J Neurol Neurosurg Psychiatry 2001;71:129–137 129

LETTERS TO THE EDITOR

Accessory middle cerebral artery and moyamoya disease

A rare association of moyamoya disease with the accessory middle cerebral artery was seen in two patients. The terminal portions of bilateral internal carotid arteries and their vicinities were markedly stenotic and so- called moyamoya vessels developed at the base of the brain. The left accessory middle cerebral artery originating near the anterior communicating artery supplied the left ante- rior frontal lobe in both patients. Although the accessory middle cerebral artery coursed Figure 2 (A) Right carotid angiography (anterior-posterior view) shows stenotic change at the in the vicinity of the markedly stenotic termi- origin of the right middle cerebral artery, but the right anterior cerebral artery is normal. The left nal portion of the left internal carotid artery, accessory middle cerebral artery (arrows) originates near the anterior communicating artery coursing the artery was not stenotic. This finding parallel to the left middle cerebral artery. (B) Left carotid angiography (anterior-posterior view) implies that the steno-occlusive changes in shows severe steno-occlusive changes at the terminal portion of the internal carotid artery with the cerebral vasculature in moyamoya disease development of moyamoya vessels. have topological predilection to the distal internal carotid arteries. and its incidence has been reported to be preted as normal, but MR angiography Moyamoya disease is characterised by 0.3–4.0%.2–4 The accessory middle cerebral showed steno-occlusive changes of the termi- angiographic features of steno-occlusive artery originates from either the proximal or nal portions of bilateral intracranial internal changes of the terminal portions of bilateral distal horizontal portion of the anterior carotid arteries. Stenotic changes were more intracranial internal carotid arteries as well as cerebral artery coursing parallel to the severe on the left than on the right side. dilated perforating arteries at the base of the horizontal portion of the middle cerebral Although the proximal portion of the left brain known as “moyamoya” vessels. The artery and reaches the anterior frontal lobe.5 middle cerebral artery was markedly stenotic clinical manifestation of moyamoya disease is Patient 1, a 32 year old woman, had on MR angiography, the left accessory typically brain ischaemia in the paediatric presented with a transient ischaemic attack of middle cerebral artery was clearly shown to population and brain haemorrhage in adults.1 right hemiparesis 2 years before the current be without stenosis (fig 1 A). The patient was The accessory middle cerebral artery is a episode. Results of initial MRI imaging treated conservatively. variation of middle cerebral artery branching performed at another hospital were inter- No recurrent ischaemic attack had oc- curred over a period of 2 years until the patient began to experience transient ischae- mic attacks of right hemiparesis several times a month. She was referred to us for further evaluation. She was neurologically normal and history was not contributory except for mild hypertension for 2 years. Digital sub- traction angiography showed progressive steno-occlusive changes of the terminal portions of the internal carotid arteries as well as development of moyamoya vessels, which were consistent with the diagnosis of moyamoya disease (fig 1 B). Most moyamoya vessels on the left side originated from the accessory middle cerebral artery. This patient subsequently underwent bypass surgery bilat- erally. Frequency of transient ischaemic attacks reduced markedly during the follow up period of 3 months postoperatively. Patient 2 was a 30 year old man admitted for re-evaluation of moyamoya disease. This patient experienced occasional headache and vomiting at the age of 5 years. At the age of 11 years, transient ischaemic attacks of right hemiparesis developed to a rate of one a week. The diagnosis of moyamoya disease was established by . He underwent bilateral bypass surgery subse- quently at the age of 13 years in another hos- pital. He had experienced no ischaemic Figure 1 (A) Initial magnetic resonance episodes for 17 years thereafter and he angiography showing the left accessory middle thought that moyamoya disease was cured. cerebral artery (arrows) and steno-occlusive He came to our hospital when he had minor changes of the terminal portions of bilateral head trauma at the age of 30 years and was internal carotid arteries. (B) Left carotid advised to re-evaluate the disease. angiography (anterior-posterior view) At admission, the patient was neurologi- performed 2 years later, showing progressive cally normal. Brain MRI showed no paren- steno-occlusive vascular changes as well as development of moyamoya vessels (arrow) chymal abnormality. Single photon emission mainly from the left accessory middle cerebral computed tomography was normal. Right artery (arrowheads). carotid angiography showed severe stenotic

www.jnnp.com 130 J Neurol Neurosurg Psychiatry 2001;71:129–137

change of the proximal right middle cerebral 6 Stabler J. Two cases of accessory middle 35 artery, but the right anterior cerebral artery cerebral artery, including one with an aneu- PD patients rysm at its origin. Br J Radiol 1970;43:314–18. 30 Controls was normal. The left accessory middle 7 Komiyama M, Nakajima H, Nishikawa M, et al. cerebral artery originated near the anterior High incidence of persistent primitive arteries 25 communicating artery (fig 2 A). Left carotid in moyamoya and quasi-moyamoya diseases. 20 angiography showed severe stenosis at the Neurol Med Chir (Tokyo) 1999;39:416–22. 8 Handa J, Shimizu Y, Matsuda M, et al. The % terminal portion of the internal carotid artery accessory middle cerebral artery: report of fur- 15 with moderate development of moyamoya ther two cases. 1970; :415–16. Clin Radiol 21 10 vessels (fig 2 B). The left accessory middle 9 Komiyama M, Nishikawa M, Yasui T. The cerebral artery was not stenotic despite the accessory middle cerebral artery as a collateral 5 blood supply. AJNR Am J Neuroradiol 1997;18: vicinity of the markedly stenotic distal 587–90. 0 internal carotid artery and middle cerebral 10–14 15–19 20–24 25–29 30–34 35–39 40–44 45–49 artery. Moyamoya vessels were not supplied Maternal age (y) Maternal age is not a risk factor for by the left accessory middle cerebral artery. Figure 1 Proportion of subjects by maternal The patient was conservatively followed up Parkinson’s disease age. for 6 months without any ischaemic episodes. An association of the accessory middle cer- The aging process is associated with an accu- age was associated with risk of Parkinson’s ebral artery and cerebral aneurysms has been mulation of oxidative damage to mitochon- 12 disease. All probability (p) values are two well documented.26 Moyamoya disease is drial DNA (mtDNA). Mutations and dele- tailed. highly associated with primitive carotid- tions of mtDNA accumulate with aging in The Parkinson’s disease group and the 34 basilar anastomosis, such as with the primi- various tissues including germ lines. The control group each comprised 106 men tive trigeminal arteries and their variants.7 To resulting mitochondrial defects, if transmit- (60%) and 71 women. The mean age (SD) of our knowledge, however, an association of the ted to oVspring through the maternal line, the patients was 67 (10) years compared with accessory middle cerebral artery with could potentially play a part in the pathogen- the control group mean age (SD) of 66 (10) moyamoya disease has not been reported. esis of several neurodegenerative illnesses (p=0.41). The age of onset of symptoms of The accessory middle cerebral artery was including Parkinson’s disease.5 In addition, Parkinson’s disease among the patient group first regarded as a hypertrophied recurrent advanced maternal age at the time of concep- ranged from 32–82 years with a mean (SD) of artery of Heubner.8 However, it is now tion results in increased risk of genetic birth 59 (11) years. Sixty three per cent of the thought to be a cortical branch of the middle defects. This is in part due to chromosomal patients received their diagnosis within 2 cerebral artery supplying the anterior frontal mutations and malfunctions within the ova years of symptom onset (range<1 year to 10 lobe, which is annexed to the embryological which are in turn due to the increased age of years, with a mean (SD) of 1 year (1.8)). early artery, the anterior cerebral artery.5 The the ovary and the ova. We examined the age The age of mothers at the time of the accessory middle cerebral artery can serve as of mothers at birth of patients with Parkin- patients’ birth ranged from 13 to 43 years a collateral blood supply when the internal son’s disease and controls to evaluate whether with a mean (SD) of 27 (7) years. In the con- carotid artery or middle cerebral artery, or maternal age may be a risk factor in the trol group, maternal age ranged from 15–49 both are stenotic or occluded,9 as was the case development of Parkinson’s disease. years with a mean (SD) of 27 (7) years. There in our patients. Subjects were recruited from the Move- wasnodiVerence between the groups for Our patients are interesting in that (a) the ment Disorders Clinic at the University of mean maternal age (p=0.63). accessory middle cerebral artery was associ- Virginia. We interviewed 629 consecutive Maternal age was recoded into eight ated with moyamoya disease and (b) the patients with Parkinson’s disease and 376 groups of 5 year increments. The percentage accessory middle cerebral artery was not consecutive spousal controls regarding their of patients and controls in each maternal age stenotic even though the distal internal mother’s age at the time of the subject’s birth group is shown in figure 1. Not surprisingly, carotid artery and the proximal middle (maternal age). The diagnosis of Parkinson’s the greatest number of patients and controls cerebral artery showed steno-occlusive disease was based on internationally accepted were born to women age 20 to 35 years. 2 changes. Stenotic changes were not seen in criteria. At least two of the following three Analysis by × showed no significant diVer- the accessory middle cerebral artery although criteria had to be present: rest tremor, ences in the proportion of patients and it coursed in the vicinity of the stenotic hori- cogwheel rigidity, and bradykinesia. Asym- controls across the maternal age categories zontal portion of the middle cerebral artery. metry of these features at the time of diagno- (p=0.09). This implies that susceptibility to arterial sis and at onset had to be present. Exclusion Logistic regression examined the associ- stenotic change is limited to the distal portion criteria included presence of atypical fea- ation between maternal age and Parkinson’s of the internal carotid artery and the tures, presence of a pre-existing possible disease. The odds ratio (0.99) for maternal proximal middle cerebral artery but not to cause, definite absence of response to levo- age was not significant (p=0.63; 95% confi- the accessory artery even though all of these dopa, and a clearly non-progressive course dence interval (95% CI) 0.96–1.02) indicat- vessels are in close proximity. The cause of over at least 3 years. Moreover, 69% of the ing that maternal age did not influence the risk of developing Parkinson’s disease. moyamoya disease is still unknown, but we patients with Parkinson’s disease had been Our data show no diVerences between think that there is a topographic diVerence in examined on multiple occasions thus increas- patients and controls for the age of mothers at the predilection to stenotic changes in the ing both our confidence in and accuracy of the time of the subjects’ birth. In fact, the cerebral vasculature in the disease. the diagnosis of Parkinson’s disease. distribution curve of mothers’ age at the time The original data set of 1005 subjects was M KOMIYAMA of birth of patients with Parkinson’s disease T YASUI reduced to obtain groups of equal size as well closely approximated the maternal birth Department of , Osaka City General as groups similar in sex proportion and age. curve of the control group. In addition, we Hospital, 2–13–22 Miyakojima-Hondohri, Firstly, we excluded 79 cases in which mater- Miyakojima, Osaka 534–0021, Japan found no diVerences in the proportion of nal age was missing. This elimination of cases patients and controls when maternal age was Correspondence to: Dr M Komiyama with missing data resulted in the groups being [email protected] grouped in 5 year increments. Although a of similar age. From the new data set of 926 greater number of patients were born to informants, we next randomly selected men women between ages 35 and 44, and only and women from each subject group. The 1 Kudo T. Spontaneous occlusion of the circle of control group subjects were born to women Willis: a disease apparently confined to Japa- number of subjects to be selected was deter- over 44 years of age, these diVerences were nese. Neurology 1968;18:485–96. mined by the minimum group, the male con- not significant. Furthermore, the odds ratio 2 Crompton MR. The pathology of ruptured trol group (n=106), for which we were able to did not approach significance, indicating that middle cerebral aneurysms with special refer- ascertain maternal age and also preserve the ence to the diVerences between the sexes. Lan- maternal age also did not aVect the risk of cet 1962;ii:421–5. 1.5:1 male to female sex ratio found in the developing Parkinson’s disease. 3 Jain KK. Some observations on the anatomy of original sample. After this final exclusion of The results of this study suggest that the middle cerebral artery. Can J Surg 1964;7: 572 cases, complete data from 177 patients genetic mutations acquired by aging oocytes 134–9. 4 Watanabe T, Togo M. Accessory middle cer- with Parkinson’s disease and 177 control are not pivotal in the development of Parkin- ebral artery. Report of four cases. J Neurosurg subjects were used for the data analyses. son’s disease. If acquired mtDNA mutations 1974;41:248–51. Data were analyzed using Student’s t test of female gametes were a significant factor in 5 Komiyama M, Nakajima H, Nishikawa M, et al. for continuous variables and ÷2 tests for the pathogenesis of Parkinson’s disease, Middle cerebral artery variations: duplicated and accessory arteries. AJNR Am J Neuroradiol categorical variables. Logistic regression was maternal age at birth would be higher for 1998;19:45–9. performed to determine whether maternal patients with Parkinson’s disease than for age

www.jnnp.com J Neurol Neurosurg Psychiatry 2001;71:129–137 131 matched controls. These data show no diVer- ence in maternal age at birth between patients and controls. Thus, transmission to oVspring of somatic mtDNA mutations that accumulate as the mothers age is unlikely to play a part in the cause of Parkinson’s disease. However, the results of this study do not exclude a role for inherited abnormalities of mtDNA mutations in this disease. Homo- plasmic polymorphisms or heteroplasmic sequence abnormalities could still account for a proportion of those with Parkinson’s disease. Indeed, either of these possibilities, especially heteroplasmy, is consistent with the high degree of variability in the clinical expression of mtDNA mutations and the apparent sporadic occurrence of this disease. L J CURRIE M B HARRISON J M TRUGMAN J P BENNETT R H SWERDLOW C A MANNING G F WOOTEN Department of Neurology, University of Virginia Health System, 500 Ray C Hunt Drive, Charlottesville, Virginia 22903, USA Correspondence to: Dr L J Currie [email protected]

1 Sohol RS, Weindruch R. Oxidative stress, caloric restriction, and aging. Science 1996;273:59–63. 2 Wallace DC. Mitochondrial diseases in man and mouse. Science 1999;283:1482–8. 3 Shigenaga MK, Hagen TM, Ames BN. Oxida- tive damage and mitochondrial decay in aging. Proc Natl Acad Sci USA 1994;91:10771–8. 4 Graeber MB, Grasbon-Frodl E, Eitzen UV, et al. Neurodegeneration and aging: role of the second genome. J Neurosci Res 1998;52:1–6. 5 Bowling AC, Beal MF. Bioenergetic and oxida- tive stress in neurodegenerative diseases. Life Figure 1 (A) and (B) show flare sequence axial MRI of twin 1 and twin 2 respectively Sci 1995;56:1151–71. demonstrating virtually identical lesions in the region of both internal capsules highlighted with arrowheads. (C) and (D) show flare sequence coronal MRI scans of twin 1 and twin 2 respectively through the temporal lobe regions showing high signal changes in the hippocamps bilaterally Hippocampal involvement in identical (highlighted with arrowheads) more marked in twin 1 than twin 2. twins with neurofibromatosis type 1 findings. Because of the acute memory prob- although the changes were less marked in the Neurofibromatosis type 1 (NF1) is an lems herpes simplex encephalitis was consid- limbic regions in twin 2 than twin 1. The sole autosomal dominant disorder due to a defect ered despite a normal EEG and CSF major diVerence was the normal optic chiasm on the long arm of chromosome 17. Since examination and he was given acyclovir, in twin 2. Neuropsychometry on twin 2 1986, when Cohen et al1 first drew attention without benefit. His brain MRI showed mul- disclosed a full scale IQ in the average range to changes on MRI in NF1, it has increas- tiple areas of increased signal in the medulla, with no evidence of significant amnesic prob- ingly been recognised that up to 75% of chil- pons, midbrain, and internal capsule regions lems. dren with NF1 have lesions, seen most often (fig 1 A), prominent bilateral lesions in the There are three main areas of interest that in the brain stem, cerebellum, optic tracts, , amygdala, and medial tempo- arise from these twins. The first is how and basal ganglia.2 They are asymptomatic, ral cortex (fig 1 C), and enlargement of the remarkably similar are the changes on brain usually less prominent with age, and the lim- optic chiasm with increased signal strongly MRI. The second is that unlike previous ited available postmortem data suggest that suggestive of an optic glioma. Extensive studies6–8 they are discordant for optic nerve they involve areas of spongiotic or vacuolar blood tests including treponemal serology gliomas. The third, and perhaps most intrigu- change.3It is unclear how often temporal lobe and chest radiographs were normal. Psycho- ing, is twin 1’s amnesiac syndrome, the cause structures are involved and figures range from metric testing confirmed twin 1 to have an IQ of which remains unestablished. He does not 0% to 16%.45We report on a pair of identical in the high average range but to have a dense seem to have any of the conditions known to twins with NF1, with prominent bilateral generalised anterograde amnesia. He was cause profound amnesia with changes on changes in hippocampal MRI, one of whom able to recall a story immediately it was told MRI, which include herpes simplex encepha- presented with a major amnesic syndrome. but at 5 minutes could only recollect one out litis, paraneoplastic limbic encephalitis, and The twins, aged 15, were the only children of 20 details. Similarly, his immediate copy of possibly neurosyphilis.9 As he has NF1 asso- of non-consanguineous unaVected parents the Rey-Osterrieth complex figure was nor- ciated with bilateral hippocampal changes on with no known family history. Twin 1 had mal (32/36) but he had almost no recall of the MRI it is tempting to speculate that the com- been considered bright throughout his figure at 30 minutes (1/2 out of 36). Repeat bination of the minor head injury and the schooling, until in summer 1999 he was MRI and neuropsychological assessment 1 lesions in the limbic cortex are responsible. involved in a relatively minor fight when year later showed no significant changes. Less likely explanations are malignant gli- another boy punched him a few times and the Because of the lack of a clear diagnosis on omatous transformation of hippocampal two boys fell onto a grass lawn without losing twin 1 we also studied his identical twin hamartoma or spread of malignant tissue consciousness or otherwise sustaining serious brother. Southern blot analysis confirmed from his optic nerve glioma. head injuries. Five days later his mother their monozygosity. Aged 2, twin 2 developed Changes in MRI specifically within the became aware of him having memory diY- a right facial plexiform neurofibroma and hippocampus have apparently not been the culties; he twice forgot to take a letter to aged 8 a plexiform neurofibroma involving subject of detailed analysis in NF1 and this school and then came home recollecting his tongue and larynx was surgically excised. paper suggests the need for further study of nothing of an evening spent at a church youth He also had typical cutaneous lesions. His hippocampal structure and function in NF1. group. On admission to hospital he was noted brain MRI showed remarkably similar to have greater than 15 café au lait spots and changes to those seen in twin 1 in the We thank Professor M Patton, Department of axillary freckling. Visual acuities were normal medulla, pons, midbrain, internal capsule (fig Clinical Genetics, St. George’s Hospital Medical and there were no major focal neurological 1 B), and hippocampal regions (fig 1 D) School for help with the Southern blots and Lister

www.jnnp.com 132 J Neurol Neurosurg Psychiatry 2001;71:129–137

Bestcare MRI Unit at Mayday Hospital, Croydon, for financial support with the MRI studies. ‡ 250 N TUBRIDY F SCHON Neurology Department, Atkinson Morley’s Hospital, Copse Hill, Wimbledon, SW20 ONE, UK BP (mm Hg) A MOSS 90 Paediatric Psychology Department, St George’s † Hospital, Cranmer Terrace, London SW17 ORE, UK 1 A CLARKE (mV) Paediatric Neurology Department MNSA 0 TCOX Radiology Department, Hospital for Sick Children, Great Ormond Street, London WC1N 3BG, UK R FERNER ECG Neurology Department, Guy’s and St Thomas’s Hospital, St Thomas’s Street, 0 1 2 London SE1 9RT,UK Time (min) Correspondence to: Dr F Schon Figure 1 A 2 minute recording of blood pressure (BP), muscle nerve sympathetic activity (MNSA), and heart rate (ECG) during an absence attack after 10 minutes of head up tilt. At 30 seconds there 1 Cohen ME, DuVner PK, Kuhn JP, et al. was sudden mental withdrawal and a rapid increase in MNSA* followed by a severe and paroxysmal in neurofibromatosis. Ann Neu- rol 1986;20:444. increase in blood pressure and heart rate. As blood decreased, MNSA increased and when blood 2DuVner PK, Cohen ME, Seidel G, et al. The pressure normalised there was a marked baseline shift in MNSA‡. During recovery, blood pressure significance of MRI abnormalities in children and MNSA oscillated reciprocally (0.1 Hz)†. with neurofibromatosis. Neurology 1989;39: 373–8. is normally maintained during head up tilt by aline (norepinephrine) concentrations were 3 DiPaulo DP, Zimmerman RA, Rorke LB, et al. 2 Neurofibromatosis Type 1: Pathologic sub- increased MNSA and vasocontriction. 1650 pmol/l and 5250 pmol/l before tilt and strate of high-signal-intensity foci in the brain. The patient showed normal blood pres- during recovery respectively. Normal values Radiology 1995;195:721–4. sure, heart rate, and MNSA responses to tilt in our laboratory before and after 10 minutes 4 Ferner RE, Chaudhuri R, Bingham J, et al.MRI initially, but after 10 minutes, he suddenly of tilt are 456 (SD 50) and 705 (SD 74) in neurofibromatosis 1. The nature and evolu- 3 tion of increased intensity T2 weighted lesions became pale, sweaty, and withdrawn for pmol/l. His absence symptoms could not be and their relationship to intellectual impair- about 30 seconds. No loss of muscle tone was reproduced by rapidly increasing blood pres- ment. J Neurol Neurosurg Psychiatry 1993;56: seen and he later confirmed that this was a sure to similar values (250/120 mm Hg for 30 492–5. typical absence attack. Coinciding with the seconds) with an intravenous bolus of epine- 5 North K, Joy P, Yuille D, et al. Specific learning disability with neurofibromatosis type 1: sig- onset of his symptoms, MNSA increased phrine (100 µg). One week later, an EEG nificance of MRI abnormalities. Neurology briefly for 3 seconds associated with a sudden during a similar absence attack showed sharp 1994;44:878–83. increase in blood pressure from 138/95 to waves arising from the left frontoparietal area 6 Cartwright SC. Concordant optic glioma in a pair of monozygotic twins with neurofibroma- 222/150 mm Hg over 10 seconds. Heart rate (fig 2). Subsequent continuous EEG and tosis. Clin Pediatr (Phila) 1982;21:236–8. simultaneously increased from 65 to 98 bpm blood pressure monitoring confirmed that 7 Crawford MJ, Buckler JMH. Optic gliomata (fig 1). Over the next 20 seconds,blood pres- focal seizure activity was simultaneous with aVecting twins with neurofibromatosis. Dev sure and heart rate decreased and there was a paroxysmal hypertension. Studies with MRI Med Child Neurol 1983;25:370–3. 8 Pascual-Castroviejo I, Verdu A, Roman M, et al. major burst of MNSA followed by reciprocal showed hippocampal atrophy consistent with Optic glioma with progressive occlusion of the oscillation of blood pressure with MNSA the diagnosis of complex partial seizure aqueduct of Sylvius in monozygotic twins with (0.1Hz) as blood pressure reached normal disorder. His absences were abolished with neurofibromatosis. Brain Dev 1988;10:24–9. 9 Denays R, Collier A, Rubinstein M, et al.A51 levels. During recovery, he complained of his 400 mg carbamazepine daily and he has year old woman with disorienatation and usual transitory headache. Venous noradren- remained free of symptoms for 6 months. amnesia. Lancet 1999;354:1786.

Paroxysmal hypertension during a complex partial seizure

The autonomic mechanisms involved in neu- rogenic paroxysmal hypertension are not understood. We present the first demonstra- tion of the precise haemodynamic and autonomic changes during a complex partial seizure. A 50 year old headmaster was investigated for an 8 year history of recurrent absence attacks, stereotyped in nature and of sudden onset, each lasting about half a minute. He became pale, sweaty, and mentally withdrawn but did not fall down. Recovery was rapid and associated with transient headache. Previous neurological investigations, including re- peated EEG and MRI, were negative. Elec- trocardiographic Holter monitoring disclosed only sinus bradycardia so he underwent head up tilt testing to exclude vasovagal syncope. Intra-arterial blood pressure and ECG were recorded continuously. Microneurography needles were positioned in the peroneal nerve of the right leg for recording eVerent postganglionic MNSA.1 This technique al- lows beat to beat monitoring and quantifica- 30 mm/s tion of MNSA (bursts/min) which controls Figure 2 EEG recording obtained during light sleep showing sharp waves arising maximally in the vascular tone in skeletal muscle. MNSA in left frontotemporal area. The montage consists of four sets of channels running anterior to posterior turn is modulated by changes in blood recorded from the right parasagittal, left parasagittal, right temporal, and left temporal areas pressure via the baroreflexes. Blood pressure respectively.

www.jnnp.com J Neurol Neurosurg Psychiatry 2001;71:129–137 133

This is the first demonstration of paroxys- vate Bag 4710, Christchurch, New Zealand A square 2 cm×2 cm was drilled in the mal neurogenic hypertension triggered by a [email protected] centre of the cap. formalin:water (1:5; seizure in a patient with complex partial 30 ml) was injected subdurally. The cadaver seizure. The diagnosis of complex partial sei- 1 Valbo AB, Hagbath KE, Torebjork HE, et al. was left in the cold room for 24 hours. Then zure was supported by the following: focal Somatosensory, proprioceptive and sympa- a mixture of phenol:formalin:water (3:1:10) EEG changes during a subsequent absence thetic activity in human peripheral . was introduced into the femoral artery about seizure; no reproduction of absence symp- Physiol Rev 1979;59:919–57. 2 Eckberg DL, Sleight P. Human baroreflexes in 5 cm below the inguinal ligament through a 3 toms during drug induced paroxysmal hyper- health and disease. Oxford: Oxford University mm diameter cannula. Before preparation the tension; characteristic hippocampal atrophy Press, 1992:255–99. cadaver remained in this phenol:formalin:wa- on MRI4; and complete response to anticon- 3 Jardine DL, Melton IC, Crozier IG, et al. The neurohormonal response to head-up tilt and its ter mixture for about 6 to 8 months. vulsant drugs. Other possible diagnoses role in vasovagal syncope. Am J Cardiology The vascular anomaly reported was noted including brain stem tumour, phaeochromo- 1997;79:1302–6. when the brain was moved carefully back- cytoma, and renal artery stenosis were 4 Lee DH, Gao F, Rogers JM, et al. MR in tempo- wards. The brain, which was otherwise excluded by appropriate imaging and neuro- ral lobe epilepsy: adults with pathological con- firmation. Am J Neuroradiol 1998;19:19–27. normal, was then dissected out of the cranial hormonal analysis. Pseudoseizures were ex- 5 Wilkinson DJC, Thompson JM, Lambert GW, cavity. The external radius of the anterior cluded on the basis of the EEG findings and et al. Sympathetic activity in patients with panic inferior cerebellar artery was determined by a the rapid response to treatment. Although disorder at rest, under laboratory mental stress, and during panic attacks. Arch Gen Psychiatry digital gauge. rapid increases in MNSA and heart rate have 1998;55:511–20. During the exploration of the posterior been found during panic attacks, paroxysmal 6 Devinsky O, Price BH, Cohen SI. Cardiac cranial fossa of the cadaver, we encountered manifestations of complex partial seizures. Am hypertension and loss of consciousness are an unusual course of the right anterior not consistent features.5 JMed1986;80:195–202. inferior cerebellar artery (1.20 mm diam- The paroxysm consisted of simultaneous eter), which arose from the lower third of the hypertension and tachycardia associated with basilar artery and passed through the facial sweating and facial pallor during the absence nerve 1.5 cm from the cerebellopontine attack. We suggest that this is secondary to a Compression by looping and angle, forming a ring around the nerve about generalised increase in sympathetic activity perforation of the facial nerve by the causing vasoconstriction and increased car- anterior inferior cerebellar artery: a 0.5 cm proximal to the point of penetration and compressing it at several points (fig 1). diac output. This is supported by (a) possible cause of facial tic increased MNSA and heart rate despite pro- The anterior inferior cerebellar artery first gressive rise in blood pressure; (b) sympto- Hemifacial spasm is a distressing, common, ran vertically downwards closely following matic blood pressure overshoot; (c) nor- and well defined condition, which is diYcult the basilar and the right vertebral arteries. adrenaline increased to over seven times the to treat. It is characterised by clonic and tonic Then it took a horizontal course until it made normal tilt levels; (d) prominent low fre- contractions of the muscles supplied by the a V shaped angle before penetrating the sev- quency (0.1Hz) oscillations in blood pressure facial nerve, is intermittent, and is usually enth cranial nerve. After it had passed and MNSA during recovery.2 These low worsened by fatigue or emotional upsets. through the facial nerve, with about one third frequency oscillations (0.1 Hz) are thought to A clinical picture similar to that of hemifa- of the fibres above it and two thirds below, its be secondary to changes in brain stem cial spasm is also seen in trigeminal neuralgia, course made a ring around the seventh sympathetic activity separate from the eVects another condition for which widely varying cranial nerve and disappeared between the of respiration, which are generally of a higher pathophysiological bases, including vascular cerebellum and the pons into the deep tissue. frequency (0.2 Hz). We emphasise that the compression, have been proposed.1 Along its course a branch went oV to the initial increase in MNSA occurred when In our case postmortem exploration of the choroid plexus of the fourth ventricle. blood pressure was increasing and so was not posterior cranial fossa disclosed a strikingly The cause and treatment of hemifacial baroreflex mediated as would be expected for abnormal relation between the anterior infe- spasm remain controversial. Whereas some respiratory or normal brain stem low fre- rior cerebellar artery and VIIth cranial nerve. theories postulate a cerebral or brain stem quency oscillations. Such abnormal vascular anomalies around mechanism for its origin, others suggest that We hypothesise that this generalised in- the facial nerve are repeatedly reported in the the causative lesion is within the facial nerve, crease in sympathetic activity is permitted by literature, and they seem to be closely corre- either within the posterior fossa or at a more a transient interruption of baroreflex feed- lated with hemifacial spasm.2 distal location. back inhibition during the seizure. We think We took the cadaver of a man who had died Our unusual finding lends support to the that this is a unique recording of transient aged 51, from the dissection material held by neurovascular compression theory to explain baroreflex failure characterised by a rapid and the First Department of Anatomy at the Uni- the aetiology of this disorder as we would like generalised increase in sympathetic activity, versity of Vienna. to point out the unusual nature of this overriding the baroreflex aVerents in the brain stem. It has long been suspected that paroxysmal hypertension occurs in complex partial seizures but to date, ambulatory monitoring has only demonstrated changes in heart rate.6 Ambulatory beat to beat blood pressure monitoring would allow closer study of this phenomenon and its possible relation to sudden cardiac death in epileptic patients. Finally, this a good example of an episodic medical condition which may be very diYcult to diagnose. Occasionally, when an episode is seen fortuitously in the laboratory, we may identify pathophysiology previously sus- pected but not actually seen.

We are grateful for the advice of Dr M Hurrell and Dr GJ Carroll and the technical assistance of Mrs J Sutherland. The figures were prepared by the Department of Medical Illustration, Christchurch Hospital. DL JARDINE I G CROZIER H IKRAM Department of Cardiology, Christchurch Hospital, New Zealand T J ANDERSON Department of Neurology Correspondence to: Dr D L Jardine, Department Figure 1 Inferior surface of the brain. A=Anterior inferior cerebellar artery; F=facial nerve (seventh of General Medicine, Christchurch Hospital, Pri- cranial nerve); C=cerebellum; P=pons; V=vestibulocochlear nerve (eighth cranial nerve).

www.jnnp.com 134 J Neurol Neurosurg Psychiatry 2001;71:129–137 anatomical variety of the anterior inferior hemifacial spasm using oblique sagittal MR 5 Vytche DH, Howard RJ. The perceptual conse- cerebellar artery. imaging. Acta Neurochir (Wien) 1998;140:565– quences of visual loss: “positive” pathologies of 71. vision. Brain 1999;122:1247–60. During an investigation of the posterior 4 Ryu H, Yamamoto S, Sugiyama K, et al. Hemi- cranial fossa, we became aware of an facial spasm caused by vascular compression of abnormal relation between a vessel and a the distal portion of the facial nerve. J V Neurosurg 1998;88:605–9. E ects of topiramate on cognition nerve, which is not described in the current 5 Jannetta PJ. Typical or atypical hemifacial textbooks and encyclopedias. Although there spasm. J Neurosurg 1998;89:346–7. This letter concerns the recently published is much literature reporting a close relation study by Thompson et al,1 reporting the between the symptom “facial tic” and vessel authors’ findings on cognitive eVects with 34 variety ; our variety—that is, with both topiramate. Firstly, I want to correct the transfixion of the facial nerve and an arterial authors’ mischaracterisation of a review loop around the same nerve—has not been CORRESPONDENCE paper of mine. The authors state that “the lit- described in the specialist literature, nor has it erature on antiepileptic drugs ... emphasised been mentioned in the most recent review of positive psychotropic eVects,” referencing variants. only a 1998 review in which I discussed cog- Several authors have provided illustrations Charles Bonnet syndrome: an example nitive and psychotropic eVects of antiepilep- of a loop formed by the anterior inferior cer- of cortical dissociation syndrome tic drugs. Although I mention some positive ebellar artery, but without elaborating further V eVects, I also discussed negative eVects, on the topographic relation between the a ecting vision? including studies from my own centre that artery and the internal acoustic meatus or the This letter is a response to that by Abhijit have shown significant negative psychotropic seventh and eighth cranial nerves. It has been Chaudhuri.1 I thank Chaudhuri for his kind and cognitive drug eVects. asserted that this vessel seldom appears on comments, but will address his surprise that I Secondly, I provide some perspective on radiographs. did not comment on the Charles Bonnet syn- the report of Thompson et al of clinically sig- Typical hemifacial spasm is caused by ves- drome in my 1999 paper.2 sels on the anterocaudal side of the nerve, nificant cognitive declines in 18 patients Chaudhuri and his associates cannot be treated with topiramate as adjunctive therapy. whereas atypical hemifacial spasm is caused really criticised for using the eponym by vessels on the posterior rostral side.5 The authors correctly conclude that “caution “Charles Bonnet syndrome” for the “triad of is warranted in the interpretation of the find- The relevant aspect of this article lies in its visual hallucinations, visual sensory depriva- emphasis on the connection between the ings due to methodological limitations of the tion, and preserved cognitive status” as there study design.” Because their study was retro- neurological symptoms and this anatomical is support in the literature for the use of the variety of a nerve. spective and observational, it is susceptible to eponym in this way. But I favour the use of considerable subject selection bias. For exam- The deceased had begun to experience the eponym, if it is to be used at all, only in intermittent symptoms of varying intensity in ple, five of the 18 patients were specifically patients with eye disease. This was what included in the topiramate sample because his face at the age of 49. These symptoms Charles Bonnet described,3 and the way the they reported cognitive eVects. took the form of uncontrollable twitching at term was initially used by de Morsier in The only way to minimise eVects that may the right corner of his mouth, ipsilateral 1936.4 I highly recommend the paper by bias study conclusions is to conduct a hearing impairment, retroauricular cramps, Vytche and Howard5 in which the authors and retroauricular sensory impairment. Ac- provide an excellent summary of Charles prospective randomised controlled study. cording to the case history, the deceased had Bonnet syndrome on pages 1253–1254. I Two such studies have recently compared undergone a full otorhinolaryngological ex- agree with the way that they accept the use of topiramate and valproate as add on therapy to amination and pure tone speech audiometry the term, and their reason for doing so. It is, carbamazepine, using comprehensive neuro- during his lifetime. Thus, it was possible to thus, a matter of personal opinion as to how psychological batteries to objectively measure diagnose the perceptive unilateral acoustic the eponym should be used. It was never used drug eVects. hypoacousia on the right. Early auditory in any of the references in my 1999 paper, but At the end of 3 months of maintenance evoked potential studies had also been I did not choose the references with this rea- therapy, only one of 17 (6%) variables in one performed showing an increase in latency and son in mind. study2 and only two of 30 variables (7%) in the a decrease in amplitude without any deterio- Is there any reason to use the eponym at other3 were significantly worse for topiramate ration of morphology of the waves. It had not all? It may remind ophthalmologists that compared with valproate. For the three been possible to use MR for the diagnosis as visual hallucinations can occur with ocular variables with statistically significant diVer- he had a pacemaker in place. None of these disease and do not necessarily suggest a ences, the mean diVerences in change scores symptoms responded to therapy with botuli- neurological lesion. But, as the eponym has were modest. Analysis of individual data num toxin. now acquired two diVerent meanings I think showed that scores were unchanged or even In our case the compression of the facial that it leads to more confusion than clarifica- improved in most patients receiving topiram- nerve at several points could have led to irri- tion. ChaudhuriQs communication is a good ate and valproate. Statistically significant tation of the region supplied by the posterior example. I would thus suggest that it no diVerences could be accounted for by a auricular nerve and in this way to the longer be used. I did not use the term because minority of patients receiving topiramate in symptoms described above. I did not, and do not, think that the subject of whom scores deteriorated>1 SD from base- M MELLING my report (myself) has the Charles Bonnet line. I suggest that the patients reported by First Department of Anatomy, University of Vienna, syndrome. Nor, presumably, did the review- Thompson et al likely represent a similar sub- A-1090 Vienna, Austria ers of my paper and the Editor of this Journal. group of patients. H GÖRZER Finally, I must tell Chaudhuri that “ad- Physicians should be aware that a subgroup Department of Neuroradiology vanced age” hardly begins at 60 years. of patients treated with topiramate may M BEHNAM M COLE experience clinically significant cognitive ef- First Department of Anatomy Department of Neurology, Case Western Reserve fects. When these eVects occur, they are gen- University, 11100 Euclid Avenue, Cleveland, D STARKEL erally apparent to the patient or family mem- OH44106–5000, USA Department of Neurology, General Hospital of Vienna, bers and can therefore be monitored with Währinger Gürtel 18–20, Vienna, Austria Correspondence to: [email protected] routine clinical evaluations. Alternatively, a D KARIMIAN-TEHERANI brief cognitive test (for example, a verbal flu- ency test or symbol digit modalities test) First Department of Anatomy 1 Chaudhuri A. Charles Bonnet syndrome: an Correspondence to: Dr M Melling, Haslingergasse example of cortical dissociation syndrome should easily detect changes of the magnitude 29/1/12, A-1170 Vienna, Austria aVecting vision [letter]? J Neurol Neurosurg Psy- reported. In a subgroup of patients, topiram- chiatry 2000;69:704–5. ate may need to be discontinued if cognitive 2 Cole M. When the left brain is not right the right brain may be left: report of personal eVects do not resolve over time with slowed 1 Kitt CA, Gruber K, Davis M, et al. Trigeminal experience of occipital hemianopia. J Neurol titration or dosage reduction. neuralgia: opportunities for research and treat- Neurosurg Psychiatry 1999;67:169–73. ment. Pain 2000;85:3–7. 3 Flournoy T. Le case de Charles Bonnet. Hallu- K J MEADOR 2 Fukuda M, Kameyama S, Honda Y, et al. Hemi- cinations visuelles chez un viellard opéré de la Medical College of Georgia, facial spasm resulting from facial nerve com- cataracte. Archives de Psychologie (Geneve) 1120 15th Street (BA 3410), pression near the internal acoustic meatus. 1902;1:1–23. Augusta, GA 30912, USA Neurol Med Chir 1997;37:771–4. 4 de Morsier G. Les automatismes visuels. 3 Nagaseki Y, Omata T, Ueno T, et al Prediction (Hallucinations visuelles retro-chiasmatiques). Correspondence to: Professor K J Meador of vertebral artery compression in patients with Schweiz Med Wochenschr 1936;66:700–3. [email protected]

www.jnnp.com J Neurol Neurosurg Psychiatry 2001;71:129–137 135

1 Thompson PJ, Baxendale SA, Duncan JS, et al. verbal diYculties are more likely to be cial, as well as in venous congestion of the EVects of topiramate on cognitive function. J attributed to the underlying pathology and inferior colliculus. Concomitant cochlear Neurol Neurosurg Psychiatry 2000;69:636–41. 2 Aldenkamp AP, Baker G, Mulder OG, et al.A ongoing seizures than to a drug eVect. Topira- dysfunction should have been considered as multicenter, randomized clinical study to evalu- mate is a useful antiepileptic drug but it may a cause of hearing loss in this patient, or ate the eVect on cognitive function of topiramate lead to adverse cognitive changes and we need ruled out by further examination. compared with valproate as add-on therapy to to be alert to this. carbamazepine in patients with partial-onset sei- M KOFLER zures. Epilepsia 2000;41:1167–78. P THOMPSON Department of Neurology, Hospital Hochzirl, 3 Loring D, Meador K, Kamin M, et al. Topiram- S BAXENDALE ate (TPM) or valproate (VPA) added to Anna-Dengel-Haus, A-6170 Zirl, Austria J DUNCAN carbamazepine (CBZ) in adults with epilepsy: Correspondence to: Dr M Kofler L SANDER eVects on subjective and objective measures of markus.kofl[email protected] cognitive function. Epilepsia 2000;41(suppl Epilepsy Research Group, Institute of Neurology, Florence):113. University College London, UK and National Society for Epilepsy, Bucks SL9 ORJ, UK 1 Strauss C, Naraghi R, BischoV B, et al. Contral- The authors reply: ateral hearing loss as an eVect of venous Correspondence to: Dr P J Thompson congestion at the ipsilateral inferior colliculus We write in response to the letter of Meador. after microvascular decompression: report of a We concur that his review article discusses case. J Neurol Neurosurg Psychiatry both positive and negative psychotropic Postoperative hearing loss due to 2000;69:679–82. 2 Buettner UW. Akustisch evozierte Potentiale eVects of antiepileptic medication. venous congestion at the inferior Meador comments on the biased nature of (AEP). In: Stöhr M , Dichgans J, Buettner UW, colliculus, or cochlear dysfunction? et al,eds.Evozierte Potentiale. Berlin: Springer our patient sample. However, we think that 1996:411–86. we adequately emphasised that our study was I read with interest the article by Strauss 1 3 Chiappa KH, Hill RA. Brainstem auditory et al evoked potentials: interpretation. In: Chiappa not randomised or controlled but carried out on postoperative contralateral hearing loss in a clinical environment. We pointed out that KH, ed. Evoked potentials in clinical medicine. which developed on the third day after Philadelphia: Lippincott-Raven, 1997:199– some of the patients had been referred due to microvascular decompression for trigeminal 249. cognitive complaints. Furthermore we ac- neuralgia. They attributed the symptoms to 4 Kofler M. Evozierte Potentiale. In: Hufschmidt knowledged studies which reported no cogni- A, Lücking CH, eds. Neurologie compact. venous congestion at the ipsilateral inferior Stuttgart: Thieme, 1999:375–85. tive eVects of the drug including the valproate colliculus after dissection of the ponto- and topiramate study and drew attention to 5 Chiappa KH. Brainstem auditory evoked trigeminal vein, which was documented by potentials: methodology. In: Chiappa KH, ed. the conflicting findings. We agree that MRI. Symptoms resolved partially after Evoked potentials in clinical medicine. prospective randomised controlled trials are Philadelphia: Lippincott-Raven, 1997:157–97. intravenous rheologic medication for a total the way to minimise selection bias. However, 6 Berry MM, Standring SM, Bannister LH. of 19 days. The authors’ explanation for the they are not the be all and end all. Bone mar- Nervous system. In: Bannister LH, Berry MM, delayed postoperative hypacusis, however, Collins P, et al,eds.Gray’s anatomy. The row aplasia with felbamate and visual field merits further discussion. Strauss 1 anatomical basis of medicine and surgery.New constriction with vigabatrin treatment were et al provided preoperative and postoperative York: Churchill Livingstone, 1995:901–1398. not found in randomised controlled trials but recordings of brain stem auditory evoked by the careful clinical study of patients. This potentials: postoperatively, after stimulation is an analogous situation. Randomised con- Strauss replies: on the operated side, ipsilateral waves I trolled trials are not without their own biases We appreciate Kofler’s comment on our as most will be sponsored by the pharmaceu- through V, and contralateral waves II paper and his interest in this unusual and still tical industry and it would be naïve to through V are all clearly identifiable, con- poorly understood clinical picture. We agree conclude that this does not influence the trasting with stimulation on the non- that the brain stem auditory potentials presentation of the results. operated side, depicting only a small wave V (BAEPs) after contralateral stimulation do We agree with Meador that the adverse bilaterally, but no other components. This not clearly point to a lesion of the ipsilateral pattern suggests a left sided lesion involving eVects of topiramate reported are likely to colliculus; however, to our knowledge the the generator of wave I—presumably the occur in a minority of patients treated with of auditory pathways within auditory nerve near the cochlea—and is also the drug. However, we think that this may the brain stem is not yet fully understood. For consistent with the patient’s pancochlear represent a clinically significant number of example, in our series of more than 300 cases hearing loss.2–4 By contrast, brain stem patients, particularly in those attending terti- of acoustic neurinoma monitoring we have lesions—unless damaging the cochlear ary referral centres. Negative eVects, however made the observation that the contralateral small the numbers, are worthy of reporting nucleus— are usually not associated with pure tone hearing loss, but rather with wave V is much more pronounced compared and of further exploration. The question our with the ipsilateral wave V. The advantage of findings raise is not does topiramate have abnormal auditory localisation or interaural 25 this case report is the preoperative and post- adverse eVects but rather why does it have time discrimination, as auditory impulses 2–4 6 operative radiological and clinical documen- adverse eVects in some people? are conveyed bilaterally in the brain stem. Furthermore, a brain stem lesion causing tation. The delayed onset of symptoms We agree with Meador’s final point and several days after the surgical procedure, the indeed this was one of the intended take profound hearing loss is likely to produce also contralateral wave IV/V abnormalities lack of eVect of calcium blocker therapy— home messages of our paper. This is why we actually the patient’s pure tone audiogram chose to submit to a journal with a broad after stimulation on the non-aVected side, and speech discrimination deteriorated under readership who would have much less experi- but even the severest brain stem lesions, such nimodipine treatment—and the hearing im- ence with topiramate. We hoped that our as in evolving brain death, do not aVect wave 3 provement after heparinisation do not sug- paper would draw attention to a group of I. The vascular supply of the mesencephalic gest vasospasm as the underlying pathophysi- patients who should be prioritised for neuro- brain stem diVers from that of the inner ear, ological mechanism. The literature on this psychological monitoring and highlight the the first being fed by mesencephalic arteries type of measures that could be employed via the posterior cerebral or superior cerebel- rare yet important phenomenon of contralat- showing that an extensive assessment is not lar artery, and drained through the superior eral hearing loss after cerebellopontine angle necessary. petrosal vein; the second being supplied by surgery is purely speculative. By contrast this We, however, do not think that the cognitive the more caudally originating labyrinthine case report follows a straight course, which changes experienced would be obviously artery via anterior—or occasionally started at surgery with dissection of the pon- attributed to topiramate treatment. Most posterior—inferior cerebellar artery, and totrigeminal vein, followed by a delayed con- patients in the study were not referred because drained by the labyrinthine vein through the tralateral hearing loss, and ended with a of cognitive complaints. Six were being seen as posterior part of the superior petrosal or lesion of the ipsilateral colliculus. This lesion part of their presurgical assessment and indeed transverse sinus, and the internal jugular was not documented on preoperative MRI. were not considered good candidates due to vein.6 Although the patient’s hearing may Taking these findings into consideration, their neuropsychological test profiles. For have been partially aVected by the docu- together with the neurophysiological findings some the cognitive complaints did not occur in mented mesencephalic lesion, hearing loss of BAEPs in a still not fully understood audi- association with the introduction of topiramate may in fact be more likely caused by tory pathway within the brain stem, the “iso- or with any change in dosage and did not seem concomitant cochlear dysfunction. An is- lated vasospasm theory” seems unlikely. to develop until they had been on the drug for chaemic lesion seems probable, presumably C STRAUSS several months. For such patients, particularly postoperative vasospasm, or—less likely— Neurochirurgische Klinik der Universität those with left hemispheric pathology, in- unrelated embolism. In either situation, Erlangen-Nürnberg, Schwabachanlage 6, 91054 creases in word finding problems and other rheologic treatment may have been benefi- Erlangen, Germany

www.jnnp.com 136 J Neurol Neurosurg Psychiatry 2001;71:129–137

The text addresses the commonly encoun- This is the second series on neuropsychologi- tered problems of headache, sensory distur- cal rehabilitation fostered by Barbara Wilson BOOK REVIEWS bance, cerebrovascular disease, dementia, and Ian Robertson (the first being Neural confusion, movement disorders, fits and repair, transplantation, and rehabilitation by R faints, and back, leg, and arm pain. The Barker and S Dunnett). The renewed interest approach is a simple one—namely, a defini- is welcome. Josef Zihl is one of the first clini- tion of the condition or conditions, who gets cal scientists, together with collaborators Behaviour and mood disorders in focal it? What causes it? How does it present? What such as von Cramon, to have put a concen- brain lesions. Edited by JULIEN is the diVerential diagnosis? How is it investi- trated eVort into rehabilitation of visual BOGOUSSLAVSKY and JEFFREY L CUMMINGS. gated and managed? This pragmatic ap- disorders, spanning a period of 2 decades. (Pp554, £49.95). Published by Cambridge proach pays special attention to the preva- Much of this pioneering work lies in Ger- many, stemming from an older tradition University Press, Cambridge, 2000. ISBN lence of each disorder in the community and going back to Poppelreuter, and the current 0-521-77482-9. contains “red flags” where there is diagnostic doubt and uncertainty. The two accompany- book should make the developments better known outside that country. It remains ing CD ROMs outline learning outcomes, This book has arisen because of the com- something of a curiosity that visual rehabilita- case histories, self assessment questionnaires, parative dearth of studies examining the tion is scarcely pursued in Britain and the and case pathways with respect to investiga- emotional and behavioural consequences of United States, especially given the seminal tion, management, and prognosis. Each such focal brain lesions and, if nothing else animal work (by Cowey in Britain and Moh- provides a useful compendium of this litera- case is linked with the text prividing assign- ler and Wurtz in the United States) clearly ture in 18 chapters written by experts in the ments in clinical audit, comment, documen- demonstrating the potential capacity for field. The first three chapters set the scene by tation of data, and “setting standards”. Incor- recovery after V1 lesions in the monkey, with providing a conceptual overview and discus- porated in all of this are useful components practice regimes. Indeed, Zihl’s own work sions of methodological issues. The rest are such as data collection sheets for specific acknowledges the influence of these animal devoted to specific topics and generally take symptoms and summary sheets to aid and studies on his own research. the form of either examinations of the improve documentation. The Series preface describes this work as a evidence for the localisation of specific neu- I found this an interesting book and “modular handbook.” Handbook it is not. ropsychiatric syndromes (depression, mania, enjoyed finding my way around the CDs, Rather it is a compact and relatively short obsessive compulsive disorder, psychosis) or which were easy to use. I would imagine that account that is more like a progress report, reviews of the eVects of lesions of particular this would be a valuable aid for the busy gen- full of interesting approaches described in structures (, basal ganglia, frontal eral practitioner and should help set stand- mid-stream of the research, much of it still to lobe, temporal lobe). In their overview, ards in primary care that would aid patients be completed. It is broad ranging, the largest Cummings and Bogousslavsky group the and ensure thorough evaluation before refer- and most advanced section (about half of the disorders of mood, thought, aVect, and ral. The only thing missing was guidance book) being on visual field disorders, but motivation considered in this book under the demarcating between evaluation in the pri- progressing to visual acuity, colour vision, rubric “fundamental dysfunctions“ which mary care setting from the need to refer on visual space perception (including Balint’s they point out are associated with interrup- for a specialist opinion. All of us occasionally syndrome), visual agnosia, and a final brief tions of limbic and frontosubcortical cir- think, under moments of work overload, that section on the special problems of a central cuitry. This diVuse nature of these circuits is our colleagues in primary care could be a lit- scotoma. Each section starts usefully with reflected in the finding that lesions of dispa- tle more circumspect in whom they refer evidence of spontaneous recovery, which is rate areas can produce, for example, psy- (although the truth of the matter is that most the baseline from which rehabilitative eVorts chotic phenomena and that lesions in referrals are thoughtful and necessary). A text must be assessed. Such eVorts are almost specific structures—for example, thalamus such as this with no obvious competitor, is entirely based on intensive practice (what I or basal ganglia, can result in a range of long overdue, written by neurologists who, have called “grindsight”). Zihl is very cau- behavioural and emotional abnormalities. It along with their general practitioner col- tious, lacing the evidence with statements is thus often diYcult to make strong league, have adopted a common sense such as “our observations produce prelimi- inferences from this literature about the neu- approach to their task. I would strongly nary evidence that spatial contrast sensitivity ral basis of emotion and behaviour. Indeed recommend this book with its accompanying can, in principle, be improved by specific and some chapters are essentially categorised CD ROMs to my general practitioner col- systematic practice. However, the limited lists of studies without much in the way of number of cases does not allow definite con- theoretical underpinning. Several chapters, leagues in the belief that it would serve to demystify our specialty. Perhaps in future clusions to be drawn ...... ” (p 96). He is also by Habib (disorders of motivation), Eslinger careful to distinguish between measurable and Geder (behavioural and emotional editions the authors might consider incorpo- rating referral guidelines that would be a use- eVects in the laboratory and their “ecological changes after focal frontal damage) and value” to the subject in terms of any benefit in Tranel (neural correlates of violent behav- ful supplement to a text that, for example, takes general practitioners through from the everyday life. In some regards he is over- iour), stand out. These stray from localisa- cautious—for example, in his speculation tion and bring in evidence from basic neuro- suspicion of brain tumour to the process of biopsy and subsequent radiotherapy. While (with von Cramon) “that recovery of visual science and human functional neuroimaging field defects can only be expected with com- to inform and interpret fascinating case his- such data informs and allows the practitioner to explain to his or her patient what they plete striate cortex injury” (p 34). Of course, tories in terms of a functional neuroanatomy complete V1 lesions are almost always might expect to encounter in the process of of emotion and behaviour. accompanied by additional damage, when investigation and treatment, it does not animal work demonstrates a reduced residual EILEEN JOYCE clearly define who does what and when. capacity. But a complete but restricted V1 In conclusion, I recommend this innovative removal in the monkey still allows further text, which, alough directed primarily at our Update in neurology for general recovery to take place. general practitioner colleagues, contains practitioners.ByP ORTON, D BATEMAN, G Anyone wishing to find practical ap- something for all on the simple approaches of FULLER, P NEWMAN, D PARK, R SHAKIR, and G proaches for helping patients across a range evaluation, the importance of documenta- of visual disorders will find this a useful, YOUNG. (Pp 218 + 2 CD—ROM, £80.00) tion, and the value of using an interactive CD Published by University of Bath/British and humane, and dedicated book. He or she will ROM as an educational tool. I think that this have to work through a compact monograph Spine Foundatain/Royal College of General text goes a long way not just to educate and and share Zihl’s experience with him, rather Practitioner’s, Bath, 2000. entertain, but also remove from primary care, than find a collection of shorthand recipes. It the notion that neurology is a diYcult and will be rewarding work. The seven authors of this text are to be con- uncertain area. gratulated on their innovative approach and L WEISKRANTZ the freshness they bring to the evaluation of IAN BONE the common neurological problems encoun- An odd kind of fame. Stories of Phineas tered in primary care. This text and the two accompanying CD ROMs represent a coop- Rehabilitation of visual disorders after Gage. Edited by MALCOLM MACMILLAN (Pp erative venture between the British Brain and brain injury.ByJZIHL(Pp 187, £29.95, 552, £24.95). Published by Massachusetts Spine Foundation, the Royal College of Gen- Published by Psychology Press, Hove, 2000. Institute of Technology, USA, 2000. ISBN 0 eral Practitioners and the University of Bath. ISBN 0 86377 898 4. 262 13363 6.

www.jnnp.com J Neurol Neurosurg Psychiatry 2001;71:129–137 137

A brief description of the contents of this idiopathic headache syndromes. He shares a Headache in primary care. By SD book, which is 562 pages long and deals wealth of clinical experience in the drug SILBERSTEIN, RB LIPTON, PJ GOADSBY, and RT purely with the case of Phineas Gage, would therapy of diYcult headache patients, with SMITH (Pp 192, £29.95). Published by Isis lead most potential readers to throw up their necessarily brief descriptions of clinical fea- Medical Media, Oxford, 1999. ISBN hands in horror and to assume that the book tures and pathogenesis. Although there is an must be extremely repetitive and boring. extensive list of references at the end of the 1-901865-66-5. Nothing could be further from the truth. book, they are not cited directly in the text, A wide range of textbooks are now available Malcolm Macmillan’s approach in this book and the book reads as one man’s recipe book on the topic of headache, ranging from short has similarities to James Joyce’s concept when rather than a systematic review of published notes of 40 or 50 pages up to encyclopaedic writing Ulysses. Joyce thought that all of life’s trials, or of the pharmacological or clinical tomes. This book falls into the middle of the experiences could be encapsulated in the evidence supporting management decisions. range, being an easy read of 165 pages. description of 1 day in the life of his Dublin It is inevitable that the range of drugs recom- The work is divided into three broad characters. Malcolm Macmillan’s unstated mended is American, with more emphasis on thesis is clearly similar and that a great deal of narcotic and barbiturate combinations and categories: (1) pathophysiology and epidemi- 20th century can be no mention of domperidone or pizotifen, but ology of headache, (2) primary headache dis- understood by the analysis of the remarkable there is a useful glossary of United States orders, and (3) secondary headache disor- case of Phineas Gage. trade names as an appendix. ders. This book is the outcome of an obsession The chapters on the acute and preventative In the first section, the starting point is the with Phineas Gage which must have extended management of migraine, tension headache, International Headache Society (IHS) classi- over many years. It is superbly written and and cluster headache are followed by illustra- fication but the bridge to the clinical provides a fantastic reference source across a tive case histories, from which it is soon approach to patients is rapid and the use of wide range of topics. The first few chapters apparent that the tendency of United States disability assessment tools is discussed. The describe, in detail, the events surrounding the patients to see many neurologists and com- epidemiology of the various forms of head- fateful afternoon of 13 September 1848 when plain more about side eVects can lead to the ache is then examined, together with impact the unfortunate Phineas Gage was tamping use of an extremely wide range of drugs, on the suVerer. An entire chapter is devoted an explosive charge close to Cavendish in to diagnostic testing, particularly to exclude Vermont. The tamping iron was propelled sometimes making one’s own multiple eVorts on behalf of a challenging tertiary referral ominous causes of headache, and finally the through Phineas’ skull damaging his orbit pathophysiology of primary headache is and presumably destroying much of the seem straightforward by comparison. There is a particularly useful chapter on headache in discussed, the focus being on migraine. orbitofrontal cortex. The iron landed 20 Genetics, anatomy, and physiology are all metres away and can be seen in the Anatomi- patients aged over 50, post-traumatic head- ache, headache, chronic investigated. This whole section is concise cal Museum at Harvard University. There are and informative, giving a very good overview many revelations in this part of the book. It paroxysmal hemicrania, and SUNCT. of the subject. was certainly not known to me that the origi- Although filled with much clinical wisdom The rest of the book is divided into nal interest in Harlow’s description of Phin- from a vastly experienced author, the book chapters dealing with specific types of eas Gage revolved around the surgical proce- has the feel of a catalogue; for European dures Harlow carried out which saved Gage’s readers it is more likely to have a role as a headache, the first section looking at mi- life. It was only 20 years later, in 1868, that desktop reference text. graine, tension-type headache, chronic daily headache, cluster headache, and related Harlow described (very briefly) the changes RICHARD PEATFIELD in personality that have become enshrined in symptoms, the final section covering second- neurological and neuropsychiatric folklore. ary headache, including post-traumatic head- . Edited by JOHN The second section of the book is in many Vascular dementia ache, headache associated with disease of the ways the most fascinating. Malcolm Macmil- STIRLING MEYER et al (Pp 320, US$88.00). intracranial cavity, sinus headache, headache lan provides a scholarly account of the origins Published by Futura Publishing, Armonk, associated with CNS infection, headache of the concept of cognitive localisation in the 2001. ISBN 0 87993 425 5. associated with pregnancy and breast feed- brain outlining the contributions of Laycock, ing, and geriatric headache. Within these Hughlings Jackson, Ferrier, Gall, and Broca. This is a short text of 306 pages in 17 chap- chapters, all types of presentation and There is also a chapter on the contribution, or ters covering vascular dementia. management are discussed, the emphasis otherwise, of Gage’s story to the origins of Although it is a multiauthor text, there are being on the use and application of the IHS . Macmillan has unearthed a only 11 authors so many of the chapters have classification in a clinical setting. great deal of overlooked early literature on the authors in common. The principal editor, Dr This book undoubtedly contains the full use of various procedures used for the surgi- Meyer, coauthors no less than 10 of the 17 range of information necessary to deal with cal treatment of insanity in the late 19th and chapters. As a result of this many authorities the treatment of headache. However, the early 20th centuries. I was particularly are not included among the list of contribu- clinical chapters are written from a predomi- fascinated by the sections describing the tors and some of the chapters are not fully nantly American perspective and, therefore, eVects of frontal lobe resection and ablation. authoritative. the therapeutic approaches in particular do The following section of the book deals The text gives a fairly good overview of the not necessarily correlate with those used in with interpretations of the changes in Gage’s subject, although there are some unusually the United Kingdom. Because of the com- personality and its contribution to both the prominent inclusions including full chapters prehensive nature of the work, there is a dan- scientific and popular literature. The final on plasmaphoresis and estrogen replacement ger that the reader may not appreciate the chapter details the life of Dr John Martin therapy in the treatment of vascular demen- relative importance of some of the various Harlow. The main text is followed by facsimi- tia, neither of which are recognised treat- conditions discussed. For instance, my own les of the Gage papers. There are also repro- ments. Some of the more important recently experience has shown that the greatest clinic ductions of the CT of Phineas Gage’s skull. recognised conceptual issues, including the problem, even in a primary care setting, I can thoroughly recommend this book to very high preponderance of mixed dementia comprises patients with chronic daily head- anyone interested in the history of neuro- (vascular dementia and Alzheimer’s disease) ache and analgesic dependence. Important science, neuropsychology, or neuropsychia- and the issue of diagnosis before reaching the try. It is also amazing value at just under £25. basic clinical issues such as this may be lost state of being demented are scarcely covered within the wealth of information set out here. JOHN HODGES at all. The limitations of the current formal My overall impression is that this is a very diagnostic criteria for vascular dementia are comprehensive book which would serve as a touched on, but not considered in detail and Management of headache and headache good reference. However, although it would the chapter on diagnosis really does not help medications, 2nd edition.ByLAWRENCE D be extremely useful for the primary care doc- in this respect. ROBBINS tor in the United States where more speciali- (Pp 296, US$49.00). Published by Overall, this book provides an adequate Springer-Verlag, New York, 2000. ISBN background to the subject but a person sation exists, its value in the United Kingdom 0-387-98944-7. interested in the topic would need to look else- may be limited to the secondary care doctor where for some of the more recent critical or to the few primary care doctors who have Dr Lawrence Robbins, from Rush Medical a specific interest in headache. College, Chicago has revised his deservedly issues. popular 1994 text on the management of JOHN BOWLER A J DOWSON

www.jnnp.com