho- examination disclosed severe expressive dys- logical diagnostic criteria for Creutzfeldt-Jakob interest from a theoretical perspective: they phasia although his eyes opened appropri- disease and other human spongiform encepha- confirm the independence of memory sub- ately to speech and the patient flexed his lopathies (prion diseases). Brain Pathol 1995;5: systems, and also imply that the pathological 459–66. process in transient global amnesia is con- limbs. There was a spastic right hemiparesis 5 Hsich G, Kennedy K, Gibbs CJ, et al. The with bilateral extensor planter responses. 14-3-3 brain protein in CSF as a marker for fined to those structures known to be critical Brain CT showed irregular low density in the transmissible spongiform encephalopathies. N for the establishment of new episodic memo- Engl J Med 1996;335:924–30. right cerebellum and low density in the left ries (the medial temporal complex, the dien- frontoparietal region. After admission, the cephalon, and the basal forebrain). In sup- patient developed frequent partial seizures Transient semantic amnesia: a new port of this hypothesis, functional brain imaging studies have shown hypoperfusion of and myoclonus and in view of the diagnostic syndrome? doubt, an open brain biopsy was carried out components of the episodic memory system, 10 on 29 April 1992. Pathological examination The term transient global amnesia was first particularly the medial temporal lobe. of this tissue (1 cm3) revealed infarcted and coined by Fisher and Adams in 1964 to If transient global amnesia represents a ischaemic cortical and subcortical tissues describe a clinical syndrome characterised by temporary disruption of episodic memory with no evidence of vasculopathy or vasculi- the abrupt onset of severe amnesia usually then it should, in principle, be possible to tis. No spongiform change was present in the accompanied by repetitive questioning, oc- identify a syndrome of transient impairment biopsy and no amyloid plaques were present. curring in middle aged or elderly people, and of semantic memory. There have in recent Immunocytochemistry for prion protein was lasting several hours.1 The aetiology of years been scattered reports of various atypi- negative.3 The patient continued to deterio- transient global amnesia has been a topic of cal forms of transient amnesia including transient verbal amnesia11 rate postoperatively and died eight weeks debate, but in the past few years the situation and transient topo- graphical amnesia,12 but only one with any later at another hospital. Necropsy was has become considerably clearer with the resemblance to the hypothesised syndrome of performed and histological examination of emergence of the following diagnostic crite- transient semantic memory loss; Kapur 13 the brain led to the diagnosis of Creutzfeldt- ria: (1) attacks must be witnessed and et al described a patient with temporary loss of Jakob disease. There were spongiform information available from an observer who memory for people which they suggested changes evident in a wide distribution was present for most of the attack, (2) there might be an analogue of the permanent form throughout the and prion must be clear cut anterograde amnesia during of memory loss of people associated with protein desposition shown by immunocyto- the attack, (3) clouding of consciousness and right temporal lobe pathology.14 This could chemistry in a compositive synaptic and loss of personal identity must be absent and be regarded, therefore, as a selective loss of a perivascular pattern. In addition, numerous the cognitive impairment limited to amnesia, specific subcomponent of semantic knowl- infarcts were noted throughout the brain, (4) there should be no accompanying focal edge. The present report details, for the first particularly in the right occipital lobe, the neurological symptoms and functionally rel- evant focal signs, (5) epileptic features must time, a patient with transient loss of general right temporal lobe, and the . The (non-person specific) semantic memory. brain biopsy tray which had been repacked be absent, (6) attacks must resolve within 24 hours, and (7) patients with recent head The patient, a 50 year old printer’s and resterilised had deliberately not been injury or known active epilepsy are assistant, was admitted to Addenbrooke’s used on another patient and the instruments excluded.23 Epidemiological studies which Hospital in October 1995 with an acute loss were subsequently destroyed. have applied these criteria have established of memory for word and object meaning. He This case highlights three major points. that thromboembolic cerebrovascular disease was in excellent general health without risk Firstly, some cases of Creutzfeldt-Jakob plays no part in the causation of transient factors for cerebrovascular disease, but had disease will present in an atypical fashion. In global amnesia, but the incidence of migraine lifelong migraine; typical attacks consisted of http://jnnp.bmj.com/ these patients, because Creutzfeldt-Jakob in patients with transient global amnesia is hemicranial headache and nausea followed by disease is not high on the diVerential diagno- greater than would be expected in the general unilateral facial numbness. There was no his- sis, brain biopsy may be required to exclude population. A small minority of cases with tory of psychiatric illness and there had been treatable causes of progressive dementia. unusually brief, and recurrent, attacks even- no recent stressful life events. Secondly, brain biopsy in Creutzfeldt-Jakob tually manifest temporal lobe epilepsy.3 On the evening of his admission, he had disease may fail to confirm the diagnosis Turning to the neuropsychological aspects, been working on his stock car in the garage because the pathological lesions of much has been learned about the organisation when he developed a typical migraine. At the Creutzfeldt-Jakob disease may be unevenly and neural basis of human memory since the crescendo of the headache, his father in law distributed within the brain and not detect- initial description of transient global amnesia. entered the garage and asked him if he on September 26, 2021 by guest. Protected copyright. able in small biopsy specimens using stand- It is clear that memory is not a unitary wanted a marrow from his garden for supper. ardised neuropathological diagnostic function but consists of interactive systems.45 The patient recalls being puzzled by this criteria.4 Thirdly, for all patients undergoing One broad distinction, which derives from question and said “what’s a marrow?” The brain biopsy, we think that a dedicated set of experimental studies of patients with amnesia, family understandably thought that he was instruments is appropriate as there may be a is that between explicit and implicit memory; joking with them, but he then seemed to be delay between performing the operation and explicit memory refers to memory available to having diYculty understanding other com- obtaining a definitive diagnosis as occurred in conscious access; implicit memory, by con- mon terms. They searched the house and this case. Such a precaution minimises the trast, refers to various types of learned found a cucumber which they presented to risk of inadvertent iatrogenic transmission. responses such as conditioning, priming, and the patient to show him what a marrow was Hopefully, in the not too distant future, motor skills which are not available to like: he was unable to name or identify the advances in “premortem”, non-surgical diag- conscious reflection. Explicit memory can be cucumber. They repeated the process with nosis of Creutzfeldt-Jakob disease,5 will subdivided into short term (working) and other fruit and vegetables all of which he was render brain biopsy redundant. long term memory; episodic and semantic unable to name. In conversation he had diY- memory are components of long term culty understanding terms such as “stock IAN R WHITTLE ROBERT G WILL memory. The first refers to memory for car” and “engine”, but his speech was fluent JAMES W IRONSIDE personally experienced and highly temporally and grammatically correct. He was aware that Department of Clinical Neurosciences, and specific events or episodes. Semantic memory he had two children but could not think of Neuropathology Laboratory, Western General Hospital, refers to the permanent store of representa- their names. Because of the family’s increas- Edinburgh, EH4 2XU, UK tional knowledge which includes facts, con- ing concern, their general practitioner ar- Correspondence to: Mr I R Whittle, Department of cepts word, and object meaning. It can be ranged for admission to hospital. Clinical Neurosciences, and Neuropathology Labo- considered as the “database” which we draw When seen by the admitting general medi- ratory, Western General Hospital, Edinburgh, EH4 on to give meaning to our sensory experi- cal team he was alert and oriented in time and 2XU, UK. ences. place. He could recount his name and J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from Letters, Correspondence, Correction, Book reviews 549 performed rapid serial subtraction. Regis- the introduction, there is now an established 11 Nishiyama KHK, Bandoh M, Ishikawa T, Sug- tration of a name and address was normal, association between transient global amnesia ishita M. Transient partial verbal amnesia. J Neurol Neurosurg Psychiatry 1993;56:1234–7. but he was unable to name the prime minis- and migraine. A wide range of physical and 12 Stracciari A, Lorusso S, Pazzaglia P. Transient ter or monarch, and said to the examiner emotional stressors have been linked to tran- topographical amnesia. J Neurol Neurosurg Psy- “what’s a monarch?” There was no evidence sient global amnesia and a unifying hypoth- chiatry 1994;57:1423–5. of repetitive questioning and the patient gave esis was proposed by Olesen and Jorgensen.20 13 Kapur N, Katifi H, El-Zawawi H, Sedgewick M, Barker S. Transient memory loss for people. J an accurate history of the events over the past They suggested that the phenomenon of Neurol Neurosurg Psychiatry 1994;57:862–4. few hours. Physical examination was normal. spreading depression could be precipitated 14 Evans JJ, Heggs AJ, Antoun N, Hodges JR. Pro- The next day, he was reviewed in the Neu- by intense volleys of sensory inputs to the gressive prosopagnosia associated with selec- rology Department. Again orientation and , releasing the tive right temporal lobe atrophy: a new syndrome? Brain 1995;118:1–13. verbal anterograde memory were normal. glutamate; glutamate is a potent cause of 15 Warrington EK. Selective impairment of seman- Digit span was recorded as six forwards and spreading depression in animals. It was tic memory. Q J Exp Psychol 1975;27A:635– four backwards. His spontaneous language subsequently suggested that this may, there- 57. fore, provide a link between the diverse 16 Snowden JS, Goulding PJ, Neary D. Semantic was normal and he was able to name 12 dementia: a form of circumscribed cerebral everyday common objects. Category fluency precipitating factors reported to provoke atrophy. Behav Neurol 1989;2:167–82. for animal names was, however, reduced in transient global amnesia. These stressful 17 Hodges JR, Patterson K, Oxbury S, Funnell E. that he produced only eight such exemplars. stimuli may act by causing a surge of excito- Semantic dementia: progressive fluent aphasia toxic neurotransmitter which then temporar- with temporal lobe atrophy. Brain 1992;115: Brain CT and EEG were normal and he was 1783–806. discharged from hospital. ily shuts down normal memory function in 18 Hodges JR, Patterson K. Non-fluent progressive When I reviewed the patient one month the hippocampus.3 A similar mechanism aphasia and semantic dementia: a comparative later he was able to give a lucid account of the could be evoked as the cause of transient neuropsychological study. J Int Neuropsychol semantic amnesia, although the neurotrans- Soc 1997;2:511–25. events surrounding his admission to hospital 19 Mummery CJ, Patterson KE, Hodges JR, Wise and remembered being unable to name and mitter systems involved in the lateral tempo- RJS. Generating ‘tiger’ as an animal name and identify vegetables and other things shown to ral neocortex are less clearly established. a word beginning with T: diVerences in brain. It is possible that transient semantic amne- Proc R Soc Lond B Biol Sci 1996; 263:989–95. him by his family. He also recalled accurately 20 Olesen J, Jorgensen MB. Leao’s spreading the sequence of events relating to his hospital sia may be more common than has been rec- depression in the hippocampus explains tran- admission. Cognitive assessment showed him ognised. In transient global amnesia, the pro- sient global amnesia. Acta Neurol Scand 1986; to now be normal. His verbal fluency for ani- found loss of anterograde episodic memory 73:344-352. mal names had risen from the previously results in repetitive questioning and a loss of impaired level to 18. More formal neuropsy- all new information after a few seconds. Such an attack will not escape the notice of observ- Alopecia with carbamazepine in two chological assessment disclosed a level of ver- patients with focal seizures bal intellectual and memory performance ers and will lead to medical attention. commensurate with his predicted back- Impairment of semantic memory results in more subtle cognitive dysfunction without In the treatment of patients with epilepsy ground education and IQ assessed by reading with antiepileptic drugs, alopecia was re- age. the dense lacuna seen in transient global amnesia, as illustrated here. ported in association with valproic acid The features of the attack suggest a 1 In conclusion, the present report describes (VPA), and on rare occasions with car- transient loss of semantic memory with pres- bamazepine (CBZ).23 We recently encoun- ervation of anterograde episodic and working a patient with what seems to be temporary reversible impairment of semantic memory tered two young women with partial seizures memory, a syndrome which has hitherto been who developed alopecia after starting CBZ. unrecognised. with preservation of anterograde memory function. Patient 1 was a 32 year old woman with a The syndrome of progressive, yet selective, diagnosis of left frontal lobe epilepsy since the loss of semantic memory was first recognised JOHN R HODGES age of 27. Her habitual seizures were simple by Warrington.15 More recently, the term University of Cambridge Neurology Unit, partial seizures consisting of tonic contraction semantic dementia has been applied to this Addenbrooke’s Hospital, Hills Rd, of her right hand and bilateral orbicularis oculi syndrome, the core features of which are as Cambridge CB2 2QQ, UK and MRC Applied Psychology Unit, muscles, followed by eye deviation to the right follows: (1) selective loss of semantic memory 15 Chaucer Rd, and clonic convulsion of the right side of her causing severe anomia, decreased word com- Cambridge CB2 2EF,UK. face, which were not usually associated with prehension, impaired production of exem- disturbance of consciousness. A cranial MRI

Correspondence to: Professor J R Hodges, MRC http://jnnp.bmj.com/ plars on tests of word fluency, and a loss of Applied Psychology Unit, 15 Chaucer Rd, Cam- was normal and routine EEG showed normal general knowledge, (2) sparing of other bridge CB2 2EF, UK. background activity and intermittent irregular aspects of language, notably syntax and pho- 2-3 Hz slow activity at the left frontotemporal nology, (3) unimpaired perceptual and non- area seen every 50 to 100 seconds. As the sei- verbal problem solving abilities, and (4) pre- 1 Fisher CM, Adams RD. Transient global amne- zures had started occurring often, the patient served anterograde episodic memory.16–18 sia. Acta Neurol Scand 1964;40(suppl 9):1-83. had been started on VPA (1200 mg/day) and 2 Hodges JR, Warlow CP. Syndromes of transient studies in semantic dementia amnesia: towards a classification. A study of diazepam (6 mg/day). Because alopecia have consistently shown atrophy of the 153 cases. J Neurol Neurosurg Psychiatry developed over a period of several months, temporal lobe involving particularly the left 1990;53:834–43. these two drugs were discontinued, and then 3 Hodges JR. Transient amnesia: clinical and on September 26, 2021 by guest. Protected copyright. inferolateral neocortex with sparing of the neuropsychological aspects.London: JB Saunders, the hair loss stopped. As her seizures became hippocampal formation.17 18 Parallel func- 1991. less frequent despite discontinuing VPA and tional activation studies, using PET, have also 4 Squire LR. Mechanisms of memory. Science diazepam, no medication had been taken in pointed to a key role for the inferolateral left 1986;232:1612–9. the next year until she visited our hospital, at 19 5 Baddeley AD. Human memory: theory and temporal lobe in semantic memory. practice.Hove: Lawrence Erlbaum Associates, which time the seizures had started occurring It can be seen from this brief description of 1990. once every month or every other month. The semantic dementia that the syndrome de- 6 Hodges JR, Ward CD. Observations during patient was then given CBZ, increased the scribed in the present paper seems to be a transient global amnesia: a behavioural and dose gradually up to 300 mg/day in two neuropsychological study of five cases. Brain transient analogue of selective semantic 1989;112:595–620. months, and the steady state trough blood memory loss. In the same way that transient 7 Kritchevsky M, Squire LR. Transient global concentration was 5.0 mg/l. Three months global amnesia is a temporary form of the amnesia: evidence for extensive, temporally after starting CBZ, the patient developed graded retrograde amnesia. Neurology 1989;39: classic amnesic syndrome resulting from 213–8. appreciable hair loss, as seen previously with reversible dysfunction of structures critical 8 Evans J, Wilson B, Wraight EP, Hodges JR. VPA. After CBZ was switched to phenobarbi- for episodic memory, transient semantic Neuropsychological and SPECT scan findings tone, hair shedding stopped. The patient had during and after transient global amnesia: amnesia is likely to result from a temporary evidence for the diVerential impairment of no pregnancy, other dermatological prob- dysfunction of the inferolateral temporal remote episodic memory. J Neurol Neurosurg lems, endocrine disorders, or autoimmune lobes. Of course, this remains a speculative Psychiatry 1993;56:1227–30. disturbance such as systemic lupus erythema- hypothesis as functional imaging studies 9 Hodges JR. Semantic memory and frontal tosus throughout the clinical course. executive function during transient global (SPECT or PET) were not obtained during, amnesia. J Neurol Neurosurg Psychiatry 1994; Patient 2 was a 25 year old woman with a or after, the present patients’ attack. 57:605–8. diagnosis of tuberous sclerosis associated It is of interest that our patient was a 10 Zeman AZ, Hodges JR. Transient amnesic syn- with complex partial seizures since the age of lifelong migraineur and that the attack dromes. In: Berrios GE, Hodges JR, ed. 17. She had several small Pringle adenomas Memory disorders: the neuropsychiatric perspec- occurred in the context of a particularly tive. Cambridge: Cambridge University Press, on her face and a depigmented macule on her severe episode of migraine. As mentioned in 1997 (in press). trunk, but otherwise no skin lesions, includ- J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from 550 Letters, Correspondence, Correction, Book reviews ing in her hair, were seen. Her intellectual association with CBZ, and what the patient this when she was investigated for anaemia at function was within the normal range, and population with this side eVect was. Although another hospital and an abdominal ultrasound her social activity was normal. Multiple angio- alopecia resulting from treatment with CBZ disclosed hyposplenism. Reticulin antibodies myolipomas in both kidneys, plaques in the is not often seen, attention must be given to were present. She was started on a gluten free retina, and a small calcified tuber by the left epileptic patients, especially when precipitat- diet and the haematological indices returned lateral ventricle were found. Her habitual sei- ing factors of hair loss are present, and also to normal. Vitamin E concentrations were zures consisted of loss of awareness, incoher- especially in young women. normal. Her other medical history included ent speech, staring, and right hand dystonic pulmonary tuberculosis at the age of 22. posture. Routine EEG showed normal back- This study was supported by grants-in-aid for On examination she had spontaneous, and ground activity. Prolonged EEG monitoring scientific research (A) 06404031 and (C) 08279106 reflex and action myoclonus in the upper recorded ictal rhythmic discharges mainly at from the Japan Ministry of Education, Science, limbs and, to a lesser extent, in the leg. In Sport and Culture, a research grant for treatment of addition, there was a postural upper limb the right posterior quadrant. The patient had intractable epilepsy from the Japan Ministry of been on VPA (200 mg/day) and pheno- Health and Welfare, and a research grant from the tremor and finger-nose bilaterally. Her barbitone (80 mg/day) for more than six Epilepsy Research Foundation. gait was ataxic and she walked with a frame. months, with steady state trough blood AKIO IKEDA Tendon reflexes were normal and plantar concentrations of 66.6 mg/l and 20.2 mg/l HIROSHI SHIBASAKI responses flexor. Vibration sensation was respectively. VPA was switched to CBZ to Department of Brain Pathophysiology absent below the knees, but proprioception in achieve better seizure control, that was AIKO SHIOZAKI the big was normal. Neurophysiological increased up to 600 mg/day in one month. JUN KIMURA studies confirmed multifocal spontaneous, Two months afterwards, when the blood con- Department of Neurology, Kyoto University School of reflex and action myoclonus particularly of centration of CBZ was 6.5 mg/l, the patient Medicine, Shogoin, Sakyo-ku, Kyoto, 606, Japan the hands. Back averaging from the action had appreciable hair loss, and especially over Correspondence to: Dr Akio Ikeda, Department of jerks disclosed time locked cortical activity the front of her head her hair became very Brain Pathophysiology, Kyoto University School of over the contralateral hemisphere, preceding sparse. When the dose of CBZ was reduced Medicine, Shogoin, Sakyo-ku, Kyoto, 606, Japan. each jerk by 27 ms. Somatosensory evoked to 200 mg/l in two weeks, moulting stopped, potentials after finger stretch were pathologi- cally enlarged. and new hair began to grow. No other skin 1 Jeavon PM, Clark JE. Sodium valproate in treat- lesions developed simultaneously, and no ment of epilepsy. BMJ 1974;2:584–6. The clinical and neurophysiological char- white tuft hair was seen. Throughout the 2 Shuper A, Stahl B, Weitz R. Carbamazepine- acteristics of her condition conform closely to induced hair loss. Drug Intelligence and Clinical those previously reported in patients with clinical course, the patient had no pregnancy, Pharmacy 1985;19:924–5. endocrine disorders, or autoimmune distur- 3 Mattson RH, Cramer JA, Collins JF, et al.A coeliac disease and the Ramsay Hunt syn- bance. CBZ had been given in a diVerent comparison of valproate with carbamazepine drome, except that the rate of progression of hospital one year before the present episode, for the treatment of complex partial seizures her disease has been much slower. Of 11 and secondarily generalized tonic-clonic sei- cases reported in the literature,1–3 seven have but at that time the patient took the drug only zures in adults. N Engl J Med 1992;327:765– for a week, because it caused dizziness and 71. been unable to walk or stand or have died ataxia. 4 Holmes GL. Carbamazepine toxicity. In: Levy within two years of the onset of their RH, Mattson RH, Meldrum BS, eds. Antiepi- neurological illness. The longest reported CBZ often causes allergic reactions such as leptic drugs. 4th ed. New York: Raven Press, 1995:567–79. survival after the onset of neurological symp- skin rash, and also systemic lupus erythema- 3 tosus, both of which can potentially cause 5 Drug causes of hair loss. Drug Ther Bull toms has been six years (case 1 of Lu et al )at 1978; :77–9. hair loss.4 Pregnancy, fever, thyroid dysfunc- 16 which time the patient was unable to walk. tion, and malnutrition are also potential Our patient has now had neurological symp- causes of hair loss.5 In the present two toms for more than 13 years and remains Neurological manifestations of coeliac ambulant, albeit with a frame. patients, there were no such accompanied disease The present case shows that myoclonic causes to explain their hair loss. In the pre- encephalopathy associated with coeliac dis- vious case report of hair loss associated with The association between progressive myo- ease does not invariably have a rapidly CBZ, it occurred a week after CBZ was clonic encephalopathy (Ramsay Hunt syn- progressive course. This case and the recent initiated.2 Human hair follicles have two drome; progressive myoclonic ataxia) was report of antigliadin antibodies in 17 out of phases: an anagen (growth) phase of four recently the subject of a clinicopathological 25 patients with ataxia of unknown cause (of years, and then a telogen (resting) phase of report in this Journal1 and of a review by

whom four subsequently had abnormal small http://jnnp.bmj.com/ three months, followed by normal hair loss.5 2 Bhatia et al. A notable feature of these cases bowel biopsies consistent with coeliac dis- Therefore, the onset of the hair loss within a was the rapidly progressive nature of the ease4) suggests that coeliac disease should be week of CBZ administration was interpreted encephalopathy. Most patients were unable to considered in all patients with ataxia, particu- as a disturbance of growing of hair follicles walk or died within two years of the onset of larly if there is myoclonus. (anagen phase).2 However, in the present their neurological symptoms. The patient report, it took two and three months in that we report is therefore remarkable in that GDPSMITH patients 1 and 2 respectively, until hair loss her neurological illness began 13 years ago. G SALDANHA started. This can be explained by the prema- Progressive myoclonic encephalopathy asso- Department of Neurology, Guy’s Hospital, St Thomas Street, London SE1 9RT,UK ture ending of the anagen phase, as usually ciated with coeliac disease may take a more seen in treatment with antithyroid drugs, benign course and we suggest that the condi- T C BRITTON on September 26, 2021 by guest. Protected copyright. anticoagulants, and VPA.5 This type of hair tion should be considered in all patients with Neurosciences Unit, King’s College Hospital, Denmark loss occurs about three months after the , particularly if there is myo- Hill, London SE5 9RS, and Institute of Psychiatry, de Crespigny Park, London SE58AZ, UK insults—namely, at the end of the telogen clonus. phase,5 as seen in the present cases. This may The patient, now aged 64, presented at the P BROWN explain why in patient 2 hair loss was not rec- age of 55 with a four year history of diYculty The National Hospital for Neurology and ognised when she had taken CBZ for a week with fine finger movements in both hands, Neurosurgery, Queen Square, London WC1N 3BG, UK previously. Both of the present patients had worse on the right such that she had difficulty been on VPA previously, and patient 1 had picking up coins and doing up buttons. Her Correspondence to: Dr TC Britton, Neurosciences hair loss most likely caused by VPA. This did had also become slower. Examin- Unit, King’s College Hospital, Denmark Hill, Lon- don SE5 9RS. not happen in patient 2. Thus it suggests that ation at the age of 55 disclosed a jerky upper hair loss caused by CBZ and VPA does not limb tremor with myoclonus but no gait 1 Fletcher NA, Ironside JW, Warlow CP. Progres- necessarily occur by a common mechanism. ataxia. The tremor and myoclonus have slowly sive ataxia, focal seizures, and malabsorption The common characteristics in these two worsened and she has developed an ataxic syndrome in a 41 year old woman. J Neurol patients are that both were young women gait. She now walks with a frame indoors, but Neurosurg Psychiatry 1996;60:225–30. 2 Bhatia KP, Brown P, Gregor R, Lennox GG, with partial seizures, which were treated by has frequent falls. She uses a wheelchair Manji H, Thompson PD, et al. Progressive polytherapy with anticonvulsant drugs. A outdoors. She has never had a seizure. myoclonic ataxia associated with coeliac dis- recent double blind, multi-institution trial to She has never had any specific bowel symp- ease. Brain 1995;118:1087–93. 3 Lu C-S, Thompson PD, Quinn NP, Parkes JD, compare CBZ and VPA showed that changes toms, but investigation of a macrocytic anae- Marsden CD. Ramsay Hunt syndrome and in hair texture or hair loss were seen in 6% of mia and folate deficiency associated with coeliac disease: a new association? Mov Disord the patients treated by CBZ.3 As the authors weight loss eventually led to the diagnosis of 1986;1:209–19. did not specifically diVerentiate between coeliac disease five years ago, confirmed by 4 Hadjivassiliou M, Gibson A, Davies-Jones GAB, Lobo AJ, Stephenson TJ, Milford-Ward changes in hair texture and hair loss, it is small bowel biopsy. However, coeliac disease A. Does cryptic gluten sensitivity play a part in uncertain how often hair loss occurred in was almost certainly present 10 years before neurological illness? Lancet 1996;347:369–71. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from Letters, Correspondence, Correction, Book reviews 551

Successful outcome with aggressive lignocaine. Despite these measures the pres- sive therapy with a combination of corticos- treatment of acute haemorrhagic sure remained high with haemodynamic teroids, plasmapheresis, and cyclophos- leukoencephalitis instability. Consequently, a partial left frontal phamide. Immunosuppressive treatment for lobectomy was performed without replace- acute haemorrhagic leukoencephalitis is Acute haemorrhagic leukoencephalitis was ment of the bone flap on day 3. based on the assumption that it is an autoim- first described as a pathological entity by Because of his deterioration on broad mune disorder. The strongest evidence for Hurst in 1941.1 It is a rare acute inflammatory spectrum antibiotics and acyclovir along with this comes from its similarity to experimental myelinopathy of the CNS characterised by the antecedent flu-like illness, a presumptive allergic encephalomyelitis and in humans progressive loss of consciousness leading to diagnosis of acute haemorrhagic leukoen- with acute disseminated encephalomyelitis coma accompanied by focal neurological dys- cephalitis was made and 1000 mg methyl- there is evidence of activation of autoreactive function. It is now generally accepted that prednisolone daily for three days was given. T cells against myelin basic protein.78Of the acute haemorrhagic leukoencephalitis is the Subsequently, histology of the brain biopsy immunological therapies that have been severe end of the range of acute disseminated confirmed the clinical diagnosis (fig 2). reported to be eVective it is not clear which is encephalomyelitis.2 Immunoperoxidase for herpes simplex virus superior or which combination is best. None- A 34 year old man presented with rapidly was negative. Polymerase chain reaction theless, in a severely disabled patient it would progressive dysphasia, right hemiparesis, and amplification of herpes simplex type 1 DNA seem prudent to maximise treatment. deteriorating level of consciousness over 24 on brain tissue and in CSF was negative. R MARKUS hours after a flu-like illness in the preceding Intravenous immunoglobulin was given at B J BREW week. He had no history of any other relevant a dose of 30 g daily for five days initially but Department of Neurology the patient remained in coma. Plasmapher- previous illnesses, travel, or immunisation J TURNER esis (five 2 l exchanges on alternate days) and and was not taking any medication. Department of Pathology On initial examination he was febrile at 100 mg/day cyclophosphamide were com- M PELL 38°C. He had an expressive dysphasia, was menced on day 9. Methylprednisolone was given at a dose of 1000 mg daily for a further Department of Neurosurgery, St Vincent’s Hospital, drowsy, and had mild neck stiVness. A right Sydney, Australia central facial weakness was present and a gag five days followed by a reducing course of Correspondence to: Associate Professor BJ Brew, reflex was absent. He moved his left arm and steroid. First signs of recovery were noted on day 376 Victoria St Darlinghurst, Sydney, Australia leg to command but only flexed his right arm 2010. and leg to painful stimuli. Computerised 14 when he opened his eyes and moved his tomography of the brain showed left fron- left hand to command. He was extubated on toparietal oedema with mass eVect causing day 17. Steady neurological recovery contin- 1 Hurst AE. Acute haemorrhagic leucoencephali- eVacement of the sulci and the frontal horn of ued and he was walking with assistance and tis: a previously undefined entity. Med J Aust the lateral ventricle and 5 mm midline shift speaking by day 30. Cyclophosphamide and 1941;1:1–6. steroids were ceased on day 40 and 50 2 Tselis AC, Lisak RP. Acute disseminated (fig 1). encephalomyelitis and isolated central nervous There was further deterioration in his con- respectively. At four months a system demyelinative syndromes. Curr Opin sciousness over the next several hours leading was performed. At nine months he is back at Neurol 1995;8:227–9. work as an accountant and sailing competi- 3 Mastrodimos B, Barnett GH, Awad IA. Inten- to intubation and transfer to the intensive sive care management of acute hemorrhagic care unit with a presumptive diVerential tively with minimal neurological deficit. leukoencephalitis with favourable neurologic diagnosis of viral encephalitis, bacterial Aggressive treatment of patients with acute outcome. Clev Clin J Med 1992;59:549–52. cerebritis, or acute disseminated encephalo- haemorrhagic leukoencephalitis may lead to a 4 Kanter DS, Horensky D, Sperling RA, Mala- favourable outcome despite the presence of chowsky ME, Churchill WH. Plasmapheresis myelitis. Treatment with acyclovir, ceftriax- in fulminant acute disseminated encephalomy- one, flucloxacillin, phenytoin, and dexam- severe neurological disability at the beginning elitis. Neurology 1995;45:824–7. ethasone was commenced. Cerebrospinal of the illness.3–5 Recognition of acute haemor- 5 Seales D, Greer M. Acute hemorrhagic leukoen- fluid from ventricular puncture showed rhagic leukoencephalitis may be diYcult but cephalitis. A successful recovery. Arch Neurol 3 1991;48:1086–8. 25 000 red blood cells/mm , 53 white blood there are several useful clinical signals: (a) the 6 Rothstein TL, Shaw CM. Computerized tom- cells/mm3 (72% polymorphonuclear cells, antecedent flu-like illness, (b) early changes ography as a diagnostic aid in acute hemor- 24% lymphocytes, 4% monocytes), 1257 on brain CT,6 (c) the presence of red cells and rhagic leukoencephalitis. Ann Neurol 1983;13: a neutrophil pleocytosis in the CSF, and (d) 331–3. mg/l protein, and 6 mmol/l glucose. Gram 7 Lisak RP, Zweiman B. In vitro cell mediated stain, bacterial cultures, and cryptococcal deterioration despite broad spectrum antibi- immunity of cerebrospinal fluid lymphocytes antigen were negative. otics and antiviral therapy. Brain biopsy, to myelin basic protein in primary demyelinat- http://jnnp.bmj.com/ On day two of the hospital admission he however, is essential for diagnosis. ing diseases. N Engl J Med 1977;297:850–3. Management consists of early diagnosis, 8 Lisak RP, Behan PO, Zweiman B, Shetty T. Cell developed signs of increasing cerebral mediated immunity to myelin basic protein in oedema and his intracranial pressure was 90 aggressive control of raised intracranial pres- acute disseminated encephalomyelitis. Neurol- mm Hg (normal 15 mm Hg). Medical sure in intensive care, and immunosuppres- ogy 1974;24:560–4. therapy to lower intracranial pressure was started with hyperventilation, mannitol, and on September 26, 2021 by guest. Protected copyright.

Figure 1 Brain CT showing left hemispheric oedema with mass eVect and midline shift. There Figure 2 Luxol fast blue stain for myelin showing perivenous demyelination and punctate is no contrast enhancement to suggest an abscess. haemorrhages (original magnification ×10). J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from 552 Letters, Correspondence, Correction, Book reviews

Adult Niemann-Pick disease type C prominent features in adult onset Niemann- mimicking features of multiple sclerosis Pick disease type C. Vertical supranuclear ophthalmoplegia, which was absent in our Niemann-Pick disease type C is a panethnic patient, is an important sign in infantile and autosomal recessive neurovisceral storage juvenile Niemann-Pick disease type C, but it disorder characterised by a unique error in is often missing in adult disease1 cellular traYcking of exogenous cholesterol. Findings on MRI and in CSF had initially The most typical phenotype is characterised suggested a diagnosis of multiple sclerosis in by hepatosplenomegaly, vertical supranuclear our patient. Dementia may occasionally be ophthalmoplegia, progressive ataxia, dysto- the leading or even only symptom in multiple nia, and dementia and manifests in late child- sclerosis and features of dementia in multiple hood. Adult onset Niemann-Pick disease sclerosis resemble those in our patient. In type C often includes psychosis and Niemann-Pick disease type C, neuroimaging dementia.12 We report an adult female is usually normal or shows non-specific patient with biochemically verified Niemann- cerebral atrophy, but some cases with periv- Pick disease type C in whom the initial diag- entricular white matter lesions have been 3 nosis of multiple sclerosis was questionable described. Neuropathological studies occa- because of a persistent splenomegaly. sionally showed demyelination in infantile 24 The 33 year old woman was referred to the cases. Therefore, clinical and neuroradio- Psychiatry Department of the University of logical findings in our patient were compat- Heidelberg in December 1994. She com- ible with a diagnosis of Niemann-Pick disease plained of progressive diYculties with type C. However, to our knowledge oligo- memory and concentration, increasing social clonal bands have not been reported in the isolation, and problems with daily activities. disease. Increased CSF immunoglobulin Axial T2 weighted images show multifocal and The patient was unsure about the onset of her concentrations are often found in partly confluent subcortical areas with adrenoleukodystrophy,5 indicating that in- problems but suggested that they had started pathologically high signal. after the birth of her daughter three years born errors of metabolism can be associated previously. History, family history, and the with an immune response. Accordingly, patient’s fetal and childhood development oligoclonal bands in our patient may also enlarged (vertical diameter 17 cm) and inho- represent an immunological response to cer- were unremarkable. Her parents are non- mogeneously structured spleen. Liver size consanguineous. School performance was ebral damage caused by Niemann-Pick dis- was borderline. Routine hematological and ease type C. In our patient, Niemann-Pick satisfactory initially but later her learning biochemical analyses were normal except for disease type C mimicked features of multiple behaviour became erratic and she failed her transient thrombopenia. Vasculitis screening, sclerosis. Therefore, in patients with sus- examinations after 10 school years. She serum tests for copper, coeruloplasmin, vita- pected multiple sclerosis in whom dementia worked as a saleswoman and after giving birth min B12, folic acid, basic endocrinological or psychosis are the leading symptoms, she lived on social support. With time she no tests including thyroid function assessment, Niemann-Pick disease type C should be con- longer managed to educate her daughter and testing for HIV and syphilis, very long chain sidered as diVerential diagnosis. neglected her housekeeping. The patient´s fatty acids, arylsulphatase A, and other serum mother reported that her problems with lysosomal enzyme activities were normal. ARMIN J GRAU memory and concentration had been variably Considering the MRI and CSF findings we TOBIAS BRANDT apparent since school but progressively dete- first diagnosed multiple sclerosis with a Department of Neurology riorated after her pregnancy. prominent picture of progressive dementia. MATTHIAS WEISBROD On neurological examination, cranial Treatment with steroids did not improve the RAINER NIETHAMMER nerves were unremarkable. Guided and clinical status. The patient was referred to a Department of Psychiatry voluntary eye movements were normal; fast rehabilitation hospital. There, CSF analysis MICHAEL FORSTING saccadic movements were slightly dysmetric. was normal and oligoclonal bands were Department of Neuroradiology Optokinetic nystagmus was absent vertically absent. but normal horizontally. The patient had a In August 1995, the patient was readmitted MICHAEL CANTZ

Department of Pathochemistry University of http://jnnp.bmj.com/ slight of her left leg, tendon reflexes in a state of severe depression and helpless- Heidelberg, Germany were brisker in the left than the right leg but ness. Her dementia had deteriorated, symmetric in both arms. Plantar response neurological examination was unchanged, MARIE T VANIER was flexor. There was a slight ataxia in both still showing slight lower limb ataxia and left INSERM U 189 and Fondation Gillet-Mérieux, Lyon-Sud Medical School and Lyon-Sud Hospital, legs. Sensory testing was normal. Snout or leg paresis. Because control sonography France palmomental reflexes were absent. The pa- disclosed persistent splenomegaly, we consid- tient was inattentive but well oriented to time, ered the diagnosis of Niemann-Pick disease KLAUS HARZER person, and place. She had diYculties in type C or Gaucher disease. Cytology of a Institute for Brain Research, University of Tübingen, Germany retrieving past and storing new information. bone marrow aspirate disclosed so-called sea

There was no aphasia or dysarthria but her blue histiocytes and Niemann-Pick cells, a on September 26, 2021 by guest. Protected copyright. Correspondence to: Dr Armin J Grau, Neurology speech was vague and circuitous with many result strongly suggestive of Niemann-Pick Department, University of Heidelberg, Im Neuen- meaningless phrases. Thinking was slowed, disease type B or C. Biochemical studies on heimer Feld 400, 69120 Heidelberg, Germany. impoverished in content, and reduced in cultured skin fibroblasts confirmed the diag- flexibility. She was emotionally labile, her nosis of Niemann-Pick disease type C. The behaviour was disorganised, and her judg- intracellular low density lipoprotein induced 1 Shulman LM, Noble JD, Weiner WJ. Psychosis ment was impaired. esterification of cholesterol was reduced (330 as the initial manifestation of adult-onset Psychometric tests resulted in an estimated (controls 2950 (SD 1200)) pmol/mg cell Niemann-Pick disease type C. Neurology 1995;45:1739–43. IQ of 71 on the Hamburg-Wechsler test. An protein of cholesteryl oleate formed after 2 Pentchev PG, Vanier MT, Suzuki K, Patterson amnesia score (Berliner Amnesietest- 4.5 h). M. Niemann-Pick disease type C: A cellular Kurzform) and the d2 test were highly patho- Cytochemical staining with filipin showed cholesterol lipidosis. In: Scriver CR, Beaudet logical. Cranial MRI showed considerable AJ, Sly WS, Valle D, eds. The metabolic and a significant but not massive accumulation of molecular basis of inherited disease. New York: cerebral atrophy and multiple non-enhancing unesterified cholesterol in perinuclear vesi- McGraw Hill 1995:2625–39. hyperintense areas on T2 weighted images cles. These combined findings were consist- 3 Fink JK, Filling-Katz MR, Sokol J, et al. Clinical localised in the periventricular white matter ent with an “intermediate” biochemical phe- spectrum of Niemann-Pick disease type C. 2 Neurology 1989;39:1040–9. and the corpus callosum (figure). Follow up notype of Niemann-Pick disease type C. 4 Harzer K, Schlote W, PeiVer J, Benz HU, Anzil examination in spring 1996 showed no The clinical picture in our patient corre- AP. Neurovisceral lipidosis compatible with changes. Protein concentration (0.36 g/l) and sponds well to late onset Niemann-Pick Niemann-Pick disease type C: morphological and biochemical studies of a late infantile case cell count (1 cell/ml) in CSF were normal; disease type C. Although her disease may and enzyme and lipid assays in a prenatal case oligoclonal bands were positive in CSF but have insidiously started during childhood or of the same family. Acta Neuropathol (Berl) negative in serum. Somatosensory evoked adolescence, definite dementia only devel- 1978;43:97–104. potentials to both legs and transcortical mag- oped around the age of 30. Progressive 5GriYn DE, Moser HW, Mendoza Q, et al. Iden- tification of the inflammatory cells in the netic evoked potential to the left leg were dementia is the most often reported sign, and of patients with adreno- delayed. Abdominal sonography disclosed an limb ataxia and splenomegaly are other leukodystrophy. Ann Neurol 1985;18:660–4. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from Letters, Correspondence, Correction, Book reviews 553

Motor cortex stimulation does not reset In three patients, while standing, transcra- The frequency of tremor ranged between primary orthostatic tremor nial magnetic stimulation (TMS) was applied 14 and 18 Hz. Leg tremor was present in all using a Magstim 200 with a double cone coil patients on standing and disappeared on Primary orthostatic tremor is a rare but well positioned over the vertex. Stimulus intensity sitting, whereas arm tremor was only seen characterised condition.1–3 The predominant was suYcient to produce consistent motor when patients leant on their arms. Within feature is unsteadiness and shaking of the evoked potentials in voluntarily activated leg each patient there was a clear and consistent legs, which occurs exclusively on standing. muscles while seated. The magnetic stimuli phase relation of tremor bursts between Electromyographic recordings disclose a were triggered, at a maximum of 0.3 Hz, by diVerent muscles. However, comparing all 13-18 Hz tremor in leg and paraspinal EMG bursts. The stimulus delay after an five patients the exact timing of these muscles.23 A similar tremor in the arms can EMG burst was varied between 0 and 50 ms relations varied. Specifically, there was no be induced by weight bearing actions. using a precise timing device, to see whether overall relation between flexor and extensor We studied five patients with previously any eVect of TMS was dependent on its tem- muscles or between left and right sides. established clinical and EMG diagnoses of poral relation to the tremor cycle. Ten stimuli We found that changes in head position primary orthostatic tremor. Surface EMG were applied at each delay setting. while standing had no eVect on either the fre- recordings were made from up to four The EMG signals were digitised and data quency or amplitude of the tremor. This would muscles simultaneously; one muscle (right were rectified and displayed in sweep dura- suggest that static vestibular inputs to spinal hamstrings) was recorded from throughout tions of 200-500 ms. The 10 individual trials motor neurons do not modulate the tremor. each study to act as a timing reference. Leg for a given delay were averaged, using the In two patients TMS reduced the ampli- muscles were studied with the patients stand- stimulus as a reference point. In addition all tude of the next EMG burst after the evoked ing and arm muscles with patients leaning interburst intervals in each individual trial compound muscle action potential (CMAP), forwards on their arms. The eVect of different were measured. This allowed calculation of but did not alter the frequency of the tremor maintained head positions (looking to the any latency shift in the first poststimulus (figure, A). The magnitude of reduction in left, to the right, upwards, and downwards) tremor burst, as well as any subsequent amplitude was variable and did not correlate was also recorded. alteration in tremor frequency. with alterations in stimulus delay. Motor neurons discharged as a result of TMS of the motor cortex would be expected to remain partially refractory and this would explain the reduction in burst amplitude. In the third patient, EMG bursts were tran- siently suppressed for between one and three tremor cycles after TMS (figure, B), before reappearing at the expected latency and with- out any subsequent modulation of frequency. There was no correlation between degree of suppression and stimulus delay. This suppres- sion can probably be attributed to the “silent period”, which is thought to be due to activa- tion of both intracortical and spinal inhibitory mechanisms. Significantly, these inhibitory mechanisms did not seem to have modulated the tremor generator, as the EMG burst reap- peared at the expected time and the tremor continued at the prestimulus frequency. The generator site underlying primary orthostatic tremor is unknown. The fre- quency of the tremor (13-18 Hz) suggests that it is central rather than peripheral. The failure of peripheral stimuli to reset primary 2

orthostatic tremor supports this. The eVect http://jnnp.bmj.com/ of transcranial magnetic stimulation, which has been reported to reset both and parkinsonian tremor,4 has not previously been reported on in this condition. We suggest that the motor cortex is neither the site nor a modulator of the presumed central generator of primary orthostatic tremor. Furthermore, other intracortical structures activated by transcranial magnetic stimulation (for example, intracortical or on September 26, 2021 by guest. Protected copyright. transcallosal inhibitory pathways) do not seem to modulate the generator.

This research was funded in part by a Wellcome Trust research grant (KAN). KERRY R MILLS KANNAN A NITHI Clinical Neurophysiology Unit, University Department of Clinical Neurology, The RadcliVe Infirmary, Oxford OX2 6HE, UK Correspondence to: Dr KR Mills.

1 Heilman KM. Orthostatic tremor. Arch Neurol 1984;41:880–1. 2 Britton TC, Thompson PD, van der Kamp W, Rothwell JC, Day BL, Findley L, Marsden CD. Primary orthostatic tremor: further observa- tions in six cases. J Neurol 1992;239:209–17. All traces show rectified and averaged (10 trials) EMG data over a sweep duration of 500 ms. (A) 3 McManis PG, Sharbrough FW. Orthostatic and (B) refer to diVerent patients, recording from right quadriceps and right hamstrings respectively. tremor: clinical and electrophysiologic charac- teristics. Muscle Nerve 1993;16:1254–60. In both cases the upper trace is from a recording without transcranial magnetic stimulation. (A) The 4 Pascual-Leone A, Vallis SJ, Toro C, Wasser- stimulus was triggered 45 ms after the start of an EMG burst. The EMG burst after the CMAP is mann EM, Hallett M. Resetting of essential reduced in amplitude, but occurs at the expected latency (vertical line). (B) The stimulus was tremor and postural tremor in Parkinson’s dis- triggered at the start of an EMG burst. EMG bursts disappear after the CMAP,but then reappear at ease with transcranial magnetic stimulation. the expected latency (vertical line). Muscle Nerve 1994;17:800–7. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from 554 Letters, Correspondence, Correction, Book reviews

Frozen section in pituitary surgery manent section diagnosis. The accuracy rate KJ, eds. The pituitary. New York: Elsevier for rapid intraoperative diagnosis for the 1987:203–27. 5 Martinez AJ, Lee A, Moossy J, Maroon JC. Most surgeons consider intraoperative frozen transsphenoidal route was 92.9%. The accu- Pituitary adenomas: clinicopathological and section examination as an important part of racy rate in subfrontal surgery was 91.7%. immunohistochemical study. Ann Neurol 1980; pituitary surgery.12 Frozen section is used to The size of the intraoperative specimen 7:24–36. establish a histological diagnosis (adenoma, ranged from 0.4×0.4×0.4 cm to 1.2×1.0×0.6 6 McKeever PE, Laverson S, Oldfield EH, Smith BH, Godille D, Chandler WF. Stromal and other tumours, hyperplasia, and non- cm in subfrontal surgery, and from nuclear markers for rapid identification of neoplastic processes) and to examine the 0.1×0.1×0.1 cm to 1.5×0.5×0.1 cm in trans- pituitary adenomas at biopsy. Arch Pathol Lab margins of resection and determine its sphenoidal surgery. Nine trans-sphenoidal Med 1985;109:509–14. adequacy.12 A reliable intraoperative histo- samples (32.1%) were smaller than 7 Dalal BI, Malik AK, Datta BN. Frozen section × × diagnosis—a review of 1057 cases. Indian J logical diagnosis will help the surgeon with 0.4 0.4 0.4 cm which was the smallest Cancer 1979;16:59–65. achieving the ultimate goal in pituitary subfrontal sample size. However, both misdi- 8 Elsner B. La biopsia por conjalacion: su valor surgery—namely, selective removal of the agnosed trans-sphenoidal samples were assistencial y en la educacion medica del lesion with preservation of endocrine larger than 0.4×0.4×0.4 cm. All smaller sam- patologo. Prenatale Medica Argentina 1968;55: 3 1741–9. function. With the increased popularity of ples were accurately diagnosed. 9 Agnantis NY, Apostolikas N, Christodoulou I, the transsphenoidal approach for microsurgi- There are a few studies available in the lit- Petrakis C, Garas J. The reliability of frozen cal removal of pituitary lesions, mainly erature describing frozen section accuracy section diagnosis in various breast lesions: a microadenomas, the reliability of the frozen rates in pituitary surgery. Overall accuracy study based on 3451 biopsies. Recent Results Cancer Res 1984;90:205–10. section technique becomes even more impor- rates were found to be between 83.1 to 10 Fessia L, Ghiringhello B, Arisio R, Botta G, tant. It has been stated that the trans- 94.3%.1–3 6 Lang et al1 emphasised that these Aimone V.Accuracy of frozen section diagnosis sphenoidal route provides the surgeon with rates are consistently lower than accuracy in breast cancer detection. Pathol Res Pract suboptimal exposure of the gland and there- rates of frozen section in other types of 1984;179:61–6. 78 910 11 Farmer P, Sher JH, Cooper P. Frozen section fore smaller pieces of tissue are submitted for surgery including breast surgery. Farmer diagnosis in neuropathology: Usefulness and evaluation by frozen section.45Some authors et al11 reported 91.0% accuracy of frozen sec- accuracy. Acta Neuropathologica and Experimen- suspect that this may result in a lesser degree tion diagnosis in neuropathology. Lang et al1 tal Neurology 1981;40:349. of accuracy of the intraoperative rapid and other authors indicate that two main fac- diagnostic techniques.45 tors contribute to lower accuracy rates in Bilateral posteroventral for There are only a few papers in the English pituitary surgery: Firstly, the size of the literature studying the accuracy rate of frozen specimen. They state that extremely small severe antipsychotic induced tardive section diagnosis of pituitary lesions.1–3 6 specimens present technical diYculties which and dystonia None of these compared the accuracy rate in result in diYcult interpretation. Secondly, the trans-sphenoidal surgery with that of the diYcult task of histological diVerentiation Tardive dyskinesia as a consequence of long term treatment with antipsychotic drugs has subfrontal approach. between normal pituitary tissue, hyperplasia, 1 We studied 55 cases of pituitary surgery. and adenoma. Many authors indicate that an average prevalence of 15%–20% and may coexist with tardive dystonia which occurs in Twenty four operations were done before trans-sphenoidal pituitary surgery provides a 2 1991 via the subfrontal craniotomy approach. limited exposure and therefore results in 1%–4% of patients. The medical treatment Since August 1991 almost all pituitary smaller pieces submitted for intraoperative of both is notoriously diYcult and it is often surgery at our institution is done via the diagnosis.45Therefore, it has been suspected impossible to withdraw the provoking medi- trans-sphenoidal route. Between 1991 and that trans-sphenoidal surgery results in lower cations. We report a patient with early onset 1993 we performed 31 trans-sphenoidal frozen section accuracy rates.45 severe and refractory tardive dyskinesia and pituitary operations. The patients’ medical The results of our study indicate that the dystonia which was significantly improved by records, operative reports, and pathological surgical approach did not play a part in bilateral simultaneous posteroventral palli- reports were studied. The histological slides determining the accuracy rates. Accuracy dotomies. of the specimens submitted for rapid intraop- rates of frozen section diagnosis were essen- A 31 year old man had a seven year history erative diagnosis, as well as those submitted tially the same for both the trans-sphenoidal of a schizophrenic illness presenting with for permanent section, were reviewed. All and the subfrontal routes. Based on these auditory hallucinations, persecutory delu- slides of frozen sections, touch preparations, findings and review of the literature,125 we sions, delusions of reference, thought inser- and permanent sections, were stained with conclude that problems with the quality of the tion, withdrawal, and broadcasting. There was no history of family psychiatric or haematoxylin and eosin. Additional stains specimen, as well as problems with histologi- http://jnnp.bmj.com/ and immunohistochemical studies were per- cal characteristics of the frozen section, play a neurological disorder and no relevant per- formed on the permanent sections only. In more important part than the size of the sonal medical or psychiatric history. After addition, the size of the intraoperative specimen as reasons for inaccurate diagnosis. initially responding well to treatment with specimen as measured by the examining Touch imprint is a very helpful histological trifluoperazine he discontinued this after a pathologist was recorded for every case. technique for intraoperative diagnosis.2 It is a few months and relapsed requiring admission Three cases were excluded from the study. In very rapid procedure which provides a to hospital two years after his first illness. His two cases the diagnosis could not be made microscopical sample with less artifacts and illness again responded to oral antipsychotic intraoperatively and was deferred. Permanent with better cellular details. Its disadvantages drugs and then depot flupenthixol decanoate section diagnosis was pituitary adenoma. The include some loss of architectural details and (maximum dose of 40 mg monthly). One specimens submitted for frozen section dur- the inability to evaluate invasion and year later he was first noted to have dyskinetic on September 26, 2021 by guest. Protected copyright. ing a third case were considered non- margins.1 movements of his mouth and tongue and his diagnostic. The patient was eventually found GADY HAR-EL depot flupenthixol decanoate was reduced to have a pituitary abscess. CHANDRAKANT RAO and then discontinued but he quickly re- RICHARD M SWANSON Of the 28 cases of transsphenoidal surgery lapsed and has required treatment with AFAF F ABDU antipsychotic medication since. His tardive included in the study, the diagnosis given by THOMAS H MILHORAT frozen section agreed with the permanent sec- dyskinesia rapidly progressed to involve his Departments of Neurosurgery and Otolaryngology, and limbs and trunk with severe choreiform and tion diagnosis in 26 patients. None of the 24 the Division of Neuropathology, State University of cases of subfrontal craniotomy were excluded New York, Health Science Center at Brooklyn, New hemiballistic movements. Severe dystonia from the study. The frozen section diagnosis York, USA developed involving his neck, larynx, upper limb girdle, and chest. These aVected his agreed with the permanent section diagnosis Correspondence to: Dr Gady Har-El, SUNY- in 22 cases. The table illustrates the cases with Health Science Center at Brooklyn, Box 126, 450 breathing and swallowing resulting in recur- discrepancy between frozen section and per- Clarkson Avenue, Brooklyn, NY 11203. rent chest infections, weight loss, and anae- mia. Communication became increasingly diYcult because of dysarthria. He was Table 1 Inaccurate intraoperative diagnosis 1 Lang HD, Saeger W, Ludecke DK, Muller D. Rapid frozen section diagnosis of pituitary increasingly unable to walk or stand and he tumors. Endocr Pathol 1990;1:116–22. had frequent falls with repeated minor Intraoperative Final 2 Lloyd RV. Frozen sections in the diagnosis of injuries necessitating wearing a helmet. Wors- Approach diagnosis diagnosis pituitary lesions. In: Lloyd RV, ed. Surgical pathology of the . Philadelphia: ening diYculties in managing at home due to Trans-sphenoidal Non-diagnostic Adenoma WB Saunders, 1993;22–4. intense distress and damage to furniture and Trans-sphenoidal Non-diagnostic Adenoma 3 Adelman LS, Post KD. Intraoperative frozen sanitary fittings resulted in increasing stays in Transfrontal Adenoma Normal section technique for pituitary adenomas. Am J hospital. pituitary Surg Pathol 1979;3:173–5. Transfrontal Meningioma Adenoma 4 Kim JH, Robinson F, Manuelidis EE. Pathology Neurological examination was normal ex- of the pituitary adenomas. In: Goodrich I, Lee cept for the presence of the involuntary J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from Letters, Correspondence, Correction, Book reviews 555 movements. On the modified Obeso dyski- resulting in retrocollis. His speech was less aggressive drug treatment involving strategies nesia scale these were rated as grade 4 mixed dysarthric than preoperatively and he was described as helpful in case reports including chorea and dystonia for all activities,3 and on now able to sit, feed himself, and walk the combination of clozapine and clon- the Fahn-Marsden dystonia rating scale at although there had been occasional falls. This azepam recently reported to be helpful in 76/120 for movement and 22/30 for disabil- was rated as grade 2 on the Obeso scale and severe cases of tardive dystonia.7 ity.4 He had a mild normochromic, normo- 21/120 and 4/30 respectively on the Fahn- Posteroventral pallidotomy is eVective in cytic anaemia which was only partially Marsden dystonia movement and disability treating both the akinetic and hyperkinetic responsive to iron supplements but investiga- scales. He had gained weight and his anaemia motor symptoms in Parkinson’s disease.5 A tions were otherwise normal (brain MRI, had improved but he had continued to recent case report described successful use in EEG, full blood count, urea and electrolytes, experience psychotic symptoms consisting of tardive dyskinesia8 but it has not to our glucose, and calcium, thyroid and liver func- auditory hallucinations and delusions of knowledge been reported in drug induced tion tests, and serum copper and caeruloplas- reference although these cause only minor tardive dystonia. min concentrations ). disability. Medication had only been reduced The short term outcome in this case has Alternative antipsychotic drugs were tried slowly and consisted of benzhexol (12 been a significant amelioration of abnormal including sulpiride, remoxipride, risperidone mg/day), clozapine (400 mg/day), fluvoxam- movements and has arguably been life saving and, for the past two years, clozapine (up to ine (100 mg/day), diazepam (6 g/day), iron with pronounced improvement in function 550 mg/day). In addition the following treat- sulphate (600 mg/day), and salbutamol (16 and quality of life. We suggest that posterov- ments were tried singly and in combination: mg/day). He had had no periods in hospital entral pallidotomy should be considered for sodium valproate (3 g/day), diltiazem (360 after the procedure. extremely severe cases of tardive dyskinesia mg/day), clonazepam (6 mg/day), tetrabena- This patient presented with tardive dyski- and dystonia but further experience is zine (175 mg/day), vitamin E (1600 IU/day), nesia of the “buccal-linguomasticatory” type required to determine its place in the wider procyclidine (80 mg/day), benzhexol (16 but rapidly progressed to severe tardive treatment of these disorders. mg/day) and diazepam (30 mg/day). Salbuta- dystonia. The picture was consistent with J WEETMAN mol (16 mg/day), given for a chest infection suggested characteristic features of patients I M ANDERSON seemed to bring modest relief. None of the with tardive dystonia; male sex, early age of University of Manchester Department of Psychiatry, treatments brought about any major or onset after a fairly short duration of antipsy- Rawnsley Building, Manchester Royal Infirmary, lasting improvement except that high doses of chotic treatment, and a high degree of Oxford Road, Manchester M13 9WL, UK 6 benzodiazepines caused sedation and relief disability. There is no recognised specific R P GREGORY from the movements when he was asleep. He treatment for tardive movement disorders but Department of Neurology, Battle Hospital, Oxford had input from a dietician, speech therapist, the patient described here had received Road, Reading, Berks RG3 1AG, UK and psychologist without success. After reports of the successful treatment of dopa induced in Parkinson’s dis- ease by pallidotomy,5 he was referred to the Department of Neurosurgery, Frenchay Hos- pital, Bristol where he underwent bilateral posteroventral pallidotomy. The day preop- eratively he underwent brain MRI on a Phil- lips Gyroscan 1.5 Tesla NT. The detailed anatomy of the was visualised using inversion recovery sequences with long acquisition times to provide high resolution images. The relation between the medial glo- bus pallidus and the anterior and posterior commissures (AC/PC) of the third ventricle were calculated, and the predicted lesion vol- ume defined for each side. A Leksell G frame was secured to the head under general anaes- thesia and the patient was transferred to a CT http://jnnp.bmj.com/ scanner where the coordinates of the AC/PC were established. From these, the target coordinates were transposed to the stereogu- ide. Through bilateral frontal burr holes a radiofrequency electrode was guided to the target, sequentially on each side. To ensure accurate placement of the proposed lesions, anaesthesia was lightened, so that the patient became partially responsive. Stimulation on September 26, 2021 by guest. Protected copyright. studies confirmed avoidance of the internal capsule and optic tract. Because of the sever- ity of the involuntary movements it was not possible to use high frequency stimulation to mimic the lesion. Thermal lesions were then made each with a predicted length of 6 mm and diameter of 4 mm. Postoperative MRI (figure) confirmed accu- rate placement of the lesions in both posterov- entral pallidums. In the immediate postopera- tive period he was completely free of abnormal movements but his swallowing deteriorated and he developed an aspiration pneumonia. This was successfully treated with a short period of ventilation and antibiotics. One week postoperatively there was a slight return of the orofacial dyskinesia but he was discharged home at 10 days with a sustained improvement in his . At reassessment eight months postopera- tively he still had mild to moderate dyskinesia of his mouth and neck with some laryngeal Figure Postoperative MRI confirming accurate placement of stereotaxic lesions in both and moderate dystonic movements posteroventral pallidums. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from 556 Letters, Correspondence, Correction, Book reviews

SSGILL due to local pain at the time of treatment. for it to be applied in a consistent manner. Department of Neurosurgery, Frenchay Hospital, Other side eVects (for example, small local For that reason we have used this approach as Bristol BS16 ILE, UK haematomas) are a minor concern. a back up for medication failure or Correspondence to: Dr Ian Anderson, University of I suspect that, after the pioneering treat- medication ineVectiveness and not as a main- Manchester Department of Psychiatry, Rawnsley ments, injection sites have been gradually stay in treatment. Building, Manchester Royal Infirmary, Oxford moved to preseptal and even to orbital ROBERT IANSEK Road, Manchester M13 9WL, UK. Telephone 0161 locations to avoid local pain. This must be 276 5396; fax 0161 273 2135; email: Geriatric Research Unit and Geriatric Neurology [email protected] taken into account considering that many Service patients still do well with preseptal botulinum MEG MORRIS toxin injections. 1 Gerlach J, Casey DE. Tardive dyskinesia. Acta Geriatric Research Unit, Kingston Centre, Psychiatr Scand 1988:369–78. ALBERTO ALBANESE Cheltenham, Victoria, 2 Raja M. Tardive dystonia. Prevalence, risk Australia 3192 factors, and comparison with tardive dyskinesia in a population of 200 acute psychiatric inpatients. Rehabilitation of gait in Parkinson’s Eur Arch Psychiatry Clin Neurosci 1 McIntosh GC, Brown SH, Rice RR, Thaut MH. 1995;245:145–51. disease 3 Goetz CG, Stebbins GT, Shale HM, Lang AE, Rhythmic auditory-motor facilitation of gait patterns in patients with Parkinson’s disease. Chernick DA, Chmura TA, et al. Utility of an J objective dyskinesia rating scale for Parkinson’s We comment on the recent publication by Neurol Neurosurg Psychiatry 1997;62:22–6. disease: inter- and intrarater reliability assess- McIntosh et al1 concerning rhythmic auditory 2 Morris ME, Iansek, R. Matyas TA, Summers JJ. The pathogenesis of gait in Parkin- ment. Mov Disord 1994;9:390–4. motor facilitation of gait patterns in patients 4 Burke RE, Fahn S, Marsden CD, Bressman SB, son’s disease. Brain 1994;117:1161–82. Moskowitz C, Friedman J. Validity and reliabil- with Parkinson’s disease. These authors have ity of a rating scale for the primary torsion dys- shown a roughly 10% improvement in gait tonias. Neurology 1985;35:73–7. velocity as a result of entrainment of cadence Thaut replies: 5 Gregory RP, Lambert MV, Rogers DGC, with a metronome embedded in background Ferguson IT, Gill SS. Medial pallidotomy for the treatment of Parkinson’s disease: a con- music. We entirely concur with Iansek and Morris in secutive series of 26 patients. J Neurol Neuro- Although this study is interesting, the level regard to the importance of stride length surg Psychiatry 1997;62:207. of improvement in gait is therapeutically regulation in gait rehabilitation of patients 6 Burke RE, Fahn S, Jankovic J, Marsden CD, Lang AE, Gollomp S, Ilson J. Tardive dystonia: small and not directed to the amelioration of with Parkinson’s disease. In regard to their late onset and persistent dystonia caused by the basic deficit in parkinsonian gait. Other comments about our paper some apparent antipsychotic drugs. Neurology 1982;32:1335– studies have clearly shown that the basic defi- errors in reading our data may have occurred: 46. 7 Shapleske J, McKay AP, McKenna PJ. Success- cit in the control of parkinsonian gait is regu- (1) Iansek and Morris state that our study ful treatment of tardive dystonia with clozapine lation of stride length. Cadence control is found a 10% increase in velocity. Most of and clonazepam. Br J Psychiatry 1996;168: intact in Parkinson’s disease (Morris et al 2) their argument against the rehabilitation 516–8. and cadence is used as a compensatory relevance of our paper is based on this 8 Wang Y, Turnbull I, Calne S, Stoessi AJ, Calne DB. Pallidotomy for tardive dyskinesia. Lancet mechanism for the defective control of stride increase. However, the actual increase in our 1997;349:777–8. length. This is done to increase gait velocity paper (see p 24) was 36% for ON patients above normal and preset values for each indi- and 25% for OFF patients. This is potentially vidual patient. We see very little benefit in a very useful increase for rehabilitation.1 entraining patients whose cadence is already (2) The second argument against the increased above normal values to improve therapeutic usefulness of our paper concerns CORRESPONDENCE velocity by values of only the order of 10%. the improvement of cadence rates which are This is particularly so when it is possible to already raised above normal. Again, this restore normal gait velocity, normal cadence assertion does not fit the cadence rates in our values, and normal stride length by use of study sample. The cadence for ON patients Pretarsal injections of botulinum toxin either visual cues and more importantly by was 98 and 91 for OFF patients, which is improve blephospasm in previously attentional strategies that are not dependent clearly below normal age matched cadence unresponsive patients on any visual cues whatsoever. Under these rates and matches gait data from other stud- circumstances it is possible to restore normal ies . The experience of Albanese 1 is very gait indices in patients with Parkinson’s (3) The emphasis of Iansek and Morris on

et al http://jnnp.bmj.com/ interesting because it is a good example of the disease even though they may be totally oV the importance of stride length is well taken. old adage that logical treatment works. The their medication or when medication is However, our study supports that point. technique as taught by Scott does involve the totally ineVective. Rhythmic cuing improved both stride length injection of pretarsal botulinum toxin be- The percentage improvement in gait veloc- and cadence in our study (p 24, table 2, and cause it is spasm of the pretarsal muscle that ity associated with auditory stimulation is of results section). This dual eVect of rhythmic causes the symptoms. the same order that can be attained by cuing on stride length and cadence in Parkin- Injection of the preseptal muscle should patients by simply asking them to walk faster. son’s disease has also been reported by other not be expected to be as eVective for control- When asked to do so, patients with Parkin- researchers.2 ling the symptom of lid spasm. Injection of son’s disease increase their cadence rate, and (4) The error of Iansek and Morris in the superciliaris is sometimes required to to a lesser degree stride length, by similar commenting on a 10% increase in velocity on September 26, 2021 by guest. Protected copyright. control the brow spasm that many of these amounts to improve their velocity. This may have stemmed from mistaking it for the patients get. requires no complex equipment—only a gen- 10% increase in cadence found in our study. A similar study to that of Albanese et al 1 tle prompt to walk at a faster rate. This, however, simply reflects the exact has been reported locally with a similar However, to obtain normal indices patients synchronisation to the rhythmic cue which result—the logical treatment of the pretarsal need to be shown the correct stride length was set only 10% faster than each patient’s muscle is more eVective (but produces that they are to attain and then told to use baseline to avoid hastening into abnormally slightly more side eVects) than the less logical attentional strategies to attain that stride high step rates. Patients were able to follow injection of the preseptal orbicularis. length. Again with very minimal equipment the rhythmic cue in close synchronisation and with a modicum of encouragement it is without compromising their stride length. LIONEL KOWAL possible to allow patients with festinant gait (5) From a clinical perspective it needs to The Private Eye Clinic, The Royal Victorian Eye and Ear Hospital, Locked Bag 8, East Melbourne 8002, to walk normally. be clearly reiterated that nowhere in this Australia We do not disagree with what the authors study do we suggest that patients with step have demonstrated in their paper, nor do we rates already above normal should be cued to 1 Albanese A, Bentivoglio AR, Galardi G, Mad- necessarily disagree with their conclusions; even higher step frequencies. The assertion of erna L, Colosimo C, Tonali P. Pretarsal however, we merely wish to put these findings Iansek and Morris in this respect is un- injections of botulinum toxin improve blephos- into some form of perspective and of founded. However, patients with Parkinson’s pasm in previously unresponsive patients. J Neurol Neurosurg Psychiatry 1996;60:693-4. relevance, particularly in the sphere of disease with slowed step rates may very well rehabilitation. We have now used stride benefit from rhythmic cuing to improve Albanese replies: length normalisation as a strategy for restor- velocity, or the entrainment eVect may ing gait to normal values in our multidiscipli- actually be used to decelerate hastened stride. I agree with the propositions of Kowel. In my nary programme and have found this very Besides the apparent errors in reading our experience, patients sometimes question the eVective. The problem is that it requires con- results, we fail to see how this is less relevant need for pretarsal botulinum toxin injections stant vigilance and continual reinforcement than or contradicts other facilitation J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from Letters, Correspondence, Correction, Book reviews 557 techniques—for example, visual cuing or 2 Woolley I, Faragher M, Spelman D. Essential Neurosurgery. Second Neurological manifestations of HIV in an out- attentional training. Edition. Edited by ANDREW H KAYE. (Pp 448 patient clinic including a case control study of MICHAEL H THAUT factors associated with peripheral neuropathy. £23.50). Published by Churchill Annual Conference Australasian Society for HIV Livingstone, Edinburgh 1996. ISBN Medicine 1994;6:271(poster 32). 0443053472. 1 Thaut MH, McIntosh GC, Rice RR, et al. 3 Woolley I, Faragher MW, Ugoni A, Spelman D. Rhythmic auditory stimulation in gait training An analysis of factors associated with HIV- for Parkinson’s disease patients. Mov Disord related peripheral neuropathy. Neurological In- The second edition of this basic textbook 1996;11:193-200. fections and Epidemiology 1997;2 (in press). gives a clear and concise introduction to neu- 2 Richards Cl, Malouin F, Bedard PJ, Cioni M. 4 Haug CJ, Aukrust P, Egil L, et al. Disseminated rosurgery. It is mainly written by one author Changes induced by L-dopa and sensory cues mycobacterium avium complex infections in with contributions from his neurological and on the gait of parkinsonian patients. In: Wolla- AIDS: immunopathogenic significance of an cott M, Horak F, eds. Posture and gait: control neurosurgical colleagues at the Royal Mel- activated tumor necrosis factor system and mechanisms. Vol II. Eugene, OR: University of bourne Hospital. The text is well written, Oregon books, 1992:126-9. depressed serum levels of 1,25 dihydroxyvita- min D. J Infect Dis 1996;173:259–62. beginning with an excellent chapter on 5 Sathe SS, Tsigler D, Saral A, et al. “Pentoxyfil- neurological assessment and examination, Association between HIV distal symmetric line impairs macrophage defense against myco- and proceeding to covering the whole of polyneuropathy and bacterium avium complex”. J Infect Dis 1995; neurosurgery in 21 chapters. There are a Mycobacterium 172:863–6. avium complex infection large number of illustrations with good qual- ity reproduction of imaging. Neurosurgical We read with interest the paper by Norton et controversies are well argued with balanced al1 showing an association between distal approaches to diYcult areas—for example, symmetric polyneuropathy and Mycobacte- the treatment of spontaneous intracerebral complex (MAC) infection but not rium avium CORRECTION haematomas. Advances in some treatments between distal symmetric polyneuropathy such as gene and immunotherapy are well and cytomegalovirus (CMV) infection. This covered, but the sections on other new confirms our findings presented in Sydney in techniques—for example, endoscopy, image- 1994,2 to be published later this year.3 K Beyer, J I Lao-Villadóniga, B Vecino- guidance and interventional radiology are In our study a cohort of 94 consecutive Bilbao, R Cacabelos, R de la Fuente- brief. patients with HIV were examined for evi- Fernández. A novel point mutation in the I have no major criticisms, but the chapter dence of peripheral neuropathy. Thirteen had GTP cyclohydrolase I gene in a Spanish fam- on head injury could have had more detail on distal symmetric polyneuropathy. We then ily with hereditary progressive and dopa head injury as a component of multiple head performed a multivariate analysis of possible responsive dystonia. J Neurol Neurosurg Psy- injury (there is no mention of advanced associated clinical factors including CD4 chiatry 1997;62:420. In the third and fifth lines of the third trauma life support) and the section on count, age, means of HIV acquisition, ddI paragraph C→A should read A→C. minor head injury does not mention alcohol (didanosine) and ddC (zalcitabine) treat- or drugs as indication for admission. Also, the ment, opportunistic infections including book would benefit from some more recent CMV, pneumonia, and Pneumocystis carinii references. MAC and candidal infections. No factor The review of the first edition of the book reached significance but a history of MAC published in this in 1991 suggested infection had borderline significance BOOK REVIEWS Journal that the author preferred intracranial work as (P=0.051). he expressed less strong opinions regarding Surprisingly, homosexual transmission of spinal surgery. However, I think the second HIV was the next most significant factor edition has redressed the balance. The author (P=0.089). Other factors including CMV Concise Text of Neuroscience. Edited by now includes anterior approaches to the tho- and other opportunistic infections were not ROBERT E KINGSLEY. (Pp 564 £29.95). racic and lumbar spine. The section on AIDS significant. Published by Williams and Wilkins, is still brief, and the driving regulations relate Distal symmetric polyneuropathy probably Baltimore 1996. ISBN 0-683-04621-7. to Australian law. Overall, however, although has heterogeneous causes, sometimes drug the author admits the book describes practice related and sometimes possibly multifacto- In his introduction to this book the author in the Asia-Pacific rim region, it is equally rial. We doubt that MAC itself is the cause of states that the primary purpose is to provide http://jnnp.bmj.com/ applicable to European neurosurgery. In distal symmetric polyneuropathy given most an integrated approach for medical students summary, I enjoyed reading the book and of our patients (seven of 13) did not have of neuroanatomy, neurophysiology and neu- considered that it gives an accurate and use- overt MAC infections as was the case with rology. This is clearly an ambitious aim which ful guide to neurosurgical management. I five of 17 definite cases of distal symmetric by and large he has succeeded in achieving. would strongly recommend it to medical stu- polyneuropathy in the study by Norton . et al The basis of the success of the book is that it dents, general and orthopaedic surgeons, and We also considered the possible role of isoni- does not attempt to be comprehensive and is junior neurosurgical trainees. azid, although only three of our patients had written with a practical clarity. In addition the taken isoniazid in the past, and concluded chapters are sensibly structured and ex- PETER HUTCHINSON that cytokines such as tumour necrosis factor tremely well illustrated. on September 26, 2021 by guest. Protected copyright. and interleukin-6, known to be raised in Unfortunately, the emphasis on neuro- MAC infections,45may be involved. London Dysmorphology Database- anatomy and physiology neglects neuropa- We agree with Norton et al that further London Neurogenetics Database- thology. This is surprising as neuropathology investigation of (1) the mechanisms by which Dysmorphology Photolibrary, CD-ROM MAC may help produce peripheral neu- can provide a bridge between the primary (price for the two databases and the neurosciences and clinical neurology. Fur- ropathy and (2) whether treatment and photolibrary on CD-ROM £1292.50 thermore, the neurology sections at the end of prophylaxis of MAC will alter the natural his- including VAT). Oxford: Oxford University tory of this condition will be of future interest. certain chapters could benefit from limited Press. IAN WOOLLEY expansion. For example, the chapter on MARK FARAGHER cerebral cortex would be enhanced by a more This is a remarkable and ongoing . DENIS SPELMAN detailed approach to the dementias which by The Oxford Medical Database consists of the Departments of Infectious Diseases and Neurology, their nature highlight the association between London Dysmorphology Database (LDDB) and Alfred Hospital, Commercial Road, Prahran, 3081, anatomy, function and pathology. Victoria, Australia the London Neurogenetics Database (LNDB). In a sense this book is ahead of its time as The Dysmorphology Database has information Correspondence to: Dr I Woolley, Division of Infec- its unitary approach to neuroscience is in on over 2800 non-chromosomal multiple tious Diseases, University Hospital of Cleveland contrast to the discrete and modular teaching congenital anomaly syndromes and includes Case Western Reserve University, Euclid Avenue, of neurophysiology and anatomy. None the over 25 000 references. The London Neuroge- Cleveland, Ohio 44106, USA. less I found the book surprisingly readable netics Database contains details of over 2600 and I would expect it to be well received by syndromes aVecting the central and periph- 1 Norton GR, Sweeney J, Marriott D, Law MG, both medical students and junior doctors eral nervous system in adults and children. Brew BJ. Association between HIV distal sym- On a separate disc is a library of clinical pho- metric polyneuropathy and mycobacterium with a developing interest in neurology. avium complex infection. J Neurol Neurosurg tographs and also x ray films illustrating many Psychiatry 1996;61:606–9. SIDOHARTHAN CHANDRAN of the conditions contained on the database. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.63.4.547 on 1 October 1997. Downloaded from 558 Letters, Correspondence, Correction, Book reviews

How does the system work in practice? For absence of discussion of the International The authors rightly emphasise the need for each programme, an alphabetical list of con- Stroke Trial (IST), a large randomised trial good clinicopathological correlation and the ditions is presented. Various options can then comparing the use of aspirin and heparin in use of a broad range of histological tech- be entered. Syndrome details provide a stroke, which recently reported its prelimi- niques, both of which are particularly essen- resumé of the condition, accompanied in nary results. Not all clinicians will agree with tial in the proper evaluation of muscle some instances by illustrations which in the the book’s failure to endorse the use of MRA biopsies. However, it must be questioned why stand alone CD-ROM are of small size and of in selected patients. The discussion on the these clinicians find it necessary to alienate future of genetics in cerebral ischaemia is no particular value. If, however, the photoli- the very pathologists with whom they wish to brief and somewhat out of keeping with the brary is used alongside the data base then cooperate and to enlighten. Pathologists who pharmaceutical industry’s enormous spend- each photograph can be clicked on to present read the preface will discover that they above a full size demonstration of the clinical ing on trying to identify those with genetic all are considered woefully ignorant and the syndrome or the relevant imaging. These predisposition to vascular diseases. At one “traditional histopathological reliance on one images are generally of high quality. They point we wondered whether “mo” was a new or two poorly orientated and often badly fixed have been acquired from various sources gene till we realised it was an abbreviation for H and E stained sections” is hardly a fair including the author’s own clinical atlas pub- “month”! Notwithstanding, these are minor lished by WolV. Understandably, some of the irritations which we hope the authors will description of the modern diagnostic spe- syndromes are not illustrated and no doubt forgive us for mentioning; we would be failing cialty. these areas of deficit will gradually shrink as in our duty if we did not cite them. The edi- This is a small slim volume that will serve time passes and more illustrations become tors should be congratulated for producing as a useful overview and introduction to available. For each syndrome an abstract is an outstandingly well referenced treatise on muscle pathology; nevertheless there is con- what is likely to become an increasing large provided along with a features section which siderable diVerence between a diagnostic subspecialty. Reviewers should recommend a details the main manifestations of a particular manual and a reference atlas. The text of this book if they themselves would have it on their syndrome, and, finally, selected references for book is superficial and, given that muscle shelves— we do and we would. But would we that particular disorder. At any time all the pathology is never a simple matter of picture pay the price asked for by the publishers? syndromes and indeed all the references on matching, those actively engaged in biopsy the individual databases can be displayed. PANKAJ SHARMA diagnosis will find it very limited as a MORRIS BROWN This is a remarkable enterprise which will diagnostic bench manual. However, other prove invaluable for those clinicians and larger more comprehensive texts are available geneticists who deal regularly with the disor- Color Atlas of Muscle Pathology. Edited and I did buy this book and wish to ders under review. Its value, of course, stands by W J KENNETH CUMMING, JOHN FULTHORPE, recommend it, but mostly for the high quality or falls by the frequency with which the PETER HUDGSON, AND MICHAEL MAHON. (Pp images that are currently unsurpassed and material is revised. It is the intention of the 202; price £95.00 hard back). London: will provide a valuable source of reference. publishers to achieve this on a regular basis Mosby Wolfe, 1994. ISBN 0-7234-2016-5. and, providing that is fulfilled, I can recom- JANICE ANDERSON mend the database unreservedly. The authors’ stated aim with this atlas is to G D PERKIN produce a bench book of practical application Practical Neurology. By JOSE BILLER (Pp752; for clinicians and pathologists. Chapters on price £46.00). Phildelphia: Lippencott- technical aspects and general approaches to Raven,1997. ISBN 0-316-09483-8. The Clinical Pharmacology of Cerebral the muscle biopsy are followed by chapters Edited by GERT J TER HORST AND covering groups of muscle disorders in a sys- Ischemia. Recognising the impossibility of compressing JAKOB KORF. (Pp 305; price $99.50). New tematic fashion. Several useful appendices are included. the whole of neurology into a single volume of Jersey: Humana Press, 1996. digestible size, Jose Biller has instead pro- ISBNO-89603-378-3. The overall presentation of the book is attractive and well set out. As befits an atlas, duced a manual designed to enable relevant the illustrations, both colour photographs of and reliable information to be easily located. Stroke is well known to be the third largest histological sections and many electron mi- To this end Practical Neurology has been writ- cause of death in the western World but its crographs, are mostly of high quality and ten with common problems in mind, each treatment is largely conservative. The recog- convincingly demonstrate the features which chapter in the first half of the book dealing http://jnnp.bmj.com/ nition that the area surrounding an infarct the authors intend. with a particular neurological complaint. The (the penumbra) is still viable during the early The chapter on preparation of the muscle information is provided in an outline format stages of ischaemia along with our increasing biopsy, including both open and closed which is concise but of necessity brief, some- understanding of the pathophysiology of cer- approaches to the biopsy procedure itself, is times resulting in accounts that lack helpful ebal ischaemia has led to the establishment of practical and succinct. The authors cover the details. The format also results in an inevita- a whole host of clinical trials using various preparation of tissue for histochemistry, bly degree of repetition between the chapters “clot-busters” as well as agents designed for immunohistochemistry, electron microscopy, together with a rather flat and dry style. The neuroprotection—for example, calcium an- and morphometry. The appearances of both book is clearly written with the North Ameri- tagonists, excitatory amino acid antagonists, normal muscle and the pathological reactions can “generalist” in mind; some of the practi- on September 26, 2021 by guest. Protected copyright. free radical scavengers etc. Physicians and demonstrable by these various techniques are cal advice, such as remembering to ask the scientists interested in stroke found them- well illustrated. The potential pitfalls of tech- families of dementing patients whether there selves at first awkward, but now increasingly nical artefacts are included and there is a are any weapons at home, is of limited comfortable, bedfellows. This timely book— useful practical scheme of systematic analy- a collection of reviews— seeks to cover the sis. The following chapters only briefly relevance on this side of the Atlantic. newest of the subspecialties drawing on the describe broad categories of disease. The The second half of the book deals with expertise of around 30 investigators from adjuvant role of immunohistochemistry in therapy and is in general more detailed. three continents. The editors of the book the inflammatory myopathies and the im- However by separating the book in this way have undertaken a brave task. Almost by defi- portance of using these techniques in the the eYciency with which information can be nition their subject is moving quickly. Books, dystrophies is emphasised. However, cover- found is reduced. no matter how well intentioned or re- age of dystrophies other than Duchenne and In conclusion I think this book largely suc- searched, are by virtue of their medium slow Becker dystrophy is cursory and the book ceeds in its primary aim; although Alice may creatures; there are few references later than does not include the emerging classification have found little use for a book “without pic- 1995. The book fails to cite much of the of the many limb girdle syndromes. A short tures or conversation” I suspect that North recent literature which has moved the argu- chapter on miscellaneous disorders glosses American generalists will not be so disap- ment of stroke units needing “further study” over several of the most diYcult and diagnos- pointed. to requiring them in every hospital. The lead tically challenging categories in muscle dis- time of publication is also responsible for the ease. STEVE SAWCER