Journal of Neurology, Neurosurgery and Psychiatry (JNNP ...example, reading span38 (span=2, 6th percen-tile) or a computerised test of verbal working memory28 (level 3, 1st percentile)

Basal forebrain amnesia: does the nucleusaccumbens contribute to human memory?

Georg Goldenberg, Uwe Schuri, Olaf Grömminger, Ursula Arnold

AbstractObjective—To analyse amnesia caused bybasal forebrain lesions.Methods—A single case study of a patientwith amnesia after bleeding into the ante-rior portion of the left basal ganglia.Neuropsychological examination includedtests of attention, executive function,working memory, recall, and recognitionof verbal and non-verbal material, andrecall from remote semantic and autobio-graphical memory. The patient’s MRI andthose of other published cases of basalforebrain amnesia were reviewed tospecify which structures within the basalforebrain are crucial for amnesia.Results—Attention and executive functionwere largely intact. There was antero-grade amnesia for verbal material whichaVected free recall and recognition. Withboth modes of testing the patient pro-duced many false positive responses andintrusions when lists of unrelated wordshad been memorised. However, he con-fabulated neither on story recall nor inday to day memory, nor in recall fromremote memory. The lesion aVectedmainly the nucleus accumbens, but en-croached on the inferior limb of thecapsula interna and the most ventral por-tion of the nucleus caudatus and globuspallidus, and there was evidence of someatrophy of the head of the caudate nu-cleus. The lesion spared the nucleus basa-lis Meynert, the diagnonal band, and theseptum, which are the sites of cholinergiccell concentrations.Conclusions—It seems unlikely that falsepositive responses were caused by insuY-cient strategic control of memory re-trieval. This speaks against a major role ofthe capsular lesion which might discon-nect the prefrontal cortex from the thala-mus. It is proposed that the lesion of thenucleus accumbens caused amnesia.(J Neurol Neurosurg Psychiatry 1999;67:163–168)

Keywords: amnesia; basal forebrain; nucleus accumbens

Research in human amnesia has identifiedthree core regions where localised braindamage can cause a lasting amnesic syndrome:the medial temporal lobe, the thalamus, andthe basal forebrain.1 However, the weight of theevidence for a selective correlation betweencircumscribed lesions and amnesia is diVerentfor these three locations. There are many casereports showing that circumscribed lesions ofthe medial temporal lobes and of the medial

and anterior portions of the thalamus cancause amnesia.2–7 The evidence for a selectiverole of the basal forebrain in human memory isweaker and more indirect. It has originallybeen derived from degeneration of cholinergiccells within the basal forebrain in Alzheimer’sdisease.8 9 As amnesia is a core symptom ofAlzheimer’s disease it has been concluded thatbasal forebrain lesions can cause amnesia. Thisconclusion is not compelling as neuronaldegeneration in Alzheimer’s disease aVectsmedial temporal structures as well.10 11 Moredirect evidence for basal forebrain amnesia isprovided by patients with amnesia after cir-cumscribed damage to only the basalforebrain.12–22 In most of them, the lesion wascaused by haemorrhage from anterior commu-nicating artery aneurysms and subsequent inf-arction of perforating arterial branches orHeubner’s recurrent artery.12–14 16 17 19–22 How-ever, subarachnoidal or intraventricular haem-orrhage may cause hydrocephalus and diVusebrain damage. This raises the possibility that atleast some of the symptoms of these patientsare due to diVuse brain damage rather than tothe basal forebrain lesion.23

The anatomicoclinical relation of basal fore-brain amnesia is further complicated by theproximity of the basal forebrain to the frontallobe. Lesions from anterior artery aneurysmsoften aVect the orbital and medial frontallobe.12 17 24 25 Disinhibition or unconcern re-lated to frontal lobe pathology may distort orconceal the amnesic syndrome.16 Particularly,confabulations which are a frequent feature ofbasal forebrain amnesia have been related toaccompanying frontal lobe pathology ratherthan to the basal forebrain lesion itself.26

The uncertainty of the anatomicoclinicalrelation increases when localisation within thebasal forebrain is considered. Inspired by theparallel with Alzheimer’s disease, some authorshave considered damage to cholinergic cells inthe septum, the diagonal band, or the nucleusbasalis of Meynert as being the cause ofamnesia.12 18 An alternative account holds thatlesions of fibre tracts rather than cells are cru-cial for amnesia.15 20 The basal forebrain is tra-versed by the inferior and anterior thalamicpeduncles which connect the dorsomedialnucleus of the thalamus with the amygdalaeand with the prefrontal cortex. Amnesia may becaused by interruption of an amygdala-thalamic-prefrontal loop rather than by basalforebrain damage itself.1 15 27

In this paper we describe a patient withamnesia from a circumscribed basal forebrainlesion which did not extend into the frontal

J Neurol Neurosurg Psychiatry 1999;67:163–168 163

NeuropsychologicalDepartment,Bogenhausen Hospital,Munich, GermanyG GoldenbergU SchuriO GrömmingerU Arnold

Correspondence to:Dr G Goldenberg,NeuropsychologischeAbteilung, KrankenhausMünchen Bogenhausen,Englschalkingerstrasse 77, D81925 München, Germany.Telephone 0049 89 92702106; fax 0049 89 92702089; [email protected]

Received 20 October 1998and in revised form12 January 1999Accepted 19 January 1999

on August 15, 2021 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.67.2.163 on 1 A

ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from



j.com/

J Neurol N


ugust 1999. Dow

nloaded from

http://jnnp.bmj.com/













lobes and which largely spared structures richon cholinergic cells.

Case reportThe patient was a 55 year old right handedengineer who held a managing position in aninternational computer company. He had arte-rial hypertension. In July 1996 he becamesomnolent and confused. On emergency ad-mission, CT showed bleeding into the anteriorportion of the left basal ganglia with invasioninto the left frontal horn of the lateral ventricle.No source of bleeding could be identified onangiography. During the subsequent weeks themedical records noted mild right sided weak-ness, word finding diYculties, and paraphasia.His ability to store new information wasseverely defective and he was disoriented totime.

He was admitted as an outpatient to ourdepartment in October 1996. The hemiparesishad resolved and conversation was inobstru-sive. He was fully oriented, but complained ofmemory problems: He rated himself as verypoor at remembering names, places, and thecontent of conversations. His spouse con-firmed that he often enquired about details

which had been talked about only a fewminutes ago. In addition he noted an impover-ishment of his autobiographical memories anda loss of professional knowledge. He could notremember geographical facts, names of othercompanies, mathematical formulae, or phonenumbers which he had known by heart before.He said that he had lost his previous knowledgeof foreign languages. He had to translate labo-riously even English which had been the every-day language in his job. Visiting a place wherehe had spent several holidays, his wife and henoted that he had diYculties in finding familiarroutes. As well as his memory problems thepatient complained of a loss of libido and of aflattening of emotional reactions. He said:“There is nothing which makes me reallyhappy or really sad. It’s all far away”.

Neurological examination showed only aslight reduction of fine motor skill of the righthand, which disappeared completely duringthe subsequent weeks.

NEURORADIOLOGICAL EXAMINATION

Brain MRI (figure) showed a sharply demar-cated, slit-like, lesion which ran from thebottom of the left frontal horn in a slightlycurved way laterally. The caudal limit of thelesion was at the level of the anterior commis-sure. The lesion destroyed a large portion ofthe nucleus accumbens and encroached on theinferior limb of the capsula interna and themost ventral portions of the anterior nucleuscaudatus and globus pallidus. There was nogeneral enlargement of ventricles, but the wallof the left frontal horn was slightly widened andflattened, indicating some atrophy of the headof the caudate nucleus.

Tc-99-HMPAO SPECT showed hypoper-fusion of left basal ganglia (92% of homologu-ous right brain region), left frontal cortex(91%) and left basal and medial temporal cor-tex (91% and 90%).

NEUROPSYCHOLOGICAL EXAMINATION

AttentionThe patient scored well within the normalrange on a battery of computerised tests prob-ing reaction time, speed of visual scanning,divided attention, and shifting of attention.28

His attention/concentration index value on theWMS- R29 was 111.

LanguageIn German, his native language, speech waswell articulated, fluent, and syntactically cor-rect. Only when talking about diYcult andcomplex matters did his expression becomeimprecise and circumstantial with occasionalconfusions of low frequency words. He per-fectly named 20 highly familiar items from theSnodgrass-Vanderwart pictures30 31 but madetwo errors for 20 unfamiliar items (“perhaps agoose?” for ostrich, and skittle for spinningtop). Knowledge of second languages was notassessed.

IntelligenceOn a multiple choice lexical decision test withwords of decreasing frequency,32 reflecting pre-

(A) T 1 weighted MRI of the lesion. (B) Anatomical scheme of the centre of the lesion,corresponding to the leftmost image of the bottom row of the MRI. The right side of thefigures corresponds to the left side of the brain. GP=globus pallidus; Cd=caudate nucleus;Acb=nucleus accumbens; CI=capsula interna; DB=diagonal band.

Cd

GP

CI

DB

B

A

Acb

164 Goldenberg, Schuri, Grömminger, et al



j.com/

J Neurol N


ugust 1999. Dow

nloaded from


morbid intelligence level, his score was equival-ent to an IQ of 124. On the German WAIS-R33

his performance IQ was 121 and his verbal IQwas 100. In the verbal part the poorest resultwas on arithmetic (scaled score 6) whereas theother subtests yielded scaled scores between 10and 13. In the performance part scaled scoreson picture completion and block design weresuperior (18 and 15) and the other scaledscores ranged from 10 to 12.

Executive functionsHe obtained normal scores on the modifiedWisconsin card sorting test,34 the Tower ofLondon test,35 and the six elements test.36

Design fluency37 was normal (25/3 minutes:40th percentile of normal controls). Fluencyfor words with a given initial letter was loweralbeit still within the normal range (20/3 min-utes: 15th percentile).

Primary and working memoryVisual span was above average (WAIS-R:forwards 98th percentile; backwards 90th per-centile) and digit span average (forwards 51stpercentile, backwards 31st percentile). A weak-ness of verbal working memory manifesteditself in tests which demand both maintainanceand processing of verbal information—forexample, reading span38 (span=2, 6th percen-tile) or a computerised test of verbal workingmemory28 (level 3, 1st percentile). The poorresult on WAIS-R arithmetic may also relate tothis weakness.

Anterograde memoryA selection of tests of anterograde memory isgiven in the table. In accordance with thepatient’s complaints, the reports of his spouseand our findings, the Rivermead behaviouralmemory test39 confirmed ecologically signifi-cant memory impairment. Recall and recogni-tion of verbal information were severely defec-tive. By contrast, recall and recognition of non-verbal visual information were normal. On theBerlin amnesia test,40 which provides norma-tive data for comparing recall and recognition,recognition of words seemed to be less affectedthan free recall. A story with emotional contentwas better recalled than a neutral one. A nota-ble feature of his performance on tests of freerecall of words was the production of intru-sions. For example, in recall after interferenceof a German adaptation of the Californiaverbal learning test (CVLT)41 he producedmore intrusions than list words. Likewise, thepoor scores on verbal recognition memorywere largely due to false positive responses.Indeed, he recognised roughly equal numbersof faces and words, but whereas the total scoreon face recognition was normal, the score onverbal recognition fell to near chance levelsbecause of the many false positive responses.However, he confabulated neither in storyrecall nor in daily life.

Semantic memoryA preservation of basic semantic knowledgewas manifested by his average score on the

Anterograde memory

Test Result Comment

Rivermead behavioural memorybattery:Profile score / screening score 15 / 6 Defective

WMS-R index score:Verbal memory 83 Below averageVisual memory 104 AverageDelayed recall 78 Below average

Berlin amnesia test z scores:Verbal -3.23 Severely defectiveNumeric -0.12 AverageRecall/recognition verbal -1.88 Below averageRecall/recognition numeric -1.06 Average

Recognition memory:25 words: immediate/delayed(correct/false positive)

6 (16/10) / 5 (21/16) Severely defective

25 faces: immediate/delayed(correct/false positive)

18 (19/1) / 15 (17/2) Average

CVLT:List a: 5 trials (total intrusions) 4/4/7/6/6 (1) DefectiveList b 5List a: recall after interference/delayed (intrusions)

0 (5)/ 1 (1) Severely defective

Delayed recognition of list a (falsepositive)

14 (9) Severely defective

Learning of eight paired associates (4trials):Words 0/0/2/2 Defective with exception of trial 1-3 of face-name associationsObjects 2/3/4/4Face- name (delayed after 48hours)

5/6/7/7 (0)

Story recall (24 units of information):immediate / delayed

9 / 1 Below average / severely defective

Story recall (16 units of information):immediate / after 30 min / after 24hours

Emotional story (with cueing) 8.5 / 0 (5) / 0 (4.5) All values severely defectiveNeutral story (with cueing) 3* / 0 (2) / 0 (1)Taylor figure recall:

immediate/delayed28 / 23 Average

Below average=<25th percentile; defective=<5th percentile; severely defective=<1st percentile.*After immediate recall the neutral story had been presented for a second learning trial which yielded an immediate recall score of5.

Basal forebrain amnesia 165



j.com/

J Neurol N


ugust 1999. Dow

nloaded from


WAIS-R subtest information and by theabsence of significant naming diYculties.However, he did have some word finding diY-culties for unfamiliar items (see above) andcategorical fluency42 was mildly impaired withthe exception of musical instruments (animals:13/1 min; birds: 4/1 min; household items: 15/1min; musical instruments: 12/1 min). Drawingfamiliar objects from memory was good. Wecould not assess his professional knowledge buthe credibly assured us that it had becomeseverely defective. On a famous faces testwhich presents faces of 15 famous personsfrom each decade and asks for the famous per-sons’ names and for semantic informationabout them his scores were at the 5th percentileof age matched controls for persons famousbetween 1966 to 1975, below the 1st percentilefor 1976 to 1985, and between the 5th and25th percentile for 1945 to 1965 and after1986. In this test he never produced a wrongname and only once wrong semantic infor-mation (“American Secretary for ForeignAVairs” for Winston Churchill). All othererrors were constituted by “don’t know”responses.

Autobiographical memoryThe patient complained of an impoverishmentof his entire autobiographical memory. On theautobiographical memory interview43 knowl-edge about autobiographical facts was good forchildhood and for recent time (17.5 and 18 outof 21) but defective for early adulthood(11/21). Recall of autobiographical incidentswas borderline to abnormal for all time periodsand showed a similar time gradient with themost severe loss for early adulthood (child-hood: 6/9; early adulthood: 1/9; recent time:4/9). Recall of autobiographical episodes wasfurther explored with a modified Crovitztechnique.44 He was presented with 60 cuewords (for example, train, friendship, hostility,swim, laugh, destroy). He was asked to producean autobiographical episode related to eachword and to say from which period of his life itstems. Episodes were rated for specifity andrichness. He produced only 35 episodes, whichis below average. In addition, he gave 12general statements without episodic character,and refused to produce more specific infor-mation when requested to do so. In parallel tothe famous faces test, production was poorestfor the period from 1976 to 1985. We could notcheck the veracity of those episodes which heproduced but they did seem to be plausiblytrue memories.

DiscussionThe patient had a severe anterograde memorydeficit for verbal material aVecting free recalland recognition. With both modes of testing asignificant source of errors was constituted byfalse positive responses and intrusions. How-ever, the patient confabulated neither on storyrecall nor in day to day memory. Retrogradememory was impoverished for both semanticand autobiographical information. On tests ofretrograde memory he virtually never pro-

duced wrong information but stated that he didnot know the response.

Primary memory for verbal material waswithin normal limits. Only when he had tomaintain and simultaneously process verbalmaterial his performance fell below normalvalues. He scored normally on tests ofexecutive functions. Verbal fluency was lowerthan graphic fluency, and there were someword finding problems for unfamiliar items.These language diYculties were, however, fartoo mild to plausibly explain his severe amnesiafor any verbal material.

The responsible lesion was unilaterally leftsided. It had its centre in the nucleusaccumbens but encroached on the inferior limbof the internal capsula and the ventral portionof the globus pallidus and nucleus caudatus.There was evidence of some atrophy of thehead of the caudate nucleus and of frontal andtemporal hypoperfusion.

Similar to patients with unilaterally left sidedmedial temporal or thalamic lesions4 6 45 andtwo other patients with strictly unilateral leftsided basal forebrain lesions15 21 he had amnesiaonly for verbal material and no significant defi-cits of general intelligence, attention, or execu-tive function. This constellation is unlikely tobe a manifestation of diVuse brain damage.23

We therefore feel justified to interprete hisamnesia as a sequel of the circumscribed basalforebrain lesion.

In addition to the anterograde memory defi-cit, he had diYculties recalling premorbidlyacquired autobiographical and semantic infor-mation. As these were tested mainly verbally, itis not clear whether the deficit aVected visualmemory as well. Drawing of objects frommemory was normal, but the husband and hiswife reported some diYculties with retrieval oftopographical memories. Recall of autobio-graphical episodes and of famous persons waspoorest for the years from 1976 to 1985 whenhe was between 35 and 45 years old. Presum-ably, the poor result for “early adulthood” inthe autobiographical memory interview in-cluded the same period. This gap may be anenhancement of the paucity of memories fromthis period of life which has been documentedin normal subjects older than 50 years.46

If anterograde and retrograde amnesia had acommon cause, this would be most likely to bea defect of retrieval rather than of memoryconsolidation or storage. There are, however,other patients with basal forebrain amnesiain whom premorbidly acquired auto-biographical20 21 and semantic20 memory werenormal. Apparently the combination of an-terograde and retrograde memory loss is notan invariable feature of basal forebrain amne-sia. We think, however, that the published dataon retrograde memory loss in basal forebrainamnesia are too scarce too permit any definiteconclusions regarding their underlying mecha-nisms. Further discussion will concentrate onthe anterograde verbal memory deficit.

The absence of a general weakness of execu-tive function does not rule out the possiblity ofa specific weakness of executive control of ver-bal memory. We discuss this possibility, and




j.com/

J Neurol N


ugust 1999. Dow

nloaded from


then the location of the critical lesion withinthe basal forebrain.

FALSE RESPONSES AND STRATEGIC CONTROL OF

MEMORY RETRIEVAL

The patient produced numerous intrusions inrecall of word lists, and the inaccuracy of hisrecognition memory for words was entirely dueto the high number of false positive responses.At the same time, he confabulated neither onstory recall, nor in recall from remote semanticand autobiographical memory, nor in daily life.False responses were thus restricted to memoryfor word lists. For these words, the decisionwhether a word coming to mind during recallwas correct or not could be based on nothingelse but on memory of this particular word’soccurrence in the word list. There was no wayto infer the plausibility of the decision frompartial recall of the memorised material orfrom preserved remote memory. By contrast,tests which did not provoke false responsesasked either for memory of organised verbalinformation (story recall, autobiographical epi-sodes), or for recall of names and informationwith a defining relation to pictures presentedduring recall (famous persons). In these tests,he could exploit partial recall of the memorisedmaterial or preserved remote memory forjudging the plausibility of information comingto mind.

On this account, supervisory control of recallwas intact. He used available knowledge forcontrolling the plausibility of informationcoming to mind. The source of false responsesmight be sought in a general enhancement of“feeling of familiarity”24 which levelled the sig-nal to noise ratio between memorised itemsand items which were similar but had not beenpresented. Feelings of familiarity were assignedindistinctly to any word that came to mindduring recall and misled his responses when hehad no external evidence for checking thecredibility of this feeling. False responses mayalso have resulted from a strategic decision atthe level of supervisory control. He may haveopted for a liberal response bias whenever hehad no resources for estimating the plausiblityof responses. In any case, intrusions and falsepositive responses do not indicate a failure ofsupervisory control of memory.

IS THE NUCLEUS ACCUMBENS A CRITICAL

STRUCTURE FOR HUMAN MEMORY?The basal forebrain consists of three function-ally distinct compartments47: the corticopetal,mainly cholinergic, system, the extended amyg-dalae, and the ventral striatopallidal system.Cholinergic cells are concentrated in theseptum, the diagonal band, and the nucleusbasalis.48 The extended amygdalae stretch fromthe centromedial nuclei of the amygdalaemedially and rostrally through the substantiainnominata into the medial portion of thenucleus accumbens where they join the ventralstriatopallidal system. The ventral striatopalli-dal system is constituted rostrally by thenucleus accumbens and caudally by subcom-missural continuations of putamen and globuspallidus. The nucleus accumbens occupies a

central position in the functional interplaybetween these components of the basal fore-brain and between the basal forebrain andother brain regions. It constitutes a junctionbetween extended amygdalae and the ventralstriatopallidal system and it has eVerentconnections to the corticopetal system. It istherefore able to modulate cholinergicoutput.47 48 The nucleus accumbens sendseVerents to the hypothalamus, substantianigra, autonomic brainstam nuclei, and thepallidum, which provides a connection to thedorsomedial nucleus of the thalamus.49 Itreceives aVerent projections from dopaminer-gic midbrain nuclei and intralaminar thalamicnuclei, from the hippocampus and basolateralamygdalae, and from the prefrontal, insular,and temporal association cortex.47 50–52 It is thusin a position to integrate inputs from multiplecortical and subcortical areas including thehippocampus and the amygdalae, and to exertmodulatory influences on widespread corticalfunction.

Animal studies have suggested a role of thenucleus accumbens in memory. Rats withselective lesions of the nucleus accumbens havebeen found to fail on tests such as the Morriswater maze or delayed win-shift foraging,which depend on mnemonic function andwhich are known to be sensitive to hippocam-pal lesions.51 53–55 It has not, however, been set-tled whether these deficits are reflections ofmnemonic dysfunction or of other behaviouraldisturbances.50 56

Published studies of basal forebrain amnesialend credibility to a role of the nucleusaccumbens in human memory. In a groupstudy of patients with operated anteriorcommunicating artery aneurysms, Irle et al17

found memory disturbances only in patientswith “combined basal forebrain-striatum” le-sions. The nucleus accumbens forms part ofthe ventral striatum. Nucleus accumbenslesions were visible on MRI or CT in severalpublished cases of basal forebrainamnesia.12–14 18 20 In a further patient15 the lesionaVected the ventral pallidum and in two otherpatients MRI showed atrophy of the caudatenucleus.21 22 It is notable that in some of thesecases,13 15 21 22 as well as in our patient, thenucleus basalis, septum, and diagonal band,which have the greatest concentration of corti-copetal cholingergic cells, seem to be unaf-fected.

In summary, there seems to be ratherconvincing evidence that aVection of parts ofthe striatopallidal system makes an importantcontribution to basal forebrain amnesia.17 Theevidence for the nucleus accumbens as criticalstructure is less unequivocal, but does seem todeserve further studies in animals andhumans.51

Because the patient’s lesion encroached onthe anterior limb of the internal capsula ourfinding would be compatible with the hypoth-esis that lesions of fibre tracts rather than cellbodies are responsible for amnesia from extra-hippocampal lesions.15 27 In particular, the cap-sular lesion may have damaged the anteriorthalamic peduncle which connects the dorso-

Basal forebrain amnesia 167



j.com/

J Neurol N


ugust 1999. Dow

nloaded from


medial nucleus of the thalamus to the prefron-tal cortex. It should be noted, however, that theinferior thalamic peduncle which connects thebasolateral amygdalae with the dorsomedialnucleus, runs caudally from the anteriorcommissure and seemed not to be aVected.Interruption of fibres between the prefrontalcortex and thalamus may disconnect strategicprocesses supported by the frontal lobes frommemory functions depending on thethalamus.57 However, as we have argued in theprevious section, strategic control of memoryprocesses seemed to be well preserved in ourpatient.

1 Zola-Morgan S, Squire LR. Neuroanatomy of memory. AnnRev Neurosci 1993;16:547–63.

2 Zola-Morgan S, Squire LR, Amaral DG. Human amnesiaand the medial temporal region: enduring memory impair-ment following a bilateral lesion limited to field CA 1 of thehippocampus. J Neurosci 1986;6:2950–67.

3 Corkin S, Amaral DG, Gonzalez RG, et al. HM’s medialtemporal lobe lesion: findings from magnetic resonanceimaging. J Neurosci 1997;17:3964–79.

4 Goldenberg G, Wimmer A, Maly J. Amnesic syndrome witha unilateral thalamic lesion: a case report. J Neurol1983;229:79–86.

5 Von Cramon DY, Hebel N, Schuri U. A contribution to theanatomical basis of thalamic amnesia. Brain 1985;108:993–1008.

6 Clarke S, Assal G, Bogousslavsky J, et al. Pure amnesia afterunilateral left polar thalamic infarct: topographic andsequential neuropsychological and metabolic (PET) corre-lations. J Neurol Neurosurg Psychiatry 1994;57:27–34.

7 Parkin AJ, Rees JE, Hunkin NM, et al. Impairment ofmemory following discrete thalamic infarction. Neuropsy-chologia 1994;32:39–52.

8 Whithehouse PJ, Price DL, Clark AW, et al. Alzheimerdisease: evidence for selective loss of cholinergic neurons inthe nucleus basalis. Ann Neurol 1981;10:122–6.

9 Coyle JM, Price DL, DeLong MR. Alzheimer’s disease: adisorder of cortical cholinergic innervation. Science 1983;219:1184–90.

10 Braak H, Braak E. Neuropathological staging of Alzheimer-related changes. Acta Neuropathol 1991;82:239–59.

11 Fibiger HC. Cholinergic mechanisms in learning, memoryand dementia: a review of recent evidence. Trends Neurosci1991;14:220–3.

12 Damasio AR, GraV-Radford NR, Eslinger PJ, et al. Amnesiafollowing basal forebrain lesions. Arch Neurol 1985;42:263–71.

13 Phillips S, Sangalang V, Sterns G. Basal forebrain infarction:a clinicopathological correlation. Arch Neurol 1987;44:1134–8.

14 Fukamachi A, Horikoshi T, Nagaseki Y, et al. Symmetricalbilateral low-density lesions in the area of supply by Heub-ner’s arteries after aneurysm surgery. Acta Neurochir (Wien)1987;84:89–92.

15 Markowitsch HJ, Von Cramon DY, Hofmann E, et al. Verbalmemory deterioration after unilateral infarct of the internalcapsule in an adolescent. Cortex 1990;26:597–609.

16 De Luca J. Cognitive dysfunction after aneurysm of theanterior communicating artery. J Clin Exp Neuropsychol1992;14:924–34.

17 Irle E, Wowra B, Kunert HJ, et al. Memory disturbances fol-lowing anterior communicating artery rupture. Ann Neurol1992;31:473–80.

18 Morris MK, Bowers D, Chatterjee A, et al. Amnesia follow-ing a discrete basal forebrain lesion. Brain 1992;115:1827–47.

19 Bondi MW, Kaszniak AW, Rapcsak SZ. Implicit and explicitmemory following anterior communicating artery aneu-rysm rupture. Brain Cogn 1993;22:213–29.

20 Von Cramon DY, Markowitsch HJ, Schuri U. The possiblecontribution of the septal region to memory. Neuropsycholo-gia 1993;31:1159–80.

21 Hanley JR, Davies ADM, Downes JJ, et al. Impaired recall ofverbal material following rupture and repair of an anteriorcommunicating artery aneurysm. Cogn Neuropsychol 1994;11:543–78.

22 Weniger G, Markowitsch HJ, Irle E. Anterograde and retro-grade mnemonic deficits after unilateral damage of neostri-atal, ventral striatal and basal forebrain structures.Neurocase 1995;1:231–8.

23 Richardson JTE. Performance in free recall followingrupture and repair of intracranial aneurysms. Brain Cogn1989;10:210–26.

24 Delbercq-Derouesné J, Beauvois MF, Shallice T. Preservedrecall versus impaired recognition. Brain 1990;113:1054–74.

25 Dalla Barba G, Cappelletti JY, Signorini M, et al.Confabulation: Remembering another past, planninganother future. Neurocase 1997;3:425–36.

26 Fischer RS, Alexander MP, D’Esposito M, et al. Neuro-psychological and neuroanatomical correlates of confabu-lation. J Clin Exp Neuropsychol 1995;17:20–8.

27 Markowitsch HJ. Diencephalic amnesia: a reorientationtowards tracts? Brain Res Rev 1988;13:351–70.

28 Zimmermann P, Fimm B. Testbatterie zur Aufmerksamkeit-sprüfung (TAP). Freiburg: Psychologische Testsysteme,1993.

29 Wechsler D. The Wechsler memory scale - revised. San Diego:Psychological Corporation, 1987.

30 Snodgrass JG, Vanderwart M. A standardized set of 260pictures: norms for name agreement, image agreement,familiarity, and visual complexity. J Exp Psychol Learn MemCogn 1980;6:174–215.

31 Genzel G, KerkhoV G, ScheVter S. PC-gestützte Standard-isierung des Bildmaterials von Snodgrass und Vanderwart(1980). Neurolinguistik 1995;41–53.

32 Lehrl S. Mehrfachwahl - Wortschatz - Intelligenztest. MWT-B.Erlangen: Verlag D Straube, 1977.

33 Tewer U. HAWIE-R: Hamburg-Wechsler-Intelligenztest fürErwachsene. Bern: Huber, 1991.

34 Nelson HE. A modified card sorting test sensitive to frontallobe defects. Cortex 1976;12:313–24.

35 Shallice T. Specific impairments of planning. Phil Trans RSoc Lond B Biol Sci 1982;298:199–209.

36 Shallice T, Burgess PW. Deficits in strategy application fol-lowing frontal lobe damage in man. Brain 1991;114:727–41.

37 Regard M, Strauss E, Knapp P. Children’s production onverbal and non-verbal fluency tasks. Percept Mot Skills1982;55:839–44.

38 Daneman M, Carpenter PA. Individual diVerences in work-ing memory and reading. Journal of Verbal Learning andVerbal Behaviour 1980;19:450–66.

39 Wilson B, Cockburn J, Baddeley A. The Rivermeadbehavioural memory test. Reading: Thames Valley TestCompany, 1985.

40 Metzler P, Voshage J, Rösler P. Berliner Amnesietest (BAT).Göttingen: Hogrefe, 1992.

41 Ilmberger J. German version of the California verbal learningtests. München: Institut für Medizinische Psychologie derUniversität, 1988:

42 Hodges JR, Patterson K. Is semantic memory consistentlyimpaired early in the course of Alzheimer’s disease? Neuro-anatomical and diagnostic implications. Neuropsychologia1995;33:441–60.

43 Kopelman MD, Wilson BA, Baddeley AD. The autobio-graphical memory interview: a new assessment of autobio-graphical and personal semantic memory in amnesicpatients. J Clin Exp Neuropsychol 1989;11:724–44.

44 Robinson JA. Sampling autobiographical memory. CognPsychol 1976;8:578–95.

45 Milner B. Interhemispheric diVerences in the localization ofpsychological processes in man. Br Med Bull 1971;27:272–7.

46 Rubin DC, Wetzler SE, Nebes RD. Autobiographicalmemory across the lifespan. In: Rubin DC, ed. Autobio-graphical memory. Cambridge: Cambridge UniversityPress, 1986:202–24.

47 Alheid GF, Heimer L. New perspectives in basal forebrainorganization of special relevance for neuropsychiatricdisorders: the striatopallidal, amygdaloid, and corticopetalcomponents of substantia innominata. Neuroscience 1988;27:1–39.

48 Mesulam MM, Mulson EJ, Levey A, et al. Cholinergicinnervation of cortex by the basal forebrain: cytochemistryand cortical connections of the septal area, diagonal bandnuclei, nucleus basalis (substantia innominata) and hy-pothalamus in the rhesus monkey. J Comp Neurol1986;214:170–97.

49 Alexander GE, DeLong MR, Strick PL. Parallel organiza-tion of functionally segregated circuits linking basal gangliaand cortex. Ann Rev Neurosci 1986;9:357–81.

50 Floresco SB, Seamans JK, Phillips AG. DiVerential eVectsof lidocaine infusions into the ventral CA1/subiculum orthe nucleus accumbens on the acquisition and retention ofspatial information. Behav Brain Res 1996;81:163–71.

51 Setlow B. The nucleus accumbens and learning andmemory. J Neurosci Res 1997;49:515–21.

52 Redish AD, Touretzky DS. Cognitive maps beyond the hip-pocampus. Hippocampus 1997;7:15–35.

53 Annett LE, McGregor A, Robbins TW. The eVects ofibotenic acid lesions of the nucleus accumbens on spatiallearning and extinction in the rat. Brain Res 1989;31:231–42.

54 Sutherland RJ, Rodriguez AJ. The role of the fornix/fimbriaand some related subcortical structures in place learningand memory. Behav Brain Res 1989;32:265–77.

55 Seamans JK, Phillips AG. Selective memory impairmentsproduced by transient lidocaine-induced lesions of thenucleus accumbens in rats. Behav Neurosci 1994;108:456–68.

56 Gal G, Joel D, Gusak O, et al. The eVects of electrolyticlesion to the shell subterritory of the nucleus accumbens ondelayed non-matching-to-sample and four-arm baitedeight-art radial-maze tasks. Behav Neurosci 1997;111:92–103.

57 Warrington EK, Weiskrantz L. Amnesia: a disconnectionsyndrome? Neuropsychologia 1982;20:233–48.




j.com/

J Neurol N


ugust 1999. Dow

nloaded from


LETTERS TOTHE EDITOR

Magnetic resonance imaging andvertebral artery dissection

Since the advent of advanced radiologicalmodalities such as MRI and magneticresonance angiography (MRA), dissectionsof cervical arteries are increasingly recog-nised as a common cause of stroke in youngadults. Auer et al1 recently advocated MRA asthe initial diagnostic tool for vertebral arterydissection. Conventional angiography mightbe avoided altogether in subjects with asuspicious history and MRA images sugges-tive of a dissection (double lumen or muralhaematoma).1 The sensitivity of MRA for thediagnosis of vertebral artery dissection was

only 20% in one study, but the specificity wasexcellent (100%).2 The sensitivity was con-siderably better in the hands of Auer et al,1

but in this study the specificity (true negativerate in subjects free of disease) was notconsidered because all patients had vertebralartery dissection. The following case reportillustrates that care must be taken to avoidfalse positive results when using MRA for thediagnosis of vertebral artery dissection.

A 47 year old male pilot suddenly experi-enced clumsiness and slight loss of strength inthe right arm and leg during a long distanceflight, while he stooped forward. During thefollowing hours, he developed a global head-ache without irradiation to the neck, but theother symptoms gradually diminished. Priorhistory was unremarkable, except for a 3 hourperiod of horizontal diplopia which suddenlydeveloped 3 months earlier. He had neversmoked. Family history was negative forcardiovascular disorders. The patient laterconfessed that he had recently picked up thehabit of gargling his throat with toothpaste

twice a day, always with his neck in extremeretroflexion.

General physical examination (8 hoursafter onset of symptoms) was normal. Neuro-logical examination showed minimal paresisand impaired dexterity of the right hand, mildcircumduction of the right leg, and aninsecure tandem gait. An MRI (including T1weighted spin echo images with and withoutfat suppression, and proton density and T2weighted fast spin echo sequences, performedon a 1.5 Tesla whole body MRI system) per-formed several hours later visualised both afresh and an old right sided cerebellar infarct(figure A). In addition, MRI showed anirregular right vertebral artery in which a pat-ent lumen was partially surrounded by asemilunar area of high signal intensity on T1and T2 weighted images. On fat suppressedimages, this area’s high signal intensitypersisted, excluding the possibility that itoriginated from perivascular fat. This imagewas suggestive of mural haematoma due tovertebral dissection (figure B). Because wewere reluctant to base any treatment deci-sions (anticoagulants) merely on MRI find-ings, digital subtraction angiography was per-formed on the day of admission. Thisexamination was normal (figure C). Shortlyafter this procedure, the patient developedvertigo and nystagmus which disappearedafter 3 hours. Because we were puzzled by thediscrepant findings on conventional angio-graphy and MRI, we performed an MRA 4days later. At this examination, the semilunararea of high signal intensity was found again(figure D), despite saturation of craniofugaland craniopetal flow respectively, which wasapplied to exclude the possibility that thehigh signal originated from flow in theperiarterial venous plexus. Therefore, thisexamination was again suggestive of rightvertebral artery dissection. An extensivesearch for other causes of stroke showed noabnormalities. Hence, due to the continuingdiscrepancy between conventional angio-graphy and MRI/MRA, and due to theabsence of any other cause of stroke, no cer-tain diagnosis could be established.

In this patient, a diagnosis of right vertebralartery dissection was initially made given theclinical course with repeated episodes ofischaemia restricted to the vertebrobasilarsystem, as well as the suggestive MRIfindings.1 We speculated that habitual gar-gling was a potential underlying cause, asneck retroflexion can cause cervical dissec-tions. However, we had to reject this diagno-sis in view of the normal conventional angio-graphy, which remains the gold standard fordiagnosing cervical artery dissection.3 In oneseries,2 conventional angiography was neverfalsely negative in patients with clinical signsor symptoms of vertebral artery dissection.The possibility that conventional angio-graphy had nevertheless yielded a false nega-tive result seems highly unlikely. In dissectedarteries, MRI/MRA can detect intimal flaps,mural haematomas, or aneurysmal dilata-tions that are sometimes missed by conven-tional angiography, but even in such patientsconventional angiography is never completelynormal in the acute stage. Follow upexaminations of patients with proven verte-bral artery dissection indicate that theappearance of a dissected artery on conven-tional angiography can normalise in a sub-stantial proportion of patients, but alwaysafter an interval of at least 1 to (usually) sev-eral weeks.1 Conventional angiography in ourpatient was performed on the day of admis-

(A) T2 weighted fast spin echo image showing high signal intensity in the right cerebellar hemisphere,indicative for a recent infarct. The older infarct cannot be seen on this section. (B) Axial T1 weightedfast spin echo image with fat saturation at the level of the base of the tongue, showing a semilunararea with high signal intensity around the flow void in the right vertebral artery. (C) Selectivecontrast injection in the right vertebral artery shows no abnormalities. The remainder of theintra-arterial angiography of the cervical and cranial arteries was also normal. (D) Axial threedimensional time of flight technique, acquired in the axial plane image at the same level showing highsignal intensity at the same location as in B.

J Neurol Neurosurg Psychiatry 1999;67:691–701 691

sion, directly after the “abnormal“ MRI andfour days prior to the “abnormal“ MRA,hence spontaneous resolution of the dissec-tion is very unlikely. Therefore, we considerour MRI/MRA examinations falsely positive,and we hypothesise that the area of semilunarhigh signal intensity originated from aperivascular venous plexus, in which we wereunable to saturate inflow of blood completely,presumably due to extremely slow flow.

Our “pilot study“ illustrates the specificityproblems of MRI/MRA for the diagnosis ofvertebral artery dissection. Two anatomicalstructures surrounding vertebral arteries con-tribute to these problems. The first structureis the venous plexus that surrounds vertebralarteries. This structure may have a semilunarappearance, and slow flow in its lumen maygive rise to high signal intensity on both MRIand MRA, creating an image suggestive ofdissection.4 5 It has been suggested that satu-ration slabs in conjunction with MRAcompletely suppress flow related high signal,thus distinguishing it from high signal froman intramural haematoma which cannot besuppressed by saturation slabs.4 5 The presentcase report illustrates that flow in this plexuscannot always be suppressed.

The second tissue that may falsely presentas a dissection is fat that directly surroundsvertebral arteries. This fat also gives rise tohigh signal intensity, but using fat suppres-sion techniques it can be readily diVerenti-ated from intramural haematoma. Further-more, the usual diameter asymmetry ofvertebral arteries, turbulence and magneticsusceptibility near sharp vessel turns can alsocause false positive MRA results.2 In somepatients, MRI cannot distinguish betweenintraluminal thrombus and intramural hae-matoma, leading to false conclusions.

Decisions based on false positive MRI/MRA results can be hazardous due to thesometimes severe side eVects of anticoagu-lants, the treatment that is recommended bysome to prevent further ischaemic events.Another danger of a false positive diagnosis ofvertebral dissection is that it may preclude thesearch for other causes of stroke that could beamenable to secondary prevention.

MRI/MRA remains important because ithelps visualise ischaemic lesions and, in somepatients, provides complementary morpho-logical information to cerebral angiography.1

Furthermore, it is a non-invasive procedure,an important advantage over cerebral angio-graphy which carries a morbidity and mor-tality risk. Our patient, who developedtransient neurological deficits shortly afterangiography, underscores this. Therefore,MRA can play a part in the diagnosis of ver-tebral artery dissection, provided that the pit-falls mentioned above are recognised to avoidfalse positive results. In case of doubt,cerebral angiography remains the gold stand-ard for vertebral artery dissection.

B R BLOEMG J LAMMERS

Department of Neurology

M A VAN BUCHEMDepartment of Radiology, Leiden University Medical

Centre, The Netherlands

Correspondence to: Dr Bastiaan R Bloem, Depart-ment of Neurology, Leiden University MedicalCentre, PO Box 9600, 2300 RC Leiden, The Neth-erlands. Telephone 0031 71 5262134; fax 0031 715248253; email [email protected]

1 Auer A, Felber S, Schmidauer C, et al. Magneticresonance angiographic and clinical features ofextracranial vertebral artery dissection. J Neu-rol Neurosurg Psychiatry 1998;64:474–81.

2 Levy C, Laissy JP, Raveau V, et al. Carotid andvertebral artery dissections: three-dimensionaltime-of-flight MR angiography and MR imag-ing versus conventional angiography. Radiology1994;190:97–103.

3 Hart RG. Vertebral artery dissection. Neurology1988;38:987–9.

4 Miaux Y, Cognard C, Martin-Duverneuil N, etal. Flow related enhancement in the vertebralplexus mimicking an intramural hematoma.Am J Neuroradiol 1996;17:191–2.

5 Dumas JL, Stanescu R, Goldlust D, et al. Verte-bral vein imaging with MR angiography. Am JNeuroradiol 1997;18:1190–2.

Catatonia due to central pontine andextrapontine myelinolysis: case report

Central pontine and extrapontine myelinoly-sis (CPEM) are recognised complications ofhyponatraemia and its overly rapidcorrection.1 CPEM usually presents withspastic tetraparesis and pseudobulbar palsy.1

We describe a patient with CPEM in whombehavioural manifestations overshadowedcorticospinal tract signs.

A 64 year old Chinese speaking womanwith a history of episodic psychoticdepression that had never required admis-sion to hospital was admitted to a hospitalbecause of vomiting and diarrhoea. Her gen-eral and neurological examination were nor-mal. On admission she had a sodiumconcentration of 105 meq /l. An infusion of3% saline at a rate of 150 ml/ hour was givenduring 6 hours. Ten hours later her sodiumwas 134 meq/l and she was mute andtetraparetic. She seemed catatonic withmotor perseveration. Transfer to our hospitalwas requested.

On admission her vital signs were normal.She was mute without any spontaneous voli-tional movements except for visual pursuit.She was tetraparetic and hyperreflexic withincreased tone and bilateral Babinski’s signs.CPEM was suspected. Admission MRI,EEG, and spinal fluid examination were nor-mal. Over the next 2 days the reflexesnormalised and the Babinski’s signs disap-peared but she continued to have mild diffuseweakness. She had waxy flexibility andassumed bizarre non-physiological posturesconsistent with catatonia. Psychogenic unre-sponsiveness was suspected and she wasstarted on risperidone and sertraline. There

was no benefit. Electroconvulsive therapy wasproposed by a psychiatry consultant but wasrefused by the patient’s family. The clinicalpicture was dominated by an akinetic mutismwith marked catatonia. Catatonia due toCPEM was considered. A repeat MRI 12days after the onset of symptoms showed highintensity areas in the pons, caudate, andputamen consistent with CPEM (figure A,B). Physical and occupational therapy wereinstituted and she gradually recovered overthe next 2 weeks. She was transferred to arehabilitation hospital where she recoveredcompletely and returned to live independ-ently. She has been followed up at the neurol-ogy clinic and has not shown any residualdeficits.

CPEM usually presents with tetraparesisand pseudobulbar palsy. Unusual clinicalpresentations include extrapyramidal syn-dromes, ataxia, and neurobehavioural syn-dromes. Although psychiatric manifestationsof CPEM have been recognised they usuallymanifest as an agitated delirium, or apseudobulbar state with pathological laugh-ing and crying.1 When present, neuropsychi-atric symptoms are usually overshadowed byflorid signs of brainstem and pyramidal tractdysfunction.2 3 Behavioural changes such asinappropriate aVect, emotional lability, per-sonality changes, paranoia, poor judgement,emotional incontinence, and disinhibitionhave been reported.1 2 Price and Mesulamdescribed a case of pontine myelinolysis inwhich transient pyramidal signs were fol-lowed by confusion, restless behaviour, pres-sured tangential speech, and disinhibition.2

Our patient also had transient long tract signsbut they were followed by a catatonic state.The extensive extrapontine myelinolysispresent in our patient may explain the behav-ioural symptoms we encountered.

CPEM may present with unusual behav-ioural symptoms. At the onset of neurologicaldeterioration MRI may be normal but subse-quent imaging studies usually disclose thelesions. CPEM presenting with neuropsychi-atric symptoms in patients with normal initialimaging studies might suggest a psychogenicaetiology. Corticospinal tract signs may betemporary. A strong index of suspicion forCPEM is required when patients with recent

(A) Axial T2 weighted image showing prominent high signal intensity within the pons suggestive ofcentral pontine myelinolysis. (B) Axial T2 image showing symmetric bilateral areas of high signal inthe caudate and putamen suggestive of extrapontine myelinolysis.

692 Letters, Correspondence, Book reviews, Correction

hyponatraemia present with behaviouralchanges. Akinetic mutism and catatonia maybe the dominant clinical features in CPEM.

JULIO CHALELAJORGE KATTAH

Department of Neurology, Georgetown UniversityMedical Center, Washington DC, USA

Correspondence to: Dr Julio Chalela, 4000 Presi-dential Boulevard, Apartment 213, Philadelphia, PA19131, USA. Telephone 001 215 878 3311;email:[email protected]

1 Illowsky B, Laureno R. Pontine and extrapon-tine myelinolysis. Pontine and extrapontinemyelinolysis: a neurologic disorder followingrapid correction of hyponatremia. Medicine1993;72:359–73.

2 Price BH, Mesulam MM. Behavioral manifesta-tions of central pontine myelinolysis. Arch Neu-rol 1987;44:671–3.

3 Laureno R, Illowsky B. Myelinolysis aftercorrection of hyponatremia. Ann Intern Med1997;126:57–62.

Association between butyrylcholin-esterase K variant and the Alzheimertype neuropathological changes inapolipoprotein E å4 carriers older than75 years

Apolipoprotein E (ApoE) å4 has a stronginfluence on the development of sporadicAlzheimer’s disease in many ethnic popula-tions. However, ApoE å4 is neither necessarynor suYcient for the development ofAlzheimer’s disease, suggesting that othergenes increase the risk of Alzheimer’s disease.One such new candidate is the butyrylcho-linesterase (BChE) gene (BCHE).1 BChE isassociated with senile plaques (SPs) and neu-rofibrillary tangles (NFTs). Lehmann et alrecently reported that the K variant of BCHE(BCHE-K) was associated with the develop-ment of Alzheimer’s disease, especially inApoE å4 carriers older than 75 years.1 A pos-sible mechanism as to how BCHE-K isrelated to Alzheimer’s disease under theinfluence of ApoE å4 is the acceleration ofAlzheimer type neuropathological changes. IfBCHE-K has an eVect on the development ofAlzheimer’s disease in ApoE å4 carriers, theformation of Alzheimer type neuropathologi-cal changes may be accelerated by BCHE-Kin the ApoE å4 carriers.

We have examined genotypes of BCHE andApoE, and densities of the senile plaques(SPs), with dystrophic neurites (NPs), andneurofibrillary tangles NFTs in the brainsfrom 51 patients with Alzheimer’s disease and90 non-demented subjects from a postmor-tem series of Japanese. Clinical and postmor-tem diagnosis of Alzheimer’s disease was car-ried out as described previously.2 Thedensities of Alzheimer type neuropathologi-cal changes were quantified by averaging the

counts of those in the hippocampus andsuperior temporal gyrus. Genotypes of BCHEand ApoE in all patients were determined asdescribed elsewhere.1 2 Genotypic and allelicdistributions of BCHE were analysed by ÷2

test. The densities of the SPs, NPs, andNFTs, and ages at onset and durations of ill-ness were compared among BCHE genotypeswith the Kruskal-Wallis test or Mann-Whitney U test in total subjects, those withAlzheimer’s disease, and non-demented sub-jects. We also examined these relations in thesubgroups divided by the ApoE å4 status orthe age of 75 years. Statistical significance wasdefined as two tailed probabilities of <0.05.

There were no significant diVerences in thefrequency of BCHE-K genotypes or allelesbetween patients with Alzheimer’s disease(0.16 in allele frequency) and non-dementedsubjects (0.18), and in the total subjects,ApoE å4 carriers or non-ApoE å4 carriers,although a strong association of ApoE å4alleles with Alzheimer’s disease was found inthis population (p=0.004). Genetic associ-ation of BCHE-K genotypes with sporadicAlzheimer’s disease was non-significant in allsubjects older than 75 years, the ApoE å4carriers older than 75 years, and non-ApoEå4 carriers older than 75 years. There was nogenetic association of BCHE-K with the den-sities of the SPs, NPs, or NFTs in thehippocampus and superior temporal gyrus inthe total subjects, in the Alzheimer’s diseaseor non-demented groups, or with ages atonset or duration of illness in Alzheimer’sdisease. However, when we divided total sub-jects into two subgroups with diVerent ApoEå4 status, there was significant associationbetween BCHE-K and the density of the SPsand NPs in the superior temporalgyrus(STG) in the ApoE å4 carriers (SPs,p=0.04; NPs, p=0.03, data not shown). Fur-ther, we analysed the correlation betweenBCHE-K and the densities of the SPs, NPs,and NFTs in the hippocampus and superiortemporal gyrus in the ApoE å4 carriers olderthan 75 years and non-ApoE å4 carriers olderthan 75 years (table). There was a significantgenetic association of BCHE-K with the den-sities of the SPs, NPs, and NFTs in the STGin the ApoE å4 carriers older than 75 years.There was a decrease of severity of Alzheimertype neuropathological changes with BCHE-K. A similar trend was seen in the hippocam-pus though this did not reach significance.

Our results showed that BCHE-K mighthave no eVect on the development of sporadicAlzheimer’s disease even in the ApoE å4 car-riers or subjects older than 75 years. By con-trast with a significant genetic association inpatients confirmed at postmortem in theBritish population,1 there was no correlationin the Japanese population. Although our

sample size was small, there were not eventrends for a positive association in our study,suggesting that the lack of association was notdue to small sample size. The frequency ofBCHE-K in our Japanese control populationwas 0.18. This was not significantly diVerentfrom that in the British population examinedby Russ et al (0.20).3 However, the frequencyof BCHE-K in the British control populationreported by Lehmann et al was 0.09, whichwas significantly lower than our results(p=0.04).1 These findings indicate that thefrequency of BCHE-K and its genetic linkagewith the development of Alzheimer’s diseasewould be diVerent among sample popula-tions.

Our neuropathological study disclosed asignificant association of BCHE-K withAlzheimer type neuropathological changes inthe ApoE å4 carriers older than 75 years, butnot in the non-ApoE å4 carriers. Lehmann etal showed that BCHE-K was strongly associ-ated with the development of Alzheimer’sdisease in the ApoE å4 carriers older than 75years.1 Analyses of the same subgroup ofApoE å4 carriers older than 75 yearsincreased statistical significance in both ourstudies and that of Lehmann et al. Thissuggests that BCHE-K as a genetic marker islinked with formation of Alzheimer type neu-ropathological changes or development ofAlzheimer’s disease in the ApoE å4 carriersolder than 75 years. However, a decrease ofthe severity of Alzheimer type neuropatho-logical changes with BCHE-K in our studywas not expected because Lehmann et alshowed an increase in frequency of theBCHE-K allele in Alzheimer’s disease.1 Sin-gleton et al also reported that BCHE-K wasnot associated with the densities of the SPsand NFTs, even in the ApoE å4 carriers.4 Inaddition, BCHE-K was not related to thedevelopment of Alzheimer’s disease in theApoE å4 carriers in our study. Russ et al3 andSingleton et al4 also showed a lack ofassociation between BCHE-K and the devel-opment of Alzheimer’s disease. However,Hiltunen et al showed that BCHE-K had aprotective eVect on the development ofAlzheimer’s disease in ApoE å4 carriersyounger than 75 years.5 The eVects ofBCHE-K on the Alzheimer type neuropatho-logical changes or development ofAlzheimer’s disease are diVerent amongstudies, suggesting that the significant geneticassociation in the studies by Lehmann et al,1

Hiltunen et al,5 and ourselves might belinkage disequilibrium with relevant variabil-ity in BCHE or other adlacent gene on chro-mosome 3, and that BCHE-K does not play adirect part in the pathogenesis of Alzheimer’sdisease.

BCHE genotypes and the densities of the SPs, NPs and NFTs in the hippocampus and superior temporal gyrus in ApoE å4 carriers older than 75 yearsand non-ApoE å4 carriers older than 75 years

BCHE genotype

ApoE å4 carriers over 75 years (n=28) non-ApoE å4 carriers over 75 years (n=95)

K/N (n=8) N/N (n=20) p K/K (n=4) K/N (n=25) N/N (n=66) p

Hippocampus:SPs 3.0 (0.0, 17.2) 12.2 (4.5, 28.7) 0.13 0.0 (0.0, 5.7) 0.0 (0.0, 16.7) 0.0 (0.0, 10.3) 0.69NPs 0.7 (0.0, 11.9) 11.0 (3.9, 25.3) 0.07 0.0 (0.0, 3.5) 0.0 (0.0, 14.0) 0.0 (0.0, 8.6) 0.63NFTs 1.1 (0.4, 23.1) 17.4 (2.5, 59.6) 0.12 3.9 (0.9, 10.0) 7.0 (0.6, 35.7) 4.6 (0.4, 15.0) 0.72

Superior temporal gyrus:SPs 0.2 (0.0, 22.8) 49.7 (12.1, 83.8) 0.007 0.0 (0.0, 58.0) 6.0 (0.0, 64.8) 1.2 (0.0, 44.0) 0.61NPs 0.2 (0.0, 8.8) 10.7 (3.6, 19.0) 0.02 0.0 (0.0, 3.7) 2.0 (0.0, 6.5) 0.4 (0.0, 7.8) 0.59NFTs 0.0 (0.0, 0.2) 0.9 (0.0, 4.9) 0.04 0.0 (0.0, 0.0) 0.0 (0.0, 0.5) 0.0 (0.0, 0.4) 0.32

Values are medians (25th percentile, 75th percentile). The density represents the average counts in 2.56 mm2 for the SPs and NPs, and in 0.64 mm2 for the NFTs.BCHE=butyrylcholinesterase gene; ApoE=apolipoprotein E; K=the K variant allele of butyrylcholinesterase gene; N = the normal allele of butyrylcholinesterase gene;SPs =senile plaques; NPs=senile plaques with dystrophic neurites; NFTs=neurofibrillary tangles.

Letters, Correspondence, Book reviews, Correction 693

We are grateful to I Isahai, M Takeda, H Konuma,and Y Miura for their expert technical assistance.The study was supported in part by a HealthScience Research Grant to MY from the Ministry ofHealth and Welfare, Japan and a Grant-in-Aid forScientific Research to MY from the Ministry ofEducation, Science, Sports and Culture, Japan.

N SODEYAMAM YAMADA

H MIZUSAWADepartment of Neurology

Y ITOHE OTOMO

Department of Internal Medicine, Tokyo Medical andDental University, Tokyo, Japan

N SUEMATSUDepartment of Pathology, Yokufukai Geriatric

Hospital, Tokyo, Japan

M MATSUSHITADepartment of Neuropathology, Tokyo Institute of

Psychiatry, Tokyo, Japan

Correspondence to: Dr Masahito Yamada, Depart-ment of Neurology, Tokyo Medical and DentalUniversity, Yushima 1–5–45, Bunkyo-ku, Tokyo113–8519, Japan. Telephone 0081 3 5803 5234; fax0081 3 5803 0169; email [email protected]

1 Lehmann DJ, Johnston C, Smith AD. Synergybetween the genes for butyrylcholinesterase Kvariant and apolipoprotein E4 in late-onsetconfirmed Alzheimer’s disease. Hum Mol Genet1997;6:1933–6.

2 Sodeyama N, Itoh Y, Suematsu N, et al.Presenilin 1 intronic polymorphism is not asso-ciated with Alzheimer type neuropathologicalchanges or sporadic Alzheimer’s disease. JNeurol Neurosurg Psychiatry 1998;64:548–51.

3 Russ C, Powell J, Lovestone S, et al. K variant ofbutyrylcholinesterase and late-onsetAlzheimer’s disease. Lancet 1998;351:881.

4 Singleton AB, Smith G, Gibson AM, et al. Noassociation between the K variant of thebutyrylcholinesterase gene and pathologicallyconfirmed Alzheimer’s disease. Hum Mol Genet1998;7:937–9.

5 Hiltunen M, Mannermaa A, Helisalmi S, et al.Butyrylcholinesterase K variant and apolipo-protein E4 genes do not act in synergy in Finn-ish late-onset Alzheimer’s disease patients.Neurosci Lett 1998;250:69–71.

Ideomotor prosodic apraxia

Prosody is a non-verbal or suprasegmentalfeature of language that conveys various levelsof information to the listener, includinglinguistic, aVective (attitudinal and emo-tional), dialectical, and idiosyncratic data.1

The acoustical features underlying prosodyinclude pitch, intonation, melody, cadence,loudness, timbre, tempo, stress, accent, andpauses.2 These acoustical features are typi-cally spared in patients with cortical demen-tias such as Alzheimer’s disease in whichtemporoparietal cortices are primarily af-fected. Patients with Alzheimer’s disease,however, often develop apraxia, which can bedefined as a disorder of skilled movement notcaused by weakness, akinesia, deaVerenta-tion, abnormal tone or posture, movementdisorders (such as tremor or chorea), intellec-tual deterioration, poor comprehension, oruncooperativeness.3 Moreover, subtypes ofapraxia have been delineated and are definedby the nature of errors made by the patientand the means by which these errors areelicited.4 5 Accordingly, a patient with prob-able dementia of the Alzheimer’s type isdescribed who had normal prosodic elementsto his spontaneous everyday speech, butcould not produce the same acousticalfeatures underlying prosody to command.The nature of his errors might constitutewhat can be termed “ideomotor prosodicapraxia.”

The patient was a 71 year old, retired phy-sician with a 3 to 4 year history of memoryimpairment. Neuropsychological evaluationdisclosed a high average to superior generalintellectual functioning, with mild impair-ment in naming to confrontation and epi-sodic memory for visual and verbal memory.His visuospatial ability remained relativelyunimpaired and was rated as average for hisage. His comprehension for verbal andwritten instruction remained intact. At thepresent time he is still well oriented to timeand place, and is somewhat independent inactivities of daily living. He is, remarkably,not depressed, but does, repeatedly, raiseconcern regarding the “burden he hasbecome to his wife.” Moreover, mild hypop-erfusion in the frontotemporal lobes bilater-ally was seen on SPECT investigation and noevidence of pathognomonic laboratory re-sults were found. Taken together, the patternof episodic memory and naming impairmentsand functional imaging findings was thoughtto be consistent with the early stages ofdementia of the Alzheimer’s type (DAT) inkeeping with National Institute of Neurologi-cal and Communicative DisordersAssociation-Alzheimer’s Disease and RelatedDisorders Association criteria.6 The patientwas consequently referred to our departmentfor “prospective memory book training” andfollow up assessments to index progression ofdisease.

During our sessions his wife had stated thatthe patient could no longer “act” andcomplained that her once “flamboyant” and“unblushing” husband could no longer “putany feeling into his lines” when they read playscripts together. She thought that he had“lost his enthusiasm to act” consequent to hisnew found memory loss and an “understand-able depressive reaction.” It became clear,however, that the patient was remarkably notdepressed and that he maintained normalprosodic speech during conversation. Whenasked to use prosody to command whenreading script, however, this once gallantactor spoke without melody, loudness, stress,nor accent, with inappropriate pauses. Toquantify this patient’s peculiar deficit, thepatient was required to read and repeat wordsand sentences to prosodic command andimitation. Observation revealed five singlewords and five sentences that the patientoften and spontaneously uttered with normalprosody, such as “Honey, PLEASE(!).”These 10 items were used to assess thepatient’s ability to produce prosody tocommand and imitation (table). For exam-ple, the patient was told to read the words

“Honey PLEASE!” with loudness, stress,accentuation of the word “please,” and as ifhe really meant it. If he failed (the words wereread without the acoustical features ex-pected), the patient was asked to imitate theexperimenter’s reading of the word(s) or sen-tence which incorporated the appropriateprosodic elements only after he was asked todescribe the aVective prosodic quality of thephrase to ensure good comprehension. Fiveage matched normal healthy controls volun-teered to read the items found in the table,and in each case, read spontaneously theword or phrase with appropriate and ex-pected prosody.

The patient was unable to read any wordsor sentences with normal (appropriate andexpected) prosody. Indeed, the patient hadlost his ability to “act.” The patient’s use ofprosody did, however, improve dramaticallywith imitation. That is, he was able to repeateight of the 10 items in the table with appro-priate and expected prosody. Interestingly,the single item that he continued to havetrouble producing was item three (Youknow . . .there was a time when . . .). He couldnot pause after the words “you know”suggesting that this patient seems to also havefeatures characteristic of motor aprosodia.Hence, because he had available the knowl-edge to successfully select and use appropri-ate prosody, but failed to produce prosodicspeech to command, dysfunction of thepraxis production system is implied ratherthan the conceptual system.3 Thus, ideomo-tor prosodic apraxia can be defined as aninability to produce prosody to commandduring speech. The precise underlyingmechanism(s)] responsible for producing thisdeficit is unknown, although Heilman et al7

and Tucker et al8 have hypothesised that theright hemisphere is indeed dominant fororganising the aVective-prosodic componentsof language and gestural behaviour and thatthe functional anatomical organisation ofaVective language in the right hemisphere wasanalogous to the organisation of proposi-tional language in the left (non-dominant)hemisphere. Conceptually, therefore, evi-dence of poor aVective prosody to commandyet normal spontaneous aVective prosody,and good aVective prosodic repetition andcomprehension would suggest a “transcorti-cal motor aprosodia.”2 Note however, that thepatient’s spontaneous prosody was unaf-fected whereas spontaneous speech is af-fected by a transcortical motor aphasia.Hence, we might place the critical lesion forprosodic apraxia in the right dorsolateralfrontal lobe, extending into the deep frontal

Ten item prosodic apraxia scale*

Script to be read Type of emphasis

1. Honey PLEASE ! Accentuate PLEASE !2. Are you hungry? Rise in pitch3. You know.......there was a time when I could

recite all the streets in my neighborhoodPause after “You know”

4. Holy COW ! With surprise5. YUP, yup, yup, yup, yup... As if you were distressed with decenting

intonation and stress6. La de da da... With melody7. O Canada, our home and native land... With proper tempo, as if you were singing8. SHIT ! As if you were frustrated and upset9. Thank you As if you sincerely meant it10. May I go to the bathroom, I really need to go

quite badly...As if you really meant it, accentuating the word“REALLY”

Directions: Read the above word[s] and sentences as if you really mean them. Pretend you are auditioningfor a play and you are required to read the lines with the type of emphasis noted beside each line.*These items were selected based on observation of the patients spontaneous speech. Therefore, they arequalitatively constructed and should not be used as a general measure of prosodic apraxia with all patients.


white matter, in keeping with typical domi-nant hemispheric lesions producing transcor-tical motor aphasia. This speculation issupported by the patient’s SPECT findings ofmild hypoperfusion in the frontotemporallobes bilaterally.

KONSTANTINE K ZAKZANISDepartment of Psychology, Division of Life Sciences,

University of Toronto, Canada

Correspondence to: Dr Konstantine K Zakzanis,Department of Psychology, Division of Life Sci-ences, University of Toronto, 1265 Military Trail,Toronto, Ontario, Canada M1C 1A4. [email protected]

1 Monrad-Krohn GH. The third element ofspeech: prosody and its disorders. In: HalpernL, ed. Problems in dynamic neurology. Jerusalem:Hebrew University Press, 1963:101–18.

2 Ross E. The aprosodias. In: Feinberg TE, FarahMJ, eds. Behavioral neurology and neuropsychol-ogy. New York: McGraw-Hill, 1997:699–709.

3 Heilman KM, Rothi LJG. Apraxia. In: HeilmanKM, Valenstein E, eds. Clinical neuropsychology.3rd ed. New York: Oxford University Press,1993.

4 Rothi LJG, Heilman KM. Apraxia: the neuropsy-chology of action. East Sussex, UK: PsychologyPress, 1997.

5 Roy EA. Hand preference, manual asymmetriesand limb apraxia. In: Elliot E, Roy EA, eds.Manual asymmetries in motor performance. BocaRaton: CRC Press, 1996:215–36.

6 McKhann G, Drachman D, Folstein M, et al.Clinical diagnosis of Alzheimer’s disease: re-port of the NINCDS-ADRDA work groupunder the auspices of the Department ofHealth and Human Services Task Force onAlzheimer’s disease, Neurology 1984;34:939–44.

7 Heilman KM, Bowers D, Speedie L, et al. Com-prehension of aVective and nonaVectivespeech. Neurology 1984;34:917–21.

8 Tucker DM, Watson RT, Heilman KM. Dis-crimination and evocation of aVectively in-toned speech in patients with right parietal dis-ease. Neurology 1977;27:947–50.

Vocal cord abductor paralysis inspinocerebellar ataxia type 1

Vocal cord abductor paralysis (VCAP) isconsidered a sign of a poor prognosis in neu-rodegenerative diseases, because severe la-ryngeal dysfunction by VCAP may result inacute airway obstruction and require emer-gency tracheotomy.1

Although VCAP is a cardinal feature inmultiple system atrophy (MSA), it has notbeen reported in several types of spinocer-ebellar ataxia with dominant inheritance. Weevaluated the movements of the vocal cordsof seven patients with SCA1 by laryngofibro-scopy.

Seven unrelated patients with SCA1 whohad the expanded CAG repeat of ataxin-1were investigated. There were two men andfive women ranging in age from 27 to 67 yearsold (mean 44.5 years). Spouses and otherfamily members, in addition to the patients,were questioned about events of stridor, dys-pnoea, and dysphagia. Vocal cord movementwas examined by laryngofibroscopy andrecorded during inspiration and phonation.

The rating scale used to evaluate maximalabduction of the vocal cords during larygofi-broscopy was as follows: (-)=normal;(+)=median position; (++)=paramidline po-sition; (+++)=midline position. For theevaluation of VCAP, we tried the respiratoryflow volume loop study as well in one patient(patient 2) in whom maximal abduction ofthe vocal cords was slightly limited (+) onlaryngofibroscopy.

The correlations between VCAP and CAGrepeat length or duration of illness were ana-lysed with the non-parametric Mann-Whitney U test.

The clinical features, including the vocalcord findings, are summarised in the table.VCAP was present in five of the sevenpatients with SCA1. Although it is diYcult toknow when the VCAP first became manifestin each patient, patient 1 showed VCAP con-firmed by laryngofibroscopy only 2 yearsafter the onset of gait disturbance.

All five patients with VCAP showed milddysphagia requiring no tube feeding, and fourpatients had a history of stridor at night.Patient 1 showed VCAP accompanyingdysphagia without stridor at night even in anearly stage of the disease. The VCAP wasfound to be severe on laryngofibroscopy in allthree patients with breathing diYculty oninspiration. Patient 5, who had the severestVCAP, developed stridor during wakefulnessas well. In patients 4 and 5, the breathing dif-ficulty on inspiration was improved bytracheostomy. The respiratory flow volumeloop study did not detect abnormality inpatient 2.

The CAG repeat number tended to behigher in the patients with VCAP than in thepatients without VCAP (p=0.05), but theduration of illness was not significantlycorrelated with the presence of VCAP(p=0.43).

This is the first report that VCAP is oftenfound in patients with SCA1. As VCAP maynot usually be a late feature in patients withSCA1, evaluation of VCAP is necessary evenin early stages of the disease. It is not surpris-ing to find VCAP in patients with stridor,because stridor is usually caused by airwayobstruction of the larynx. However, VCAPwas detected by laryngofibroscopy in apatient without stridor who had dysphagia.Furthermore, all patients with VCAP exhib-ited dysphagia. We therefore think that laryn-gofibroscopy should be performed in SCA1patients with dysphagia as well as stridor.

The mechanism of VCAP may be dividedinto some types, the paralytic type, the non-paralytic type, and these two combined type.2

The first is possibly caused by loss ofneurons in the nucleus ambiguus.2 3 Thesecond is considered to be due to over-activity of the intrinsic laryngeal muscles.2

Stridor due to paralysis has been found to bemore prominent in sleep than during wake-fulness; whereas stridor by non-paralyticdysfunction has been found both during the

daytime and during sleep.2 4 We suspect thatthe VCAP in patients with SCA1 may bedominantly paralytic, because the nucleusambiguus is sometimes pathologically in-volved in SCA1 and because stridor in ourpatients with SCA1 was more marked insleep.5

Our laryngofibroscopic findings suggestedthat severe VCAP caused breathing diYcultyon inspiration in the patients with SCA1 byobstructing the airway. Moreover, the stridorduring wakefulness as well as sleep indicatedit to be very serious. The important questionconcerns when tracheostomy should becarried out after the diagnosis of VCAP toprevent respiratory abnormalities leading tosudden death. Although we consider tracheo-stomy at the stage when breathing diYcultyon inspiration or stridor during wakefulnessis noted, it awaits further study with a largenumber of patients to decide which stage isbest for tracheostomy.

Furthermore, we now consider endoscopiccord lateralisation as another possible man-agement for VCAP.

T SHIOJIRIT TSUNEMI

T MATSUNAGADepartment of Neurology, Asahi General Hospital,

Chiba, Japan

H SASAKII YABE

K TASHIRODepartment of Neurology

N NISHIZAWADepartment of Oto-Rhino-Laryngology, Hokkaido

University School of Medicine, Hokkaido, Japan

K TAKAMOTODepartment of Neurology, Tokyo Metropolitan

Neurological Hospital, Tokyo, Japan

T YOKOTAH MIZUSAWA

Department of Neurology,Tokyo Medical and Dental University, Tokyo, Japan

Correspondence to: Dr Toshiaki Shiojiri, Depart-ment of Neurology, Asahi General Hospital, I-1345,Asahi-city, Chiba 289–2511, Japan. Telephone 0081479 63 8111; fax 0081 479 60 1210.

1 Williams A, Hanson D, Calne DB. Vocal cordparalysis in the Shy-Drager syndrome. J NeurolNeurosurg Psychiatry 1979;42:151–3.

2 Isozaki E, Shimizu T, Takamoto K, et al. Vocalcord abductor paralysis (VCAP) in Parkinson’sdisease: diVerence from VCAP in multiple sys-tem atrophy. J Neurol Sci 1995;130:197–202.

3 Hayashi M, Isozaki M, Oda M, et al. Loss oflarge myelinated nerve fibers of the recurrentlaryngeal nerve in patients with multiplesystem atrophy and vocal cord palsy. J NeurolNeurosurg Psychiatry 1997;62:234–8.

4 Isozaki E, Naito A, Horiguchi S, et al. Earlydiagnosis and stage classification of vocal cordabductor paralysis in patients with multiplesystem atrophy. J Neurol Neurosurg Psychiatry1996;60:399–402.

5 Genis D, Matilla T, Volpini V, et al. Clinical,neuropathologic and genetic studies of a largespinocerebellar ataxia type 1 (SCA1) kindred:(CAG)n expansion and early premonitorysigns and symptoms. Neurology 1995;45:24–30.

Brief summary of the clinical features and CAG repeat numbers in the patients with SCA1

Patient Age/sex Onset (y)Diseaseduration (y)

Vocal cordparalysis*

Stridor atnight

Stridor duringwakefulness Dysphagia

BreathingdiYculty Tracheotomy

(CAG)n ofmutant allele

1 30/M 28 2 + − − + − − 532 48/F 39 9 + + − + − − 483 27/F 20 7 ++ + − + + − 594 34/M 23 11 ++ + − + + + 525 56/F 46 9 +++ + + + + + 476 67/F 51 14 − + − + − − 467 50/F 41 8 − − − − − − 46

*(−)=normal, (+)=median position, (++)=paramidline position, (+++)=midline position.


Lateral gaze synkinesis on downwardsaccade attempts with paramedianthalamic and midbrain infarct

The symptoms of paramedian thalamic andmidbrain infarct include ocular motor distur-bances mainly in the vertical plane.1 We heredescribe a patient with the additional featureof an unusual horizontal eye movementsynkinesis.

A 60 year old overweight man,with diabetes and mild hypertension,suddenly fell into a coma that lastedfor 4 hours and was followed by slight righthemiparesis, recent memory impairment,hypersomnia, and vertical gaze impair-ment.

On admission to our centre, about 10 daysafter symptom onset, the patient still pre-sented fluctuating drowsiness from which hecould be easily aroused, normal cognitivefunctions with mild attention disturbance,slight right facial weakness, and mild incoor-dination at the finger-to-nose test with hisright arm.

The most important findings involvedocular motor function. Both pupils werenormal in diameter and reacted normallyboth to light and to convergence. The covertest did not disclose any eye misalignment.During attempted fixation, the patient

showed saccade oscillations (usually squareand macrosquare wave jerks)—that is, backto back involuntary horizontal saccades withan amplitude ranging from about 2 to about10 degrees and with an intersaccadic intervalof about 200 ms, that brought the eyes awayfrom and back to the fixation point, at anapproximate rate of three every 2 seconds.Clinical examination of eye movements inthe horizontal plane and visually guidedreflexive saccades recorded by the infraredreflection technique were both normal,whereas the amplitude range of vertical sac-cade and smooth pursuit eye movementscovered only a few degrees of upward gaze.Vertical amplitude range was slightly greaterfor the vestibulo-ocular reflex in the pitch(yes-yes) plane. Moreover, when the patientattempted to make a downward saccade, heshowed a gaze deviation to the left (figure).This synkinesis was more evident when theexaminer lifted the patient’s lids, thuspreventing lid synkinesis during downgaze.Attempted upward saccades did notproduce any horizontal gaze deviation.Finally, the patient showed normal Bell’sphenomenon.

An EEG showed frontal, bilateraltheta and theta/delta activity and sporadicdrowsiness, and MRI (figure) disclosed

a bilateral thalamomesencephalic infarctwhich involved predominantly the rightside.

Horizontal gaze deviation on attempteddownward saccades disappeared after about15 days, whereas vertical gaze impairmentand hypersomnia were unchanged 1 yearlater. Subsequent polisomnographic testingdisclosed sleep apnoea.

The clinical features of our patientare those reported for thalamic infarctinvolving the rostrointerstitial nucleus ofthe medial longitudinal fasciculus (riMLF).By contrast, the leftward gaze deviation elic-ited by the attempt to make a downward sac-cade is at variance with all previous descrip-tions.

The triggering of a saccade requires notonly the activation of the exitatory burst neu-rons (EBNs), but also the deactivation of theomnipause neurons (OPNs), which providetonic inhibition of both horizontal and verti-cal EBNs.

Accordingly, any attempt to activate alesioned riMLF should be associated withmaximal OPN inhibition. However, OPNsdischarge for saccade in any direction and arenot strictly direction selective,2 as shown byhorizontal oscillations during vertical sac-cades detectable in normal subjects.3 Theseoscillations suggest that during vertical sac-cades the inhibition of OPNs disinhibits bothvertical and, to a lesser extent, horizontalEBNs.

In our patient, the horizontal gaze devia-tion was always directed to the left rather thanin both directions as during oscillations.Many ocular motor structures, includingthose located in the midbrain,4 trigger apurely vertical (downward) saccade onlywhen stimulated bilaterally, so as to nullifyhorizontal components with diVerent direc-tion depending on the stimulation side. Thisprobably occurs for the riMLF too, as itshows ipsilateral projections to the abducensnucleus.5 In our patient, the projections to theleft nucleus were probably spared by the factthat the lesion predominantly aVected theright side.

Overall, our patient’s horizontal ocularmotor synkinesis is unusual, and probablyderives from a strong inhibition of OPNs,which in turn frees the horizontal EBNs, andfrom an unbalanced activation of the leftabducens neurons via riMLF projectionsspared from the lesion, although it is not pos-sible to exclude the possibility that the unbal-anced activation of abducens neurons origi-nated from frontal or parietal cortical areas orfrom the superior colliculus rather than fromriMLF projections.

This hypothesis is strengthened bythe reinforcement of the leftward eyedeviation when the examiner kept thepatient’s lids lifted. Since this manoeuvreprevents lid synkinesis, it results in whatresembles an attempted forced lid closurewhich, on the basis of blink induced eyeoscillations,6 is likely to be an additionalstimulus for OPN inhibition. Moreover,although they occur in various conditions,saccade oscillations during fixation are inkeeping with a reduction of OPN inhibitionlevel.7

In conclusion, our patient presented anocular motor synkinesis that should be listedamong those occurring in thalamomesen-cephalic infarcts. This sign is unusual and itis likely to be overlooked, but it is fullyexplicable both by neurophysiology and

(A) and (B) show a thalamomesencephalic ischaemic lesion, hyperintense in T2 weightedscans (SE, TR=2300 ms; TE=25 ms). The lesion involves the anteromedial portion of both thalami,but the right one to a larger exrtent. In the midbrain, the lesion is located around the Sylvianacqueduct, and symmetrically, but prevalently right sided, and involves the area that is locatedposteromedially with respect to both red nuclei. (C) Recording of the horizontal (upper tracing) andof the vertical (bottom tracing) movement recorded respectively from the left and right eye with theinfrared reflection technique (Skalar, IRIS system) during an attempted downward saccade. Thevertical tracing is flat, as the patient was unable to move his eyes downward. By contrast, thehorizontal tracing shows a concomitant leftward saccade. At outset, both tracings show a blinkartifact.

C

0 degR

L

8 deg

left eye horizontal

right eye vertical

0.5 s


by anatomical connections of the saccadesystem.

M VERSINOF SIMONETTIM G EGITTO

M CERONIV COSI

Fondazione IRCCS Istituto Neurologico C Mondino,Pavia, Italy

M VERSINOM CERONI

V COSIDipartimento Scienze Neurologiche

G BELTRAMIDipartimento di Informatica e Sistemistica,

Università di Pavia, Italy

Correspondence to: Dr Maurizio Versino, Diparti-mento Scienze Neurologiche, Università di Pavia,Fondazione Istituto Neurologico C MondinoIRCCS, Via Palestro 3, 27100 Pavia, Italy. Tele-phone 0039 0382 380340; fax 0039 0382 380286;email [email protected]

1 Castaigne P, Lhermitte F, Buge A, et al.Paramedian thalamic and midbrain infarcts:clinical and neuropathological study. Ann Neu-rol 1981;10:127–48.

2 Nakao S, Shiraishi Y, Li W, et al. Cat pontineomnipause neurons: direct inhibitory connec-tion with Forel’ s field burst neurons participat-ing in the genesis of vertical saccades. ActaOtolaryngol 1991;481(suppl):199–204.

3 Zee DS, Fitzgibbon EJ, Optican LM. Saccade-vergence interactions in humans. J Neurophysiol1992;68:1624–41.

4 Kömpf D, Pasik T, Pasik P, et al. Downwardgaze in monkeys. Stimulation and lesionstudies. Brain 1979;102:527–58.

5 Langer T, Kaneko CRS, Scudder CA, et al.AVerents to the abducens nucleus in the mon-key and cat. J Comp Neurol 1986;245:379–400.

6 Hain TC, Zee DS, Mordes M. Blink-inducedsaccadic oscillations. Ann Neurol 1986;19:299–301

7 Averbuch-Heller L, Kori AA, Rottach KG, et al.Dysfunction of pontine omnipause neuronscauses impaired fixation: macrosaccadic oscil-lations with unilateral pontine lesion. Neu-roophthalmology 1996;16:99–206.

Botulinum toxin is a useful treatment inexcessive drooling of saliva

Excessive drooling of saliva or hypersialor-rhea is a common problem in neurodegenera-tive disorders such as motor neuron diseaseor Parkinson´s disease. It is usually caused byswallowing dysfunction and can facilitatechoking, aspiration, and chest infections.Socially it is embarrassing and disabling.There are not many treatment options. Anti-cholinergic drugs are sometimes tried but areusually of little benefit and side eVects(orthostatic hypotension, dizziness, and men-tal confusion, particularly in the elderly) limittheir usefulness.

Occasionally, as a more drastic treatmentirradiation of the parotid gland is carried outwhen hypersialorrhea becomes intractable.

Apart from its established usefulness indystonia, spasticity and strabismus there aresome data showing that botulinum toxininjections are eVective in autonomicdisorders—for example, gustatory sweating1

and hyperhidrosis of the palm.2 It has beenhypothesised that botulinum toxin may helpin sialorrhea.3 In a historical note Erbguthrecently pointed out the potential use ofbotulinum toxin for hypersalivation, quotinga paper by the German physician and poetJustinus Kerner written in 1817.4

Botulinum toxin inhibits acetylcholine re-lease in nerve terminals mainly at the neuro-muscular junction, but also in sympatheticand parasympathetic ganglion cells and inpostganglionic parasympathetic nerves, by

blocking SNAP-25, a protein involved in thefusion of acetylcholine containing vesicleswith the plasma membrane.5

Against this background we evaluated theusefulness of botulinum toxin injections intothe parotid gland in four patients with exces-sive drooling of saliva, with their consent.One patient had young onset secondary gen-eralised dystonia with severe mouth openingspasms, one had advanced Parkinson´s dis-ease, the third patient had progressive supra-nuclear palsy, and the fourth patient hadmotor neuron disease. Drooling in thesepatients was so severe that they had to wear abib or carry a towel around their neck. Withone exception 20 units of Dysport( (Ipsen)were injected superficially subcutaneouslyabove the angle of the mandible at the poste-rior margin of the masseter muscle, avoidingthe bulk of the muscle. The shorter version(5/8”) of the 25 gauge needle was used.Because worsening of dysphagia was fearedonly 10 U Dysport were injected into eachparotid gland in the patient with motor neu-ron disease. Drooling did not significantlyimprove in this patient, possibly due to thelow dose of botulinum toxin used. Hedeclined further treatment.

All the other patients had a beneficialresponse beginning by the end of the firstweek and lasting 6 weeks in one patient and 3to 4 months in the others. Apart from subjec-tive improvement reported by the patientsand caregivers, reduction of drooling wasdemonstrated by the fact that the patients didnot have to use a bib or towel. One patienthad mild worsening of existing dysphagia.Two patients had mild chewing diYculties,possibly due to diVusion of the toxin into themasseter and one patient complained of a drymouth. None developed facial weakness. Allthree patients considered the response goodenough and side eVects suYciently minimalfor them to continue botulinum toxin treat-ment at regular intervals.

Injections of botulinum toxin into theparotid gland (and other salivary glands) maybe an eVective and simple treatment forexcessive disabling drooling of saliva inselected patients.

AM was supported by the Ernst Jung-Stiftung fürWissenschaft und Forschung in Hamburg, Ger-many.

KP BHATIAA MÜNCHAU

P BROWNInstitute of Neurology, Queen Square,

London, UK

Correspondence to: Dr Kailash P Bhatia, Univer-sity Department of Clinical Neurology, Institute ofNeurology, Queen Square, London WC1N 3BG,UK. Tel 0044 171 837 3611 extention 4228; fax0044 171 278 5616; email HYPERLINKmailto:[email protected]

1 Naumann M, Zellner M, Toyka KV, et al. Treat-ment of gustatory sweating with botulinumtoxin. Ann Neurol 1997;42:973–5.

2 Schnider P, Binder M, AuV E, et al. Double-blind trial of botulinum A toxin for thetreatment of focal hyperhidrosis of the palms.Br J Dermatol 1997;136:548–52.

3 Bushara KO. Sialorrhea in amyotrophic lateralsclerosis: a hypothesis of a new treatment:botulinum toxin A injections of the parotidglands. Med Hypothesis 1997;48:337–9.

4 Erbguth FJ. Botulinum toxin, a historical note.Lancet 1998;351:1820.

5 Blasi J, Chapman ER, Link E, et al. Botulinumtoxin selectively cleaves the synaptic proteinSNAP-25. Nature 1993;365:160–3.

Extensive brain calcifications insystemic sclerosis: two cases

Systemic sclerosis (scleroderma) is a multi-system connective tissue disease of unknownaetiology, characterised by progressive fibro-sis of the skin and internal organs includingthe lungs and gastrointestinal tract.1 Patho-logical calcification of soft tissues (known ascalcinosis) is a common feature in theCREST syndrome of scleroderma (calcino-sis, Raynaud’s phenomenon, oesophagealdysmotility, sclerodactyly, telangiectasia). Bycontrast, internal organ calcification is rare,and isolated cases of spinal calcinosis andcalcific constrictive pericarditis have beenreported.2 3 We report here the cases of twopatients with systemic sclerosis whose CTexamination disclosed extensive brain calcifi-cations.

Case 1, a 48 year old man was referred tohospital because of polyarthralgia involvingthe wrists and ankles, Raynaud’s phenom-enon, and tightness of hand skin. Thepatient’s wife had noticed that during theprevious months, he had a slowed mentationand a depressive mood. Physical examinationdisclosed a sclerodactyly but no telangiecta-sias. Routine haematological tests were nor-mal. Antinuclear antibodies were positive at a1/2000 dilution with nucleolar fluorescence.Rheumatoid factors, antidouble strandedDNA and antiphospholipid antibodies werenegative. There was no cryoglobulinaemia.Complement was normal. Lung functiontests showed a restrictive syndrome (forcedvital capacity=75% predicted). Chest radio-graphy was normal, as were oesophagealmanometry and cardiac ultrasonographicexamination. A diagnosis of systemic sclerosiswas made and the patient was given diltiazem(180 mg/day) and ketoprofene (150 mg/day).

Six months later the patient’s neurologicalstatus had worsened. He complained ofmemory loss, poor concentration, and insom-nia. On neurological examination he wasanxious and very slow in answering ques-tions. Mini mental state examination scorewas 22/30. The patient was oriented to place,but not to time. Anterograde amnesia wasnoted. Agnosia, apraxia, and aphasia wereabsent. There was no muscle weakness andmuscle tone was normal, as were tendonreflexes. Plantar responses were both flexor.There was no sensory loss or impairment ofcranial nerves. Systemic sclerosis signs wereunchanged. Routine hematological tests werenormal. Results of blood chemical tests werealso unremarkable (serum electrolytes, urea,creatinine, iron), including phosphorus andcalcium metabolism (serum parathyroid hor-mone concentration, blood calcium andphosphorus, 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D, calciuria, and phospha-turia). Serum concentrations of free trio-dothyroxine, free thyroxine, and thyroidstimulating hormone were normal. Serologi-cal tests for syphilis, HIV-1,2, and Lyme dis-ease were negative. Brain CT showed bilat-eral extensive calcification in the dentatenuclei (figure 1A), basal ganglia, and subcor-tical white matter (figure 1B). On MRI T1weighted images and T2 weighted images,calcification was visible as a low intensity sig-nal. The patient was given fluoxetine (20mg/day) and bromazepam (6 mg/day). At fol-low up, 1 year later, the patient’s clinical sta-tus was unchanged, as was brain CT.

Case 2, a 64 year old right handed womanwas admitted to hospital for evaluation of aRaynaud’s phenomenon which had lasted formore than 10 years. At physical examination,sclerodactyly and tightness of the facial skin


were noted. Telangiectasias were present onthe face, hands, and palate. The patient com-plained of pyrosis. Oesophageal manometryshowed abnormalities of oesophageal motil-ity. Hand radiography disclosed soft tissuecalcifications. Anticentromere antibodieswere positive at a 1/1000 dilution. A CRESTsyndrome was diagnosed and the patient wasgiven buflomedil (600 mg/day) and pred-nisone (25 mg/day).

One year later she was admitted for theevaluation of recent transient ischaemicattacks (TIAs). During the previous week shehad experienced three bouts of expressiveaphasia and right hemiplegia, each lastingabout 10 minutes. She never smoked and didnot have diabetes, hypertension, or dyslipi-daemia. The neurological examination wasnormal. Routine blood chemical tests werenormal (serum electrolytes, urea, creatinine)including phosphorus and calcium metabo-lism (serum parathyroid hormone concentra-tion, blood calcium and phosphorus, 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D,calciuria, and phosphaturia). Cranial CTshowed bilateral calcifications of the basalganglia, and faint calcifications of the dentatenuclei and rubrum nucleus. Moderate cer-ebral and cerebellar atrophy was noted.Duplex carotid ultrasound and 24 hour ECGrecording were normal. Echocardiographyshowed a normal left ventricle with anejection fraction of 60%. There was mild cal-cification and a thickening of the mitral valveleaflets. Aspirin (250 mg daily) was given athospital discharge. No further TIA occurredduring a 5 year follow up and the patient’sclinical status has remained unchanged.

Systemic sclerosis is a multisystem diseasepredominantly aVecting the skin, lungs,vascular system, and gastrointestinal tract.1

Neurological involvement occurs in a fewpatients (ranging from 0.8 to 18.5%) includ-ing cranial nerve abnormalities, peripheralneuropathy, CNS vasculitis, and autonomicperipheral neuropathy. To our knowledge,extensive cerebral calcifications have not yetbeen reported.

Calcification of the brain is discovered in0.8 to 1.2% of subjects undergoing routineCT examination, mainly in the globuspallidus. In most cases the deposits are smalland involve older patients who remainasymptomatic, leading to the concept of“physiological” senescent basal gangliacalcifications.4 On the other hand, basal gan-glia calcifications, often associated with den-tate nuclei calcifications, have been reported

in more than 30 conditions, including abnor-malities of calcium phosphorus metabolismsuch as pseudohypoparathyroidism.5

Systemic sclerosis leads to the formation ofcalcium deposits in the subcutaneous tissue.Rarely, the calcific process has been shown toinvolve the spine or pericardium.2 3 Recently,Heron et al described two cases of cerebralinvolvement in systemic sclerosis.6 In bothcases necropsy showed extensive wall calcifi-cation of the small arteries and arterioles ofthe brain. Our two patients have sclerodermaand extensive striopallidodentate calcifica-tions and metabolic investigations failed todisclose any specific aetiology in either case.We think that scleroderma should be addedto the list of conditions described as occur-ring with basal ganglia calcification.

The pathogenesis of the formation ofcalcium deposits in systemic diseases remainspoorly understood. However, pathologicalcalcification can be subdivided into meta-static (occurring in undamaged tissues whenextracellular calcium and phosphate concen-trations are increased) and dystrophic (ocur-ring in injured tissue when extracellularcalcium and phosphate concentrations arenormal) calcification.7 In our patients, as inthe patients of Heron et al, the brain calcifyingprocess may be related to primary cerebrov-ascular changes induced by systemic sclero-sis.

Routine brain CT examination in systemicsclerosis could help to determine the trueincidence of basal ganglia calcifications andtheir clinical relevance.

PATRICK BLANCOJEAN-FRANÇOIS VIALLARD

EMMANUEL ELLIEISABELLE FAURE

PATRICK MERCIÉJEAN-LUC PELLEGRIN

BERNARD LENGClinique de Médecine Interne, Hôpital Haut-Lévêque,

avenue de Magellan, 33604 Pessac, France

Correspondence to: Dr Jean-François Viallard,Clinique de Médecine Interne, Hôpital Haut-Lévêque, Centre François Magendie, 33604 Pessac,France. Telephone 0033 556556483; fax 0033556556484.

1 Seibold JR. Scleroderma. In: Kelley WN, HarrisED, Ruddy S, et al, eds. Textbook of Rheumatol-ogy. Philadelphia: WB Saunders 1997:1133–62.

2 Ward M, Curé J, Schabel S, et al. Symptomaticspinal calcinosis in systemic sclerosis (sclero-derma). Arthritis Rheum 1997;40:1892–5.

3 Panchal P, Adams E, Hsieh A. Calcific constric-tive pericarditis. A rare complication of

CREST syndrome. Arthritis Rheum 1996;39:347–50.

4 Harrington MG, Macpherson P, McIntosh WB,et al. The significance of the incidental findingof basal ganglia calcification on computedtomography. J Neurol Neurosurg Psychiatry1981;44:1168–70.

5 Ellie E, Julien J, Ferrer X, et al. Extensivecerebral calcification and retinal changes inpseudohypoparathyroidism. J Neurol 1989;236:432–4.

6 Heron E, Fornes P, Rance A, et al. Braininvolvement in scleroderma: two autopsy cases.Stroke 1998;29:719–21.

7 Anderson HC. Calcific lesions: a concept. ArchPathol Lab Med 1983;107:341–8.

CORRESPONDENCE

All tibial foot: an electrophysiologicalartifact

Yamashita et al1 claim they have proved an“all tibial foot” for the motor innervation, ananomalous dual innervation of the tibialisanterior muscle by the deep peroneal andposterior tibial nerve, and a sensory coinner-vation of the skin between the first andsecond toes by the tibial and deep peronealnerve in a patient. To support their view theyquote the letters of Linden and Berlit2 and ofGlocker et al,3 ignoring our letter4 and that ofMagistris and TruVert5, both considering theconclusions of Linden and Berlit and Glockeret al to be wrong. I point out that thementioned letter of Linden and Berlit2 andour response to it were published in the sameissue.

We have recorded a compound muscleaction potential (CMAP) with a negative ini-tial deflection on tibial nerve stimulation in83% of 50 subjects, using a surface electrodeover the extensor digitorum brevis.4 In thesame subjects no potential was recorded bymeans of a concentric needle electrodeinserted in the extensor digitorum brevis.1 Inour view, this proves that the CMAP recordedby surface electrode over the extensor digito-rum brevis on tibial nerve stimulation is aremote potential originated in the plantarmuscles (volume conducted potential). Fur-thermore, we consider that the CMAPrecorded over the tibialis anterior muscle bysurface electrode on tibial nerve stimulationin the popliteal fossa, as reported by Yamas-hita et al,1 represents a volume conductionpotential originating in the foot and toe flex-ors. The sensory nerve action potentialrecorded dorsally in the space between thefirst and the second toes on tibial nervestimulation could also be a volume con-ducted potential originating in the first com-mon plantar digital nerve, as the distancebetween this nerve and the recording elec-trode is short. Such volume conduction phe-nomena are known to occur on surfacerecordings from the median nerve at the wristin severe carpal tunnel syndrome, when theforth finger is stimulated. It is unclear whyYamashita et al could not record a CMAPover the extensor digitorum brevis bilaterallyon deep peroneal nerve stimulation in theiryoung patient who did not have neuropathy.A probable explanation is a bilateral aplasia ofthe extensor digitorum brevis, comparablewith the known aplasia of the thenar.6 The

Brain CT of case 1 shows dense symmetric calcifications in the dentate nuclei (1A) and basal ganglia(1B).


appropriate examination would have been aneedle EMG of the extensor digitorumbrevis.

GEORGIOS AMOIRIDISDepartment of Neurology, University of Crete, PO Box

1393, 71 110 Heraklion, Crete, Greece. Telephone0030 81 394651; email [email protected]

1 Yamashita M, Mezaki T, Yamamoto T. “Alltibial foot” with sensory crossover innervationbetween the tibial and deep peroneal nerves. JNeurol Neurosurg Psychiatry 1998;65:798–9.

2 Linden D, Berlit P. The intrinsic foot musclesare purely innervated by the tibial nerve (alltibial foot): an unusual innervation anomaly.Muscle Nerve 1994;5:560–1.

3 Glocker FX, Deuschl G, Lucking CH. Trau-matic lesion of the common peroneal nervewith complete foot drop and preserved dorsi-flexion of the toes: an innervation anomaly.Muscle Nerve 1995;8:926–7.

4 Amoiridis G, Schols L, Meves S, et al. Fact andfallacy in clinical and electrophysiologicalstudies of anomalous innervation of the intrin-sic foot muscles. Muscle Nerve 1996;9:1227–9.

5 Magistris MR, TruVert A. Extensor digitorumbrevis innervated by the tibial nerve (all tibialfoot): anomalous innervation or technicalpitfall? Muscle Nerve 1997;7:906–8.

6 Iyer KM, Stanley JK. Congenital absence offlexor pollicis brevis and abductor pollicisbrevis. Hand 1982;3:313–6.

BOOK REVIEWS

Assessment Scales in Old AgePsychiatry. by ALISTAIR BURNS, BRIAN

LAWLOR, and SARAH CRAIG. (Pp 302, £29.95).Published by Martin Dunitz, London, 1999.IBSN 1-85317-562-5.

If it moves - measure it. Such is the trend inpsychiatry and this has led to a proliferationof assessment scales of variable utility; fromthe esoteric to the ubiquitous. This book hasthem all and is, quite genuinely, one of themost useful volumes I have seen for a verylong time. It covers everything from theAMTS through the MMSE to the KEW cog-nitive test. Each scale is presented in fulltogether with a short commentary, criticalreferences and, usefully, an estimated timetaken to perform the test together with anaddress to contact the original author.

The scales presented are divided into thosecovering depression, neuropsychiatric assess-ments, activities of daily living, global assess-ments, visible assessments, delirium, care-giver assessments and scales for memoryfunction. This organisation, together with auseful and functional index, will make thetask of selecting an appropriate scale mucheasier in the future. Some of these scales arecovered by patent law and my only quibble isthat it would have been useful to know whichinstruments can be reproduced and for whatpurpose without fear of being billed for theprivilege. Maybe highlighting such infor-mation would give unpatented authors astimulus to visit their lawyer and thereforeincrease this deplorable practice.

This aside, I cannot recommend this bookhighly enough. By contrast with books thatcome highly recommended in diverse reviewsbut remain pristine on your library shelf, ifyou are ever fortunate enough to get hold ofthis volume from your institution library thenyou will, I guarantee, find it well thumbed.For academics, the book is a prayer answeredand will become an essential resource in the

planning and execution of any clinicalresearch. For clinicians, psychologists, occu-pational therapists, and other members of themultidisciplinary old age psychiatry team, thecollection of so many scales covering allaspects of mental health of elderly peopleshould encourage the use of valid and reliableassessment procedures that enhance ratherthan detract from good clinical practice. Myonly question is “Why didn’t somebody dothis before”.

SIMON LOVESTONE

Lumbar Disc Herniation. By FRANCO

POSTACCHINI. (Pp 623, US$299.00).Published by Springer-Verlag, Wien, 1999.ISBN 3-211-83118-5.

Professor Franco Postacchini is an orthopae-dic surgeon at the University “La Sapienza” inRome and is a well known widely respectedspinal surgeon. He is to be congratulated onthe production of this book which is wideranging, comprehensive, and beautifully illus-trated. The management of lumbar discdisease is fraught with uncertainty and thereare many diagnostic and therapeutic pitfalls.The author has succeeded in addressing mostof these controversies in a clear and logicalfashion. He occasionally blurs the distinctionbetween theories and established scientific factforgetting that the practice of medicine and inparticular the management of spinal disordersis full of paradoxes. For example, he states thatlarge extruded disc fragments are unlikely toresolve spontaneously and will usually requiresurgical treatment. This seems a logicalproposition but my experience is that many ofthese large extrusions undergo complete clini-cal and radiological resolution within 2 or 3months. Paradoxically, it is often the smallercontained disc prolapses which fail to improvewith conservative measures. Like many ortho-paedic surgeons he is persuaded by the alluringtheories of discogenic low back pain and wor-ships at the altar of segmental microinstability.However, I agree with much that he haswritten and diVerences of emphasis areinevitable in a field that is strong on dogmaand short of established truths.

I would have no hesitation in recommend-ing the text to trainees as the book is veryreadable and makes a good introduction tothe management of lumbar disc disease.Nearly all aspects of diagnosis and treatmentare covered but I was disappointed with thechapter discussing results of surgery. There isno mention of the use of objective validateddisability and quality of life instruments inthe assessment of outcome. For a text thataims to be comprehensive this constitutes aserious omission. It is because practitionershave failed to use objective outcome meas-ures to establish the natural history of lumbardisc disease and the eVects of therapeuticinterventions that there remains so muchuncertainty about management. These un-certainties cover (among others) physi-otherapy, manipulation, timing of radiology,timing of surgery, whether spinal fusion isever indicated, and what treatments are clini-cally and cost eVective. Despite these draw-backs, surgeons who manage lumbar discdisorders will want to have a copy of thisbook, either on their own or their departmen-tal library’s shelf.

RODNEY LAING

Chronic Fatigue and its Syndromes. BySIMON WESSELY, MATTHEW HOTOPF, andMICHAEL SHARPE. (Pp 428, £29.50).Published by Oxford University Press,Oxford, 1998. ISBN 0-19-263046-6.

My first sensation is that such a book is longoverdue. There is a paucity of books written byclinicians on this subject compared with themany gloom and doom tracts written bysuVerers who have apparently been ill withCFS/PVS/ME for 20 years and which depressthe hell out of the patients that I see. Why thislack? Undoubtedly because relatively few doc-tors want to put their heads above the parapetand profess to a great expertise or desire to seepatients with it. It is a condition for which thereis no objective test, no objective monitoring ofprogress, whose symptoms are so vague thatthey repeatedly defy classification and can onlybe catalogued. Among those patients with agenuine postinfectious fatigue (who mainly getbetter relatively quickly) there is a large body ofthose with hospital records measured by thekilogram who arrive with sheaves of self com-piled additional notes and occupy hugeperiods in the clinic to little eVect other thanfrustration for both doctor and patient. Whythey do this is a diVerent story.

So what about the book? Firstly I must sayit comes armour plated against criticism witha pack of glowing references from ennobledand famous physicians on the back. Presum-ably the authors chose these referees whosemajor feature in common is distinction infields other than chronic fatigue syndrome.So, for whatever reason they were chosen theresult is also to make this a poisoned chaliceto review if one has the temerity to disagreewith such a company.

Chronic fatigue and its Syndromes is written bythree psychiatrists. That in itself is a little oddas although psychiatric illness is quite commonin the chronic fatigue syndrome group, mostpatients with chronic fatigue syndrome arevery reluctant to be seen on first presentationby psychiatrists. One usually has to reassurethem that one is not dismissing their symp-toms as psychiatric if one does suggest such areferral. What proportion of patients I wondergo to psychiatrists as their first hospitalreferral? I would guess very few.

The book is very comprehensive. As a sourceof references on the subject of fatigue it isencyclopaedic. It begins with a nice history ofthe syndrome. Of a 428 page book only 44pages cover assessment of the patient and avery telling paltry 14 are devoted to treatment.What is all the rest? The answer is that theauthors have attempted to deal with everyfatigue associated subject including: aetiologyof chronic fatigue syndrome (summary - notknown) and in a move which Sir HumphreyAppleby might have described as “courageous“they have ventured way outside their own areasof expertise into subjects such as viruses andimmunity. The result is a collection of compre-hensively referenced and uncritically selectedfacts from the literature, many of which are veryuseful, such as the collected data on enterovi-ruses and chronic fatigue syndrome but muchof which is of uncertain value—for example,“divorced and separated women have highertitres of EBV-VCA antibodies” quoted out ofcontext but not unrepresentatively. To gleaninteresting and important information fromthis book requires a fair degree of skill indistinguishing timber from forest.

When in their own field of psychiatry thereis, unsurprisingly, a much more confidentand informative air to the book. It is


disappointing and somewhat introspectivethat they did not think that subjects such asneurobiology, microbiology, and immunol-ogy might justify equally expert contributors.

Is it an easy read? There is no easy way towrite on a subject such as this, bedevilled bylack of objective facts and the writers havechosen a discursive, debating style whichwhen not tightly controlled can slip towardsverbosity. This, however, is not a standardmedical text book and it would be unfair tomake direct comparisons..

Is this a useful book? As a source ofreferences, yes. As a guide to clinicians I amless convinced. The two commonest questionspatients ask are hardly mentioned: “How longdoes the illness last?” and “What are mychances of recovery” Cognitive behaviouraltherapy is concisely and usefully summarised.There is a single page on complementarytreatment, which again is often an area of con-siderable interest to patients notwithstandingthe lack of controlled evidence for or against it.

On that note it is perhaps appropriate toquote one very intelligent patient withchronic fatigue syndrome I saw who becameill during his PhD. “I’ve done a lot of readingand internet searching about the causes andpossible cures of this, before I came to seeyou” he said “It seems to me that most peo-ple have an illness a bit like that which youcommonly see after glandular fever andnobody seems to think it odd that after glan-dular fever you can feel unwell for quite along time” he continued “If most people getbetter from this” (and many do) and if you tryall sorts of other treatments like homoeopa-thy, acupuncture, meditation then the oneyou were doing when you got better will bethe one you think cured you”. He had ofcourse discovered the maxim of entertainingthe patient while nature gets them better. Onecould do worse perhaps than keep patientswith chronic fatigue syndrome occupied, ifnot necessarily always entertained for quite awhile, by recommending this book for themto read. They might end up with a greaterunderstanding of fatigue and they would cer-tainly realise how little really is known andhow the search for an instant cure (whichdrives many of them) is currently futile.

ANDREW LEVER

Migraine and Headache Pathophysio-logy. Edited by LARS EDVINSSON. (pp 184,£65.00). Published by Martin DunitzPublishers, London, 1999. ISBN1-85317-737-7.

New information about how and why mi-graine happens continues to break on us in adizzying succession of waves coming fromvarious journals in diVerent disciplines. Weneed an accessible, understandable, and uni-tary vehicle to collect, organise, and presentthis information. Journals, the Internet, andthe abstracting services have their place, butfor this purpose nothing beats the book. Howwell does Lars Edvinsson’s Migraine andHeadache Pathophysiology meet this need?

This book has several attractive features.Recognising that some of the world’s best sci-ence is now being done in the laboratories ofindustry, it has enlisted as authors severalleading researchers from the major pharma-ceutical manufacturing companies, in additionto “the usual suspects” from academe. Notonly has this introduced some exciting writersto the “review book” audience, but it has pro-vided a particular insight into the science of

determining why drugs work, or don’t work, inmigraine. The book is up to date, containingmany 1998 references. Another strong point ofthe book it its comprehensiveness; though only184 pages long, it covers every major aspect ofthe pathophysiology of migraine. There arechapters on innervation of cranial bloodvessels, receptor physiology, neurotransmit-ters, 5-hydroxytrptamine receptor subtypes,cortical spreading, depression, neurogenicinflammation, arteriovenous shunts, cerebralhaemodynamics, and animal modelling. Thisis achieved at the expense of some pretty terseprose at times, which can make it diYcult forthe non-expert to follow. It should be notedthat, the title notwithstanding, this book dealsalmost exclusively with the pathophysiology ofmigraine, and the reader who buys it to getsome insight into the mechanisms of tension-type headaches, or other kinds of headache, isgoing to be disappointed.

Who should read this book? Certainly themigraine research should. Though much of itwill be familiar to those who have kept up withthe literature, it is nevertheless an attractiveand handy compendium of current researchinformation. Moreover, the first chapter (byLars Edvinsson) and the last chapter (by PeterGoadsby) are very pretty syntheses of the field.What about clinicians? Some of it is heavygoing for people like me, who are not basic sci-entists. But I got through it all in about 6hours, and found I knew more about migrainecoming out than I did going in—which makesreading it a very worthwhile exercise.

JOHN EDMEADS

Systemic Diseases, Part II. By MJ AMINOFF

and CG GOETZ. (Handbook of ClinicalNeurology, series edited by PJ VINKEN and GW

BRUYN. (Pp 486, US$224.00). Published byElsevier Science, Amsterdam, 1998. ISBN044481289X.

There is something about the anachronisticbinding of the Handbook of Clinical Neurol-ogy series that is rather reassuring. Surely ifclassic phenomenological neurology is to befound anywhere, it will be between these fakeleather embossed covers. This volume, the sec-ond of three on the neurology of systemicdiseases, does nor disappoint. Here, in 450pages of close type and few illustrations, arecovered the neurology of orthopaedic, endo-crine, gastrointestinal, and metabolic disorders.Goetz and AminoV, the volume editors, haveassembled an authoritative panel of authorsthat equitably straddle the Atlantic. There aredetailed reviews of familiar territory such asdiabetes, orthopaedic trauma, thyroid diseases,and porphyria. Cole’s historical survey of B12deficiency is particularly fine. In addition thereare excellent chapters on more arcane topics—for instance the neurology of pancreatic trans-plantation and intestinal pseudo-obstruction.Perhaps the movement disorders associatedwith coeliac disease could have been men-tioned and a chapter on the neurology ofinflammatory bowel disorders is certainlylacking. But these are trifling complaintsagainst a text that, with its twin volumes, is sig-nificantly more comprehensive than any otheraccount of the neurology of systemic diseases.It is hard to imagine a practicing neurologistrequiring (or easily aVording) a personal copyof all three volumes, but the local medicallibrary should certainly buy them; bothneurologists and general physicians will workthe better for having them close to hand.

ALASTAIR COLES

Advances and Technical Standards InNeurosurgery. Volume 24. Editor In ChiefF COHADON. (Pp 57, US$159.00). Publishedby Springer-Verlag, Wien, 1998. ISBN3-211-83064-2.

Advances and Technical Standards in Neurosur-gery is sponsored by the European Associ-ation of Neurosurgical Societies. The inten-tion is to publish reviews of topics in whichrecent advances have been made, and toinvite acknowledged experts to present indepth accounts of established knowledge invarious fields of neurosurgery.

The advances under review in this volumeare the contribution of the septal region tomemory, the in vivo metabolic investigationof cerebral gliomas with PET, and the use ofimage guidance in neurosurgery. In the tech-nical standards section, Professors Valavanisand Yasargil discuss the endovascular treat-ment of arteriovenous malformations, DrGuglielmi reports on the interventional neu-roradiological treatment of intracranial aneu-rysms, and Dr Sussman and colleaguesdescribe the management of benign intracra-nial hypertension.

This book is aimed primarily at young neu-rosurgeons, but is an excellent source of refer-ence for those who are already trained. Thefact that it is in its 24th volume is a testamentto its success in achieving this objective.

ROBERT MACFARLANE

Hospitalist Neurology. Blue Books ofPractical Neurology. Edited by MARTIN A

SAMUELS. (Pp 748, £70.00). Published byButterworth Heinemann, Oxford, 1999.ISBN 0750697792.

It seems that there is a new specialty in NorthAmerican neurology, hospitalist neurology.The drive to promote managed health carehas apparently resulted in hospitals “filled tooverflowing with more acutely ill patientsrequiring a pace of evaluation unprecedentedanywhere in the world”. Enter the hospitalistneurologist. Unencumbered by the demandsof outpatient neurology, he or she stumblesthrough the wards of the general hospital“faced with a dizzying array of neurologicproblems”. Most British neurologists have award referral practice and will not beimpressed by its elevation to the status of aspecialty and still less by the agrammaticaltitle Martin Samuels has chosen for it. Whichis a shame, because this book deserves a widereadership. One in the Butterworth Hein-mann series of Blue Books of Practical Neu-rology, it is attractively produced and reason-ably well illustrated. Its place on yourbookshelf is earned by collating the neuro-logical aspects of diverse medical specialties:to name a few, organ transplantation, ortho-paedics, oncology, and urology. A quickglance here before a ward referral might wellbe rewarding. However, the chapters on moreconventional neurological topics, such asneuro-ophthalmology, stroke, and seizuresare probably briefer than most neurologistswould require. So, for those made dizzy bythe delirious patient after bypass, the en-cephalopathic flapping on the transplantward, or the weak and wasted on intensivecare units, this is for you. And remember: youare a hospitalist neurologist.

ALASTAIR COLES


Symptom Management in MultipleSclerosis, 3rd Edition. By RANDALL T

SCHAPIRO. (Pp 176, US$19.95). Published byDemos Medical Publishing, New York,1998. ISBN 1-888799-22-6.

This is one of a new type of medicaltextbook written to meet the needsof an increasingly informed patientpopulation. Aimed very much at those withmultiple sclerosis, their families, and care-workers, it is simple and clearly written withjargon and technical terms kept to aminimum but without patronising. Chronicdiseases, and especially multiple sclerosis,are not always well managed by the physi-cian. Too many of us think that there is nocure and feel helpless in a busy clinic facedwith the patient with a long list of com-plaints. There are too few specialty multiplesclerosis clinics in which neurologists, painspecialists, uroneurologists, physiotherapistsetc liase.

Patients often feel left in the dark,unaware which of their symptoms can beattributed to their multiple sclerosis andwhether it is “worth bothering a busydoctor”. Many can cite bad experiences intheir past when they have been fobbedoV with well meaning reassurance butwithout practical help. Dysaesthesia, sexualproblems, and urinary incontinence areonly a few of the symptoms that can bringmisery to the lives of patients and their fami-lies and which are poorly addressed by doc-tors. This book, in a language accessible tomost (and with a glossary to explain someunavoidable jargon), explains multiplesclerosis, its symptoms, and what mightrealistically be obtained in terms ofsymptom control. All aspects are coveredand nothing considered too trivial; constipa-tion or cold feet might be extremely tryingfor an individual patient and each is consid-ered.

The old idea that is doesn’t help apatient to know too much about his disease(“it will only make him introspective andhypochondrial”) is outdated. Multiple scle-rosis can hit anyone and patients now want,and deserve, to be informed. While doctorsfind it challenging to be faced with a patientequipped with the latest information downloaded from the internet or well informedhaving read a book such as this, this is achallenge to which we must be ready to rise.This textbook provides the informationpatients want and fills the gap left by busydoctors. It should be marketed appropriatelyand we must be ready to respond to thereaction of patients. Perhaps someone withmultiple sclerosis should have been invited towrite this review.

GILLIAN HALL

Goldenberg G, Schuri U, Grömminger O,Arnold U. Basal forebrain amnesia: does thenucleus accumbens contribute to humanmemory? J Neurol Neurosurg Psychiatry 1999;67:163-8.

During printing, the figure in this paper (p164) was made darker than the original. Thecorrect version appears below.

Cd

GP

CI

DB

B

A

Acb

(A) T1 weighted MRI of the lesion. (B)Anatomical scheme of the centre of the lesion, correspondingto the leftmost image of the bottom row of the MRI. The right side of the figures corresponds to the leftside of the brain. GP=globus pallidus; Cd=caudate nucleus, Acb=nucleus accumbens; CI=capsulainterna; DB=diagonal band.

CORRECTION


Documents

Journal of Neurology, Neurosurgery and Psychiatry (JNNP ...example, reading span38 (span=2, 6th percen-tile) or a computerised test of verbal working memory28 (level 3, 1st percentile)