rournal ofNeurology, Neurosurgery, and Psychiatry 1995;59:293-298

Arachnoid cysts of the left temporal fossa:impaired preoperative cognition and postoperativeimprovement

Knut Wester, Kenneth Hugdahl

AbstractThirteen adult patients were operated onfor symptomatic arachnoid cysts in theleft temporal fossa; seven with an inter-nal shunt procedure during local anaes-thesia, and five with a craniotomy withfenestration of the cyst to the basal cis-terns. In one patient, an initial internalshunt was transformed to a cystoperi-toneal shunt. After surgery, all patientsexperienced relief of symptoms.Reduction of cyst volume occurred in 11patients. The patients were tested forbrain asymmetries related to languageand verbal memory before and afteroperation, with a dichotic listening tech-nique with simultaneous presentation ofdifferent auditory stimuli to the two ears.In the preoperative memory test, thepatients showed impaired total recallcompared with healthy control subjects,and recall from the right ear was signifi-cantly impaired. The patients also per-formed poorly in a forced attention taskconsisting of dichotic presentations ofconsonant-vowel syllables. In addition toclinical improvement, the surgical proce-dures led to improvements in bothdichotic perception and memory. Overallmemory performance was enhanced,mainly because of improved recall fromthe right ear. This normalisation ofmemory function was found as early asfour hours after the operation. Theresults indicate that arachnoid cysts inthe left temporal fossa may impair cogni-tive function, that neuropsychologicaltests are necessary to disclose theseimpairments, and that cognitiveimprovement occurs after surgery.

(3 Neurol Neurosurg Psychiatry 1995;59:293-298)

Key words: arachnoid cysts; language; memory.

Lesions occupying intracranial space, such ashaematomas, tumours, or cysts may impairbrain function. The underlying mechanismsfor this disturbance in the case of rapidlyexpanding haematomas are well understood,and may be related to changes in general orlocal tissue perfusion. The reduced cerebralperfusion may be due to raised intracranialpressure, complicated by the presence ofoedema, itself a mass lesion, changes incerebrovascular reactivity, or the diffusion of

possible toxic products. The mechanismsunderlying neurological deficit in such indo-lent lesions as arachnoid cysts are far fromclear. The intracranial pressure is seldomraised enough for reduced tissue perfusion tobe the mechanism underlying loss of function.Because clinical improvement may occur afterdecompression, it would seem likely that theneurological deficit is related to processeswhich, although impairing neuronal function,do not necessarily result in cell death.

Arachnoid cysts are thought to be due to amaldevelopment of the cerebral meninges."2This may be present at birth or develop soonafter.34 After an initial growth in early life,most cysts are thought to remain stable overmany years,5 having reached a permanent sizewhich may well be substantial. Very few cystshave been reported to disappear sponta-neously.6-'0 Moreover, some seem to continueto grow in adult life; although slowly."The symptoms associated with arachnoid

cysts vary according to the site of the lesionand the neural structures surrounding them.In adults, subjective complaints such asheadache and dizziness are common. Majorneurological symptoms such as motor or sen-sory deficits, epilepsy, or language problemsmay also occur, although they are often slightand tend to develop late in life. The lack ofdramatic symptoms, even in patients withlarge arachnoid cysts, probably reflects thebrain's ability to compensate for the presenceof a slowly growing or stable expansion.

If arachnoid cysts cause disabling symp-toms such as a paresis, dysphasia, epilepsy, orsevere headache, or they are complicated by asubdural haematoma, there is a clear indica-tion for surgery. On the other hand, if thesymptoms are less dramatic, some authorsfavour conservative treatment.'2-'7 The rela-tive paucity of symptoms despite massive dis-placement of brain tissue is taken to indicatesatisfactory cerebral compensation.'8Moreover, some earlier series reported a sig-nificant mortality and morbidity after surgeryfor these "uncomplicated" cases, and this hasalso prompted caution.'3 19-22

In common with any other intracranialexpansive lesion, arachnoid cysts have thepotential to impaire cognitive function. Suchimpairments may well contribute to thepatient's incapacity, but are not easily shownby a clinical neurological examination. Todate, subtle cognitive impairments associatedwith intracranial expansions have receivedmuch less attention than the more dramaticand easily recognisable overt neurological

Department ofNeurosurgery,University of Bergen,School ofMedicine,Haukeland Hospital,N-5021, Bergen,NorwayKnut WesterDepartment ofBiological andMedical Psychology,University ofBergen,Arstadveien 21, N-5011,Bergen, NorwayKenneth HugdahlCorrespondence to: DrKnut Wester, Departmentof Neurosurgery, HaukelandHospital, N-5021, Bergen,Norway.Received 3 January 1995and in revised form3 April 1995Accepted 19 April 1995

293 on N

ovember 5, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.59.3.293 on 1 S

eptember 1995. D

ownloaded from

Wester, Hugdahl

symptoms and findings. Indeed, there are fewreports dealing with cognitive functions inpatients with arachnoid cysts. Lang et al3found a reduced ability to learn and memorisehemisphere specific material in patients withmiddle fossa cysts, whereas other reports havefailed to show cognitive impairments in suchpatients.'5 24

If cognitive deficits could be shown inpatients with otherwise "silent" cysts, and ifcyst surgery improves these functions, theindications for surgery may be expanded con-siderably, and the decision to operate wouldbe taken on a broader basis.

Nearly half of the arachnoid cysts occur inthe middle cranial fossa,' with a pronouncedpreference for the left side.25 Most cysts there-fore affect the speech dominant hemisphere.For this reason, two tests were chosen for thepresent study, based on the dichotic listeningtechnique.26 These tests are sensitive to differ-ences in hemispheric functions and in particu-lar probe left hemispheric verbal functions.Dichotic listening procedures have previouslyproved sensitive in reflecting changes in pro-cessing of verbal stimuli produced by discreteelectrical stimulation of the ventrolateralthalamus27 28 and in recovery of language func-tions after a stroke.29 Moreover, a memoryversion of the dichotic listening test was usedby Christianson et aP' when investigatinghemispheric specific memory impairments intwo cases of localised head trauma.

Patients and methodsPATIENTSThirteen consecutive patients (11 men, twowomen) were included in the study as theexperimental group. To keep the material ashomogenous as possible, only patients withcysts in the left temporal fossa were included.Another inclusion criterion was clinicalimprovement after surgery (experienced by allthe patients). Handedness was defined withthe help of a Norwegian translation of thequestionnaire developed by Raczkowski et al."The patient had to indicate 13 of the 15 itemsperformed with the right hand to be classifiedas right handed. Two of the patients (1 and 7)indicated preference for the left hand; theothers were classified as right handers.Hearing acuity was determined by a Tegnerscreening audiometer. All patients had normalhearing, and patients with an imbalance inhearing between the ears of more than 5 dBwere not included in the study.

All patients were referred to surgerybecause of pronounced clinical symptoms.Headache was the most prominent, present inall the patients before operation. One patient(1 1) had had epilepsy for many years, and twopatients had recently developed epilepticseizures (5 and 6). One of these (5) alsoshowed a slight right sided hemiparesisalthough without dysphasia. Another patient(3) had had a transient, slight right sidedhemiparesis four months before the operation.None of the remaining patients had experi-enced any gross neurological disturbance. In

one patient (4) with a recent history of headtrauma, a subchronic subdural effusion/haematoma overlying the cyst was drainedseparately before the preoperative dichotic lis-tening tests and the shunt operation thateventually drained the cyst.

CONTROL GROUPSSeven patients with arachnoid cysts of theright temporal fossa served as a control groupfor the preoperative scores. As only three ofthese patients were operated on statisticalanalyses of the effects of decompressivesurgery in patients with right sided cysts couldnot be performed.

Healthy subjects (age <50 years) alsoserved as normal controls in both the dichoticmemory (n = 32), and the dichotic listening(n = 52) tests.

DICHOTIC LISTENING STIMULUS MATERIALSAND TEST PROCEDURESWe used a modified version of the dichoticmemory test described by Christianson et al.30Three different series of 10 nouns were pre-sented two seconds apart to the right or theleft ear, with the same nouns played simulta-neously backwards to the other ear. Thepatients were asked to recall all items remem-bered for 30 seconds after each series of 10words. The results were averaged over thethree series, the maximum score for each earthus being 10. To control for possible learningeffects due to repeated presentations of thedichotic memory test, three parallel versionsof the test were used. There might, however,still be some learning of the test situationitself. To control for this possibility, sevenpatients with left sided cysts and three with asimilar cyst on the right side were tested (pre-operatively) more than once.

Preoperatively, each patient was tested atleast once. Only the score from the first pre-operative test was used as the preoperativescore in the statistical analyses. For seven ofthe patients operated on under local anaesthe-sia, the immediate postoperative conditionwas so good that it allowed systematic collec-tion of data two or three times during the firstpostoperative week. The first of these postop-erative tests took place four hours after theoperation, except in patient 1 who was testedafter 24 hours. The immediate postoperativecondition did not allow a similar systematictesting during the first postoperative week inthe patients operated on with a full cran-iotomy under general anaesthesia. All thepatients were tested when returning to thehospital for the routine follow up three to sixmonths postoperatively.

In the dichotic memory test, the 32 controlsubjects showed equal recalls from the rightand left ears.The overall memory performance (mean of

the recalls from the right ear and the left ear)were also calculated for analyses between theexperimental group and the two controlgroups, and between the results before andafter operation in the experimental group.The patients were also examined with a

294 on N

ovember 5, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.59.3.293 on 1 S

eptember 1995. D

ownloaded from

Arachnoid cysts of the left temporalfossa: impaired preoperative cognition and postoperative improvement

dichotic test emphasising verbal perception,described in detail elsewhere.26 In this dichoticlistening test, the stimuli consisted of the sixstop consonants b, d, g, p, t, k, all paired withthe vowel a to form six consonant-vowel (CV)syllables (ba, da, ga, etc). The syllables werepaired with each other for all possible combi-nations, thus yielding 36 dichotic pairs. Thedichotic tape consisted of three lists of 36dichotic randomly ordered CV pairs each.Each syllable had a duration of 320 ms, andsynchronisation of onset between channelswas performed for both the consonant andvowel segment onsets. The interstimulusinterval was 4 ± 1 s.

In the dichotic listening test, the patientswere tested in three different conditions. Inthe non-forced (NF) condition they were sim-ply instructed to report freely all the syllablesthey heard. In the forced right (FR) or forcedleft (FL) conditions, they were instructed toattend to and report only what they heard inthe right or left ear."2The stimuli for the dichotic listening and

memory tests were played to the patient from aminicassette player through plug in type ear-phones. The intensity of the output from theearphones were on average 75 dBA (repeatedmeasurements) when tested with a soundlevel meter.To facilitate comparisons with other stud-

ies, the raw scores were transformed to per-centage scores. To evaluate any change in earadvantage for both the dichotic listening andthe dichotic memory tests, a laterality indexwas calculated according to the formula:

(RE-LE)/(RE+LE) x 100

where RE and LE reflect the recalls from theright and the left ears respectively. This yieldsa positive score for right ear advantage (REA),a negative score for a left ear advantage(LEA), and zero for no ear advantage (NEA).The laterality index compensates for individ-ual differences in overall performance.

Table 1 Data from 13 patients treated, with clinical improvement, for arachnoid cysts inthe left temporalfossa showing the size (type) of the cyst, the type of operation, andpreoperative and postoperative scores for correct recallfrom right (RE) and left (LE) ear inthe dichotic memory task

Preoperative Follow upPatient No Cyst type* Operation RE LE RE LE

1 II-III Shunt 3-0 < 5 0 (4 0) 5-0 > 4-3 (4-7)2 II Shunt 3-0 < 4-0 (3 5) 4 0 - 4-0 (4 0)3 III Shunt 4-3 < 5-3 (4 8) 6-0 > 5-3 (5-7)4 III Shunt 4-3 < 6-3 (5-3) 5-0 < 5-7 (5 3)5t II-III Shunt 1-3 < 4-7 (3-0) 6-0 < 6-7 (6 3)6 I-II Shunt 3-5 > 3-2 (3 3) 5 0 > 4 0 (4 5)7 II-III Craniotomy 7-7 > 6-3 (7 0) 8-7 > 7-3 (8-0)8 III Shunt 1-3 < 4-7 (3 0) 5-7 > 4-3 (5-0)9 II Craniotomy 1-7 < 3-7 (2-7) 3-3 < 5-3 (4-3)10 II-III Craniotomy 4 0 < 5 0 (4-5) 5-7 < 6-7 (6-2)11 II Craniotomy 5 0 > 4-3 (4 7) 6-3 > 4-3 (5 3)12 I-II Shunt 4-0 < 4-3 (4 2) 5-7 < 6-7 (6 2)13 II Craniotomy 2-3 < 3-3 (2 8) 5-3 > 3.3 (4 3)

Total 3-5 < 4-6 (4-1) 5-5 > 5-2 (5 4)

Scores in parenthesis indicate the overall performance as the means of right and left ear recalls.*According to the classification of Galassi et al of middle fossa arachnoid cysts, type I is thesmallest, situated entirely within the sylvian fissure. Type III is the largest cyst type, alsoextending over the cerebral convexity.39tLost to three to six months follow up because he developed a left subdural chronic haematomatwo months after the shunt operation. The data listed under follow up for this patient are there-fore those obtained in a test two days after the evacuation of this haematoma through a burrhole.

SURGERYFive of the patients underwent a craniotomywith fenestration of the cyst to the basal cis-terns during general anaesthesia. The remain-ing eight patients were operated on with aninternal shunt procedure under local anaes-thesia. This procedure and the clinical resultswill be reported in detail elsewhere." Briefly, asmall craniectomy was made and the wall ofthe cyst and the adjacent cortex were exposed.The cyst membrane was opened at the edge,and an angled Holter ventricular catheterinserted into the cavity. The distal end of thecatheter was placed in the subdural spaceoverlying normal cortex, thus allowing thecyst fluid to be drained into the subduralcompartment. This procedure reduced oreliminated the symptoms in seven patients,and proved sufficient as the only treatment insix. In one patient (5) with a large cyst, theinternal shunt operation caused a near com-plete collapse of the cyst, followed by consid-erable clinical improvement. This collapsewas associated with the development of a sub-dural haematoma before the planned followup. The follow up data presented for thispatient were therefore obtained after removalof this haematoma. In the last patient (3), theoperation induced a contralateral, expandingsubdural effusion. The shunt was thereforetransformed to a conventional cystoperitonealshunt, resulting in an immediate improve-ment. The postoperative scores reported forthis patient are therefore those obtained afterthe shunt revision.

All the patients experienced a postoperativeclinical improvement, and CT at the three tosix months follow up showed reduction of thecyst volumes in all the patients except two (3and 6).

ResultsMEMORYTable 1 shows the performance in thedichotic memory test. Ten of the 13 patientsshowed better recall from the left ear (LE)than from the right ear (RE) in the preopera-tive dichotic memory test, with a significantleft ear advantage (LEA) for the group as awhole (RE 3.5 v LE 4-6, t(12) = 2.84, P <0-01). These preoperative scores differed sig-nificantly from the scores obtained in the nor-mal control group (n = 32; RE 5.9 v LE 5-9),both with respect to overall memory perfor-mance (4 1 v 5-9 for the patient and controlgroups respectively, t(43) = 3-65, p < 0-01),and the presence of an LEA in the patientgroup (t(43) = 2 01, P < 0-01).The preoperative memory ear advantage

scores in the 13 patients with left sided cysts(table 1) were also significantly different fromthose of the seven control patients with cystsin the right middle fossa, who as a groupexhibited an REA (RE 3.9 v LE 3-5; t(18) =

2-69, P < 0-05).At follow up, the preoperative LEA of the

experimental group disappeared for all com-parisons, with an improvement of the right earperformance in all the patients (table 1). The

295 on N

ovember 5, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.59.3.293 on 1 S

eptember 1995. D

ownloaded from

Wester, Hugdahl

Table 2 Dichotic memory data (raw scores) from seven local anaesthesia patients wherethe immediate postoperative condition was so good that it allowed early (four hours andthree to seven days) postoperative testing

Preoperative 4 hours 3-7 days Follow upPatient No RE LE RE LE RE LE RE LE

1 30 < 50 (40) 30 <40 (35)* 50> 27 (38) 50 > 43 (47)2 30 <40 (35) 3-3<37 (35) 40<4-3 (42) 40 -40 (40)3 4-3 < 5-3 (48) 57 > 43 (50) 53 > 47 (50) 6-0> 53 (57)4 4-3 < 6-3 (5-3) 5-0 < 5-3 (5 2) 6-0 > 5-7 (5-8) 5 0 < 5-7 (5 3)5 1-3 < 4-7 (3-0) 5-7 > 4-7 (5 2) 7-7 > 6-7 (7-2) 6-0 < 6-7 (6 3)6 35 > 32 (33) 43 > 30 (37) 40 > 3.3 (37) 50 > 40 (45)

12 4 0 < 4-3 (4 2) 5-7 > 3.3 (4-5) 4-7 < 5-3 (5-0) 5-7 < 6-7 (6-2)

Total 3-3 < 4-7 (4 0) 4 7 > 4 0 (4 4) 5-2 > 4.7 (4 9) 5-2 - 5-2 (5 2)

*Results from 24 hour test, as four hour test was not performed.

difference between the laterality index beforeand after operation was significant (- 11 1 v2-8, t(12) = 3.28, P < 005).The patients also improved their overall

memory performance postoperatively. At thetime of follow up, 12 patients had improvedtheir overall performance, whereas itremained unchanged in the last patient. Thegroup average increased from a preoperativescore of 4-1 to 5A4 at follow up (table 1).Theincrease in overall performance was signifi-cant (t(12) = 5 54, P < 0 001).The seven patients operated on under local

anaesthesia showed rapid improvement dur-ing the first postoperative week (table 2).There was a significant improvement of theoverall memory score from the preoperativetest to the postoperative tests (F(3,18) =4 70, P < 0 05). Further tests (Newman-Keuls) showed that the significant improve-ment occurred three to seven days after theoperation, and at the follow up control (allP < 0 05).

There was also a rapid postoperativechange in the ear advantage in these patients(table 2). The ear advantage in the memorytest changed significantly from a preoperativeLEA to an REA during the first two postoper-ative tests (at four hours and seven days) and ano ear advantage (NEA) at follow up werealso significant (F(3,18) = 4-83, P < 0 05).The Newman-Keuls test showed that thechanges were significant for all three compar-isons (all P < 0 05). These comparisons wereperformed on laterality index scores asdescribed. In the 10 patients who were testedpreoperatively more than once to control forpossible learning effects of repeated testing,

Table 3 Individual dichotic memory scores from the postoperative period after thefirst,unsuccessful, operation in patient 3

Postoperative

Preoperative 4 hours 24 hours Day 6RE LE RE LE RE LE RE LE

4-3 < 5-3 (4-8) 5-7 > 4-3 (5 0) 5-7 > 4-7 (5-2) 3-3 < 4-0 (3-7)

Note the rapid nornalisation of the response pattern and overall performance (in parentheses) thatwas evident four hours after the operation. The right ear superiority was also present the nextday but disappeared on the sixth postoperative day, indicating the shunt failure that was verifiedlater. The immediate postoperative normalisation possibly reflects a temporarily lowered intra-cystic pressure after the first operation.

there was no significant difference betweenthe results from the first and last tests. Thiswas true both for overall memory perfor-mance and ear advantage, and also whenpatients with cysts on the left and the rightside were analysed separately.The dichotic memory test seemed sensitive

not only in disclosing preoperative impair-ments and postoperative improvements, butalso treatment failures. In one patient (3), theshunt operation caused an immediate andpronounced improvement in the dichoticmemory test performance. During the firstpostoperative week, it became clear that thisimprovement was only temporary, as the ini-tial postoperative REA disappeared at thesixth day (table 3). At this time there was noother indication of the shunt failure thatweeks later necessitated a shunt revision.

PERCEPTIONThe patients were also tested with thedichotic listening test, emphasising lateralityfor perception of dichotically presented CVsyllables. Table 4 summarises the results.Preoperatively, the group displayed a weakand statistically non-significant LEA duringthe NF test. In normal controls, a significantREA is seen.26 The ear advantages varied con-siderably between the patients. The results inthe NF test differed, however, significantlyfrom those of the normal control group(n = 52) for the laterality index scores as wellas total number of correctly perceived sylla-bles. t Tests showed both the comparisons tobe significant (t(63) = 8-79, P < 0 001 andt(63) = 6-39, P < 0 001 respectively).The preoperative performance during the

FR and FL tests were below normal on bothsides.32 As a group, the patients were able todirect their attention to and increase theirresponses from the right and the left ear.Their scores were significantly below those ofthe normal control group, however, withrespect to both the laterality indices and theabsolute number of correctly perceived sylla-bles (t(63) = 4-27 and 3-16 for the FR and FLconditions respectively; P < 0.01).The cyst operations did not change the

non-forced dichotic listening pattern in anysignificant way, and the distinct REA seen innormal subjects was not seen in the postoper-ative controls either. In the forced attentiontests, the operation caused significantchanges. In the FR test at follow up, thepatients displayed the expected superiority(advantage) from the right ear (to whichattention was directed). The scores from thatear were significantly higher than the scoresobtained in the preoperative FR and the post-operative NF tests, and the scores were nolonger different from those of the normal con-trol group in the similar (FR) situation. Thesame general response pattern (with an LEA)was seen in the FL condition. Comparing thepostoperative laterality indices with the preop-erative indices for each of the two forced lis-tening conditions (FR, FL), showedsignificant changes (t(12) = 2-15, P < 0-05,and 1-83, P < 0-05, respectively).

296 on N

ovember 5, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.59.3.293 on 1 S

eptember 1995. D

ownloaded from

Arachnoid cysts of the left temporalfossa: impaired preoperative cognition and postoperative improvement

Table 4 Data (% correct reports) from the dichotic listening test, coUected in three different test situations

Non-forced (NF) Forced nght (FR) Forced left (FL)

Preoperative Follow up Preoperative Follow up Preoperative Follow upPatient No RE LE RE LE RE LE RE LE RE LE RE LE

1 47 > 43 47 > 40 63 > 17 63 > 23 50 > 40 40 < 472 57>40 33<67 67>23 70>30 27<60 10<903 43>50 47<50 40<47 60>37 60>40 50>434 27 < 73 47 - 47 47 > 27 67 > 7 13 < 47 27 < 605 53>40 50>47 37<53 57>33 40<57 23<706 63>33 47>40 67> 13 60>23 60>40 30<577 43-43 40<53 73> 17 80> 7 17<73 7<608 23 < 40 47 > 37 27 < 47 43 - 43 27 < 43 43 < 479 23<30 17- 17 30-30 27>23 13<33 13<3710 30<40 43>37 50>20 60>20 23<43 27<5711 30>23 30-30 47> 17 47>27 23<33 33<3712 20<37 57>23 20<37 63>20 27<33 30<5713 50 - 50 40 < 60 50 > 40 67 > 30 10 < 67 17 < 83

Total 39 < 42 42 - 42 47 > 30 59 > 25 30 < 47 27 < 57Controls 53 > 41 62 > 26 30 < 56(n = 52)

In the non-forced condition, the patients were instructed to report freely the syllables heard in either ear. In the forced right orforced left conditions, they were instructed to pay attention to, and report only the syllables presented to the right or the left ear. Notethe improvement in the forced conditions.

DiscussionIn the present study, dichotic listening tech-niques emphasising memory and perceptionshowed lateralised cognitive deficits inpatients with arachnoid cysts of the left mid-dle fossa. The impairments affected memoryand the ability to direct attention in an audi-tory, perceptual task, and disappeared afterdecompressive cyst surgery. Normalisation ofthe memory functions occurred within hoursor days.

DICHOTIC LISTENING AND MEMORYThese non-invasive techniques have previ-ously proved sensitive in reflecting discretechanges in language functions produced byelectrical stimulation of the brain,2728 inpatients with localised brain damage,'0'3'8and in recovery of language functions after astroke.29The clinical and behavioural implications

of the cognitive impairments reported hereremain uncertain, but they may mirror deficitsthat are of importance to the patient. In thiscontext it is interesting to note that none ofour patients had experienced any gross prob-lems with speech or memory; neither weresuch deficits disclosed by neurological exami-nation before surgery.The present results thus indicate that later-

alised neuropsychological tests may have thepotential to disclose subtle, subclinicalcognitive impairments in patients with lesionsin the region of the left sylvian fissure. Someearlier studies have failed to show cognitiveimpairment in patients with arachnoidcysts,'524 whereas Lang et al23 found learningand memory deficits for hemisphere specificmaterials in their patients. One reason for thecontradictory results may be that the neuro-psychological tests used in these studiesdiffered, and that some of them possiblylacked the appropriate specificity for thehemisphere or brain region affected by thecyst. Our tests were chosen specifically toshow dysfunction of temporal areas in the leftand the right hemispheres. Cognitive func-tions other than memory and perception mayalso be impaired in these patients. For lesions

with other locations in the left or the righthemisphere, different tests must be selected.Normal adults have the ability to direct

their auditory attention to any of the two sidesin the dichotic listening test.'2 Our patientslacked this ability preoperatively and regainedit after surgery, indicating that cerebral struc-tures close to the left sylvian fissure are instru-mental in an auditory attention mechanism,and not only important for perception fromthe contralateral right ear. These findings maywell be explained within Kimura's model fordichotic listening.35 In her model, auditoryinformation from the left ear reaches the lefttemporal lobe mainly via the right temporallobe and the corpus callosum, thus beingweakened and delayed compared with themore direct input from the right ear. Thelesions in our patients may have disturbed thenormal ability to enhance the weakened anddelayed information from the ipsilateral leftear, and in some cases also the perceptionfrom the contralateral right ear.

ARACHNOID CYSTS AND COGNITIVEIMPAIRMENTArachnoid cysts are space occupying intracra-nial lesions, and the symptoms they cause areoften surprisingly moderate considering theirsubstantial volumes. One reasonable explana-tion of the stable appearance and relativepaucity of symptoms may be that the intra-cystic and intracranial pressures are onlymoderately raised, a common finding duringoperation. The pressure exerted on neigh-bouring cerebral structures is therefore proba-bly also moderate, although the physicaldisplacement of the same structures may bemassive. Whether the neurological and neu-ropsychological deficits caused by arachnoidcysts are precipitated by an increased tissuepressure or the displacement of brain tissue,or a combination of these or other factors,remain unsolved.

There is every reason to assume thatarachnoid cysts in adults have affected thesurrounding cerebral structures for manyyears. It would therefore be a reasonableassumption that the associated brain damage

297 on N

ovember 5, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.59.3.293 on 1 S

eptember 1995. D

ownloaded from

Wester, Hugdahl

and concomitant symptoms might be perma-nent, and refractory to any form of surgicalcorrection. The rapid restoration of cognitivefunctions in our patients after decompressivesurgery shows that this is not the case. Thisindicates that middle fossa arachnoid cystsmay cause a reversible suppression of neu-ronal function rather than a permanentdestruction of cerebral tissue.The present results may have implications

for future preoperative considerations inpatients with arachnoid cysts. The cognitiveimprovement after decompression in thesepatients is so important that it may lower thethreshold for operative treatment, even inpatients with moderate clinical symptoms.Consequently, specific neuropsychologicaltest batteries that tap lateralised functionsshould be developed for clinical use. Our testswere sensitive to, and therefore suitable forprobing asymmetric hemisphere functionsrelated to memory and language. For cystswith other locations (frontal, parietal, oroccipital), tests should be added for temporalsequencing, neglect, and visuospatial func-tions.

We are indebted to Janniche Alvaer and Cathrine Hovland forassistance in the collection of data. The research wassupported by grants from the Norwegian Medical ResearchCouncil (NAVF-RMF) and Nansen-fondet to KW and KH.

1 Starkman SP, Brown TC, Linell EA. Cerebral arachnoidcysts. J7Neuropathol Exp Neurol 1958; 17:484-500.

2 Rengachary SS, Watanabe I. Ultrastructure and patho-genesis of intracranial arachnoid cysts. J Neuropathol ExpNeurol 1981;40:61-83.

3 Geissinger JD, Kohler WC, Robinson BW, Davis FM.Arachnoid cysts of the middle cranial fossa: surgicalconsiderations. Surg Neurol 1978;10:27-33.

4 Kumagai M, Sakai N, Yamada H, Shinoda J, NakashimaT, Iwama T, Ando T. Postnatal development andenlargement of primary middle cranial fossa arachnoidcyst recognized on repeat CT scans. Child's Nerv Syst1986;2:211-5.

5 Rengachary SS. Intracranial arachnoid and ependymalcysts. In: Wilkins RH, Rengachary SS, eds. Neurosurgery,Vol III. New York: McGraw-Hill, 2160-72.

6 Beltramello A, Mazza C. Spontaneous disappearance of alarge middle fossa arachnoid cyst. Surg Neurol 1985;24:181-3.

7 Yamanouchi Y, Someda K, Oka N. (1986) Spontaneousdisappearance of middle fossa arachnoid cyst after headinjury. Child's Nerv Syst 1986;2:40-3.

8 Inoue T, Matsushima T, Tashima S, Fukui M, Hasuo K.Spontaneous disappearance of a middle fossa arachnoidcyst associated with subdural hematoma. Surg Neurol1987;28:447-50.

9 Weber R, Voit T, Lumenta C, Lenard HG. Spontaneousregression of a temporal arachnoid cyst. Child's Nerv Syst199 1;7:414-5.

10 Wester K, Gilhus NE, Hugdahl K, Larsen JL.Spontaneous disappearance of an arachnoid cyst in themiddle intracranial fossa. Neurology 1991 ;41:1524-6.

11 Becker T, Wagner M, Hofman E, Warmuth-Metz M,Nadjmi M. Do arachnoid cysts grow? A retrospectiveCT volumetric study. Neuroradiology 1991 ;33:341-5.

12 Mayr U, Aichner F, Bauer G, Mohsenipour I, Pallua A.Supratentorial extracerebral cysts of the middle cranialfossa. Neurochirurgia (Stuttg) 1982;25:51-6.

13 Cilluffo JM, Onofrio BM, Miller RH. The diagnosis and

surgical treatment of intracranial arachnoid cysts. ActaNeurochir 1983;67:215-29.

14 Gandy SE, Heier LA. Clinical and magnetic resonancefeature of primary intracranial arachnoid cysts. AnnNeurol 1987;21:342-8.

15 Kunz U, Ruckert N, Tagert J, Dietz H. Clinical and neu-ropsychological results after operative and conservativetreatment of arachnoidal cysts of the perisylvian region.Acta Neurochir 1988;Suppl42:216-20.

16 Dei-Anang K, Voth D. Cerebral arachnoid cyst: a lesion ofthe child's brain. Neurosurg Rev 1989:12:59-62.

17 Robertson SJ, Wolpert SM, Runge VM. MR imaging ofmiddle fossa arachnoid cysts: temporal lobe agenesissyndrome revisited. AJ7NR Am I Neuroradiol 1989;10:1007-10.

18 Holst S. Congenital intracranial arachnoidal cysts. Casereports and discussion of the pathogenesis. J Oslo CityHosp 1965;15:113-20.

19 Aicardi J, Bauman F. Supratentorial extracerebral cysts ininfants and children. J Neurol Neurosurg Psychiatry 1975;38:57-68.

20 Choux M, Raybaud C, Pinsard N, Hassoun J, GambarelliG. Intracranial supratentorial cysts in children excludingtumor and parasitic cysts. Child's Brain 1978;4:15-32.

21 Lodrini S, Lasio G, Fornari M, Miglivacca F. Treatmentof supratentorial primary arachnoid cysts. Acta Neurochir1985;76: 105-10.

22 Marinov M, Undjian S, Wetzka P. An evaluation of thesurgical treatment of intracranial arachnoid cysts in chil-dren. Child's Nerv Syst 1989;5:177-83.

23 Lang W, Lang M, Kornhuber A, Gallwitz A, Kriebel J.Neuropsychological and neuroendocrinological distur-bances associated with extracerebral cysts of the anteriorand middle cranial fossa. Eur Arch Psychiatr Neurol Sci1985;235:38-41.

24 Gallassi R, Ciardulli C, Ferrara R, Lorusso S, Galassi E,Lugaresi E. Asymptomatic large cyst of the middlecranial fossa. A clinical and neuropsychological study.Eur Neurol 1985;24:140-4.

25 Wester K. Gender distribution and sidedness of middlefossa arachnoid cysts: a review of cases diagnosed withcomputed imaging. Neurosurgery 1992;31:940-44.

26 Hugdahl K, ed. Handbook ofdichotic listening: theory, methodsand research. New York: Wiley and Sons 1988.

27 Ojemann GA. Enhancement of memory with humanventrolateral thalamic stimulation. Effect evident on adichotic listening task. Appl Neurophysiol 1985;48:212-5.

28 Hugdahl K, Wester K, Asbjornsen A. The role of the leftand right thalamus in language asymmetry: dichoticlistening in Parkinson patients undergoing stereotacticthalamotomy. Brain Lang 1990;39:1-13.

29 Hugdahl K, Wester K, Asbj0rnsen A. Dichotic listening inan aphasic male patient after a hemorrhage in the leftfronto-parietal region. Intern J Neuroscience 1990;54:139-46.

30 Christianson SA, Nilsson LG, Silfvenius H. Initialmemory deficits and subsequent recovery in two cases ofhead trauma. ScandJ Psychol 1987;28:267-80.

31 Raczkowsky D, Kalat JW, Nebes R. Reliability andvalidity of some handedness questionnaire items.Neuropsychologia 1974;12:43-7.

32 Hugdahl K, Andersson L. The "forced-attentionparadigm" in dichotic listening to CV-syllables: a com-

parison between adults and children. Cortex 1986;22:417-32.

33 Wester K. Arachnoid cysts in adults. Experience withinternal shunts to the subdural compartment. SurgNeurol 1995 (in press).

34 Kimura D. Some effects of temporal lobe damage on

auditory perception. Can J Psychol 1961;15:156-65.35 Kimura D. Functional asymmetry of the brain in dichotic

listening. Cortex 1967;3:163-78.36 Eslinger PJ, Damasio H. Anatomical correlates of

paradoxic ear extinction. In: Hugdahl K, ed. Handbookof dichotic listening: theory, methods and research. NewYork: Wiley and Sons, 1988:139-60.

37 Hugdahl K, Wester K. Dichotic listening studies of hemi-spheric asymmetry in brain damaged patients. Intern J

Neuroscience 1992;63: 17-29.38 Hugdahl K, Wester K, Asbjornsen A. Auditory neglect

after right frontal lobe and right pulvinar thalamiclesions. Brain Lang 1991;41:465-73.

39 Galassi E, Tognetti F, Gaist G, Fagioli L, Frank F, FrankG. CT scan and metrizamide CT cisternography inarachnoid cysts of the middle cranial fossa: classificationand pathophysiological aspects. Surg Neurol 1982;17:363-9.

298 on N

ovember 5, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.59.3.293 on 1 S

eptember 1995. D

ownloaded from