aetiologies - bmj · 5-18, p

48ournal ofNeurology, Neurosurgery, and Psychiatry 1996;61:478-484

Cognitive impairment after acute encephalitis:comparison of herpes simplex and otheraetiologies

Laura Hokkanen, Eria Poutiainen, Leena Valanne, Oili Salonen, M livanainen,Jyrki Launes

Department ofNeurologyL HokkanenE PoutiainenJ LaunesDepartment ofRadiologyL ValanneO SalonenDepartment of ChildNeurology, Universityof Helsinki, FinlandM IivanainenCorrespondence to:Laura Hokkanen,Department of Neurology,University of Helsinki,Haartmaninkatu 4,00290 Helsinki, Finland.Received 14 February 1996and in revised form20 May 1996Accepted 5 July 1996

AbstractObjective-To compare the cognitivedefects after acute acyclovir treated herpessimplex encephalitis with those after othertypes of acute encephalitis.Methods-Seventy seven consecutivepatients between 1985 and 1995 and 29normal controls were studied. Of the 77patients without concomitant neurologicalconditions, 17 had herpes simplex, onevirus encephalitis (HSVE group), 27 hadsome other identified aetiology (non-HSVE group), and in 33 patients the causewas unknown. Acyclovir treatment wasstarted less than four days after the firstmental symptoms in 12 of 17 patients withHSVE. A thorough neuropsychologicalassessment was carried out about onemonth after the onset.Results-The HSVE group had deficits inverbal memory, verbal-semantic func-tions, and visuoperceptual functions moreoften than the non-HSVE group. The riskfor cognitive defects was twofold to four-fold in the patients with HSVE comparedwith the non-HSVE patients. Two (12%) ofthe patients with HSVE and 12 (44%) of thenon-HSVE patients were cognitivelyintact. Six patients with HSVE (46%) and17 (89%) non-HSVE patients laterreturned to work. The lesions on CT orMRI were bilateral only in one patient withHSVE. The defects in the three patientswith adenovirus infection were severe andresembled the amnesia after HSVE.Cognitive impairment, not previouslyreported, was found in encephalitis afterrotavirus infection and epidemicnephropathy.Conclusion-The recovery in the HSVEgroup was better than expected based onthe medical literature. On the other handthere were surprisingly severe cognitivedefects in encephalitis after other viruses.With early acyclovir treatment patientswith the least severe HSVE were equiva-lent to those with non-HSV encephalitiswith good outcome whereas those with themost severe non-HSV encephalitis wereequivalent to those with HSVE with pooroutcome.

brain parenchyma, in most patients caused byviral agents, although bacterial, other micro-bial, and postinfectious forms are known. Inthe United States 20 000 cases occur eachyear.' In Finland the prevalence of menin-goencephalitis and encephalitis is estimated tobe 3-5 cases per 100 000.2 The aetiology maybe difficult to determine and one third to twothirds of the patients lack specific microbialconfirmation.3 4 Worldwide, the arbovirusesare the most frequent cause of epidemics ofencephalitis. In western Europe sporadicinfections account for most cases, and themost common single aetiology is herpes sim-plex 1 virus. In Sweden the incidence of herpessimplex encephalitis (HSVE) is at least 2-3cases per million inhabitants per year.6

Mental changes as an initial consequence ofencephalitis have been reported in all epidemi-ological studies.478 The cognitive defects thatpersist after the acute stage have seldom beenreported in detail. In studies from the UnitedStates, the United Kingdom, and Germanythe frequency of residual neurological disabil-ity has been 10%-30%, including aphasia,mental deterioration, and paralysis.3479Neuropsychological investigation was not usedin any of these studies.

It has been reported that HSVE results inbilateral temporal lobe damage'O1' and persis-tent memory defect.'2 'I The outcome has beenconsidered poor; death or severe disablementand dependency being the common endpoints.'4 15 Most of the reported cases originatefrom the 1970s and early 1980s and since thenacyclovir treatment has changed the prognosisof this disease.6 Early treatment may have aneffect on the long term neuropsychologicalsymptomatology as well, but only small serieshave so far been published.'6 18

Even fewer reports have described the neu-ropsychological findings in acute encephalitisof other aetiologies. In the United States,HSVE accounts for 5%-10% of the annualcases of encephalitis.' In the German series of53 patients with acute encephalitis, the mostcommon aetiology, HSVE, was identified innine patients, but 83% were either caused byother specified agents or were unidentified.4We studied the cognitive sequelae in 77

patients with acute encephalitis between 1985and 1995.

(_ Neurol Neurosurg Psychiatry 1996;61:478-484)

Keywords: encephalitis; cognitive performance; herpessimplex

Acute encephalitis is an inflammation of the

Material and methodsPATIENT GROUPS AND CONTROLSBetween 1 January 1985 and 31 December1994, 92 consecutive adult patients under the

478

on March 26, 2021 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.61.5.478 on 1 N

ovember 1996. D

ownloaded from

http://jnnp.bmj.com/

Cognitive impairment after acute encephalitis: comparison of herpes simplex and other aetiologies

Table 1 Diagnostic methods for determining the specific aetiologies of 44 patients with encephalitisAetiology n CSF Serum Other

Herpes simplex 1 12 CF or EIA, CF or EIApositive PCR in one and pathological s/CSF

4 CF1 High ab titre CF

and pathological s/CSFVaricella zoster 4 CF or EIA High ab titre 3 with rash

3 CF or EIA 1 with rash2 High ab titre 2 with rash and

positive cultureInfluenza B 2 High ab titre High ab titre

and pathological s/CSF2 CF, EIA

Mpneumoniae 2 Mycoplasma RNA High ab titrein CSF and pathological s/CSF

Rubella 2 High ab titre,IgM positive

Tick borne encephalitis 2 High ab titre, High ab titreIgM positive

Adenovirus 1 EIA High ab titreand pathological s/CSF

1 High ab titreand pathological s/CSF

Adenovirus + borreliosis 1 High ab titre for both High ab titre for both,decline in ab titres

Rotavirus 2 High ab titre High ab titres,positive IgM orelectron microscopy

Herpes simplex 2 1 High ab titre Genital lesionsand positive culture

Cryptococcus maserans 1 Positive antigen testEpidemic nephropathy 1 High ab titres High ab titre

(Puumala virus) and pathological s/CSF

CF = fourfold increase in the antibody titre in paired samples using the complement fixation test; EIA = twofold increase in the anti-body titre in paired samples using the enzyme linked immunosorbent assay; PCR = polymerase chain reaction; ab = antibody;s/CSF = serum/CSF ratio.

age of 75 with acute encephalitis were referredfor neuropsychological examination at theDepartment of Neurology, Helsinki UniversityCentral Hospital. It is the only hospital with24 hour neurological emergency services in theprovince of Uusimaa (population 1 million),and thus the material includes nearly all adultpatients with serious CNS infections in thisarea during this time. The diagnosis ofencephalitis was based on the history, clinicalphysical and mental findings, and EEG sug-gesting involvement of the brain parenchyma,findings compatible with infection of the CNS,and CT excluding other causes. Fifteenpatients with alcohol misuse, or coexisting orprevious neurological disease were excluded.Thus 77 patients with acute encephalitis with-out concomitant conditions were analysed.The aetiology was established in 44 (57%).

The analyses were carried out at theDepartment of Virology, University ofHelsinki. Table 1 gives the methods for reach-ing specific diagnoses. The polymerase chainreaction (PCR) method for the detection ofherpes simplex virus-DNA was used in ninepatients during 1993-4. Seventeen patientshad herpes simplex 1 virus encephalitis(HSVE), nine had varicella zoster virusencephalitis (VZVE), and one had herpes sim-plex 2 virus (HSV-2) encephalitis. Seventeenhad some other identified non-herpetic aetiol-ogy as the underlying cause for the encephalitis.The patients were grouped as the HSVE group(n = 17), the non-HSVE group (n = 27),and the unspecified group (n = 33).

Follow up EEG and standard laboratorytests were performed on all patients.Neuroradiological follow up was performedusing CT in 18 patients and MRI in 31patients. For analysis, the CT and MRI werescored blindly by two neuroradiologists (OSand LV). In the first CT, which was per-

formed within the first three days in mostpatients, cortical or central atrophy was foundin 16 of 77 patients (20%). At follow up, CTor MRI disclosed newly developed focal atro-phy or focal lesions in 28 (36%) patients.

All patients received specific antibiotictreatment. Along with other medication, 73,including all patients with HSVE, were givenintravenous acyclovir (30 mg/kg per day) for amean of 11-2 (SD 4-1) days. One patient withinfluenza B, one with tick borne encephalitisvirus, one with epidemic nephropathy, andone with suspicion of tuberculous meningoen-cephalitis were treated accordingly, withoutacyclovir medication. In the HSVE group acy-clovir treatment was started on the same dayas the mental symptoms appeared in twopatients, on the next day in five patients, onday 2 in three patients, on day 3 in onepatient, on day 4 in one patient, and five ormore days after the onset in five patients.The duration of stay in hospital was 36-4

(SD 38 2), range 8-213 days. After that thepatients visited the hospital if necessary. Thefollow up extended up to the time of subse-quent return to work or a decision to retire,maximally one year.The performance of the patients in the neu-

ropsychological investigation was comparedwith that of a group of 29 nornmal controls.Twenty five of the controls had initially volun-teered for a study of quantitative EEG (Q-EEG) and were later willing to participatefurther. One control was excluded due to a his-tory of probable meningitis. The others had nomedical problems and their Q-EEG was nor-mal. Five of the controls were additional volun-teers who had no medical complaints andwhose neurological examination was normal.

Table 2 shows distributions of age, sex, andeducation for the HSVE, non-HSVE, and thecontrol groups. There was no difference in age

479



j.com/

J Neurol N


ovember 1996. D

ownloaded from


Hokkanen, Poutiainen, Valanne, Salonen, Iivanainen, Launes

Table 2 Sex, age, and education of the patients and the controls, and location offocalfindings in follow up CT or MRI in HSVE, non-HSVE, and control groups

HSVE non-HSVE Controls(n = 17) (n= 27) (n = 29)

Sex (MIF) 8/9 16/11 11/18Age (y): 50-2 (12 7) 38-7 (20 0) 44-4 (14-1)

Range 23-68 17-73 22-73Education (y) 10-9 (4 3) 11-2 (3 7) 12-9 (3 1)Focal lesions:None 3 23 -

Temporal left 7 -

Temporal right 6 1*Temporal bilateral 1 itOther - 1 left internal capsule4 -

1 left thalamus§

Values for age and education are means (SD).*Patient with varicellar zoster viral encephalitis.tPatient with adenoviral infection and borreliosis (patient 13).tPatient with varicellar zoster viral encephalitis.§Patient with Mycoplasma pneumoniae infection (patient 5).

or duration of education between the threegroups (analysis of variance, ANOVA). Therewas a difference in the Beck depression inven-tory (BDI) score at the time of the neuropsy-chological investigation (mean 7-2 (7-8) in theHSVE group, 6-9 (7 8) in the non-HSVEgroup, 2-1 (2-4) in the control group, F =5-18, P < 0 01). Pairwise post hoc analysesdisclosed a higher depression score in patientsthan in controls (P < 0-02, Tukey's honest sig-nificant difference), but no difference betweenthe HSVE and the non-HSVE groups. Therewas no difference between the HSVE andnon-HSVE groups in the frequency of brainatrophy or focal lesions found in the first CT(Pearson's x2 test). Fourteen patients withHSVE (82%) had focal lesions in the followup neuroradiology, whereas the correspondingfrequency for the non-HSVE group was 15%(Pearson's X2 = 19-7, P < 0-0001). Table 2gives the radiological findings.To rule out the possibility of error due to

selection, we also compared the patients withaetiologically specified encephalitis (bothHSVE and non-HSVE, n = 44) with thepatients with non-specified encephalitis (n =33). Age, duration of education, and depres-sion scores were similar in the two groups(Student's t test).

NEUROPSYCHOLOGICAL ASSESSMENTThe neuropsychological examination was car-ried out in a postacute stage, as soon as the ini-tial confusion had subsided and the patientcould cooperate adequately. This was a mean27-5 (22 8), range 5-128 days, after the onsetof symptoms. The controls underwent a similarneuropsychological examination.

Table 3 Comparison of mean psychometric test results of the HSVE group, the non-HSVE group, and the controls

HSVE non-HSVE Controls Significant differences(n = 17) (n =27) (n =29) (P<005)

VIQ 97-5 (25 7) 109-6 (16-8) 121 8 (12-5) HSVE v controlsRange 37-138 72-130 104-152 Non-HSVE v controls

PIQ 95 5 (21-3) 108-1 (15-2) 126-2 (13-2) HSVE v controlsRange 74-143 81-146 104-159 Non-HSVE v controls

Verbal memory 14-4 (9 3) 27-0 (6 0) 31 8 (4 5) HSVE v controlsRange 2-5-29-5 14-36 22 5-38 Non-HSVE v controls

HSVE v non-HSVEVisual memory 23-6 (8-9) 31 3 (4 9) 35-4 (3 5) HSVE v controls

Range 12-38 21-39 27-40 Non-HSVE v controlsHSVE v non-HSVE

Values in parentheses are SD.Significance measured by post hoc Tukey pairwise analysis.

The cognitive functions were evaluated inseven different areas. Verbal logical thinkingwas assessed using the Wechsler adult intelli-gence scale (WAIS) verbal intelligence quotient(VIQ)19 estimated by subtests information,arithmetic, and vocabulary or similarities.Visual logical thinking was assessed with theWAIS performance intelligence quotient (PIQ)estimated by subtests digit symbols, picturecompletion, and block design. Deficit was con-sidered to exist if the VIQ or the PIQ fell below95, the lower limit of the normal range inFinnish norms.20 Verbal memory was assessedwith the Wechsler memory scale (WMS)21 sub-tests logical memory and associative learning,in which the raw scores were added together.Visual memory was assessed by addingtogether the raw scores of the WMS visualreproduction and the Benton visual retentiontest.22 Verbal-semantic functions were assessedwith analysis of the overt speech and under-standing of the instructions, confrontationnaming of body parts and pictures of familiarobjects, and comprehension of sentences withcomplex semantic structures.23 24 Visuo-perceptual functions were assessed with tasksof copying a cube and a Greek cross, and ofdrawing clock hands. Voluntary motor func-tions were assessed with Luria based tasks ofcopying spatial movements and sequentialmovements with arms, hands, and fingers.24Deficits in verbal memory, visual memory, ver-bal-semantic, visuoperceptual, and voluntarymotor functions was considered to exist if theperformance in any of the relevant tests fellbelow the arbitrary cut off point, which wasdefined as the score two SD below the mean ofthe controls. The level of depression wasassessed with the short form of the Beckdepression inventory.25 Persistent psychiatricsymptoms and personality change wereassessed in interviews during the follow upvisits.The occurrence of neuropsychological find-

ings in the HSVE group (n = 17) was com-pared with that in the non-HSVE group (n =

27), and risk ratios as well as 95% confidenceintervals were calculated. The findings inpatients with non-herpetic encephalitis are alsogiven in detail. Cases of unspecified aetiologywere omitted from these analyses.

ResultsSignificant psychometric differences werefound between the HSVE, the non-HSVE,and the control groups (ANOVA: VIQ F (2,69) = 10-2, P < 0-001; PIQ F (2, 69) = 20 9,P < 0-001; verbal memory F (2, 66) = 38-8,P < 0-001; and visual memory F (2, 63) =

21-4, P < 0-001). In pairwise comparison(Tukey's honest significant difference,Spjotvoll/Stoline test), both the HSVE and thenon-HSVE group performed worse than thecontrols in all tests (P < 005). In the verbaland the visual memory the HSVE group wasmore impaired than the non-HSVE group(P < 0-001). Table 3 gives the neuropsycho-logical findings.

480



j.com/

J Neurol N


ovember 1996. D

ownloaded from



Table 4 Occurrence ofdifferent cognitive deficits of the HSVE and non-HSVE groups and the increased risk forcognitive defects in the HSVE group

HSVE (n 17) Non-HSVE (n =27) Risk ratioAreas of cognitive defect No (%/0) No (%) (95% CI) x2 (P value)

Verbal logical thinking 7 (41) 6 (22) 1-67 1 00Visual logical thinking 9 (53) 6 (22) 2-17 3 12Verbal memory 13 (76) 8 (30) 3-56 (1-4-8-9) 7-39 (P < 0-01)Visual memory 10 (59) 9 (33) 1-88 0-81Verbal-semantic 10 (59) 6 (22) 2 50 (1-1-5-8) 4-56 (P < 0 05)Visuoperceptual 14 (82) 11 (41) 3-55 (1-3-5-8) 5-76 (P < 0 05)Voluntary motor 3 (18) 3 (11) 1-36 0-03Any cognitive defects 15 (88) 15 (56) 3-50 3-74

Table 4 gives the frequencies of cognitivedeficits in HSVE and non-HSVE groups. TheHSVE group had a higher frequency of deficitsin all measures, and significant differenceswere found in verbal memory, verbal-semanticfunctions, and visuoperceptual functions. Therisk of defects in the HSVE group was twofoldto fourfold compared with the non-HSVEgroup (table 5). Two of the 17 patients withHSVE (12%), and 12 of 27 of the non-HSVEpatients (44%) were cognitively intact. Elevenpatients with HSVE (65%) had persistingmood or personality change as opposed tothree (I 1 %) non-HSVE patients (Pearson's x2= 13-8, P<0-001). These disturbancesincluded euphoria and manic behaviour,aggressiveness and irritability, and depressivemood. No Kluver-Bucy type of behaviour was

seen.

Table 5 presents the neuropsychologicalfindings of patients with non-herpetic aetiol-ogy. Neuropsychological test performance ofthree of the four patients with encephalitisafter an influenza B infection (patients 1-3),and both patients with Mycoplasma pneumoniae(patients 5-6), rubella (patients 7-8), andarbovirus (patients 9-10) encephalitis wererated intact. One patient with influenza Binfection had slightly lowered VIQ and mildinaccuracy in visuoperceptual performance.The two patients with adenovirus infectionand the one with simultaneous borrelia andadenovirus infection (patients 11-13) all had aclear memory defect, as well as circumscribedneuropsychological findings, including verbal-semantic or visuoperceptual difficulty. Onepatient with rotavirus infection (patient 14)was inaccurate in visuoperceptual functions,

the other (patient 15) had memory defects.Cryptococcal infection (patient 16) led to per-sistent epilepsy, slowness, general deteriora-tion, and retirement. Epidemic nephropathy(patient 17) resulted in visuoperceptual diffi-culty and hand dyspraxia. Findings forpatients with VZVE will be reported in detailseparately, and are not discussed here. Thepatient with HSV-2 had marginal difficulty inattention and mild inaccuracy in visuopercep-tual functions.Two patients died of cancer but no deaths

related to encephalitis occurred in this studypopulation during the follow up period.Twenty nine patients with a non-specifiedaetiology (total n = 33) were employed beforethe onset of illness. Of these, 23 (79%) wereable to return to their initial occupation.Thirty two patients with a specified aetiology(total n = 44) were employed before the onsetof illness, and of those, 23 (72%) were able toreturn to work. The percentage of patientswho were able to return to work was similar inthe specified and non-specified group(Pearson's x2 test). Six of the HSVE group(46% of the 13 who were employed) and 17 ofthe non-HSVE patients (89% of the 19 whowere employed) returned to work after illness,which means a poorer outcome for the HSVEpatients (Pearson's X2 = 7-2, P < 001). Inpatients with HSVE, the duration of sick leavewas 133-5 (80.1), range 63-251 days, and innon-HSVE patients 84-2 (67 3), range 13-271days, after the onset of symptoms (no differ-ence by Student's t test).No differences in any of the cognitive mea-

sures were found between patients with aspecified aetiology (both HSVE and non-

Table S Neuropsychologicalfindings in patients with non-herpetic encephalitisPatient Verbal Visual Verbal- Visuoper- VoluntaryNo Aetiology VIQ PIQ MQ memory* memory* semantic* ceptual* motor*

1 Influenza B 125 120 132 n n n n n2 Influenza B 113 118 123 n n n n n3 Influenza B 129 146 124 n n n n n4 InfluenzaB 93 114 118 n n n + n5 Mpneumoniae 101 104 99 n n n n n6 Mpneumoniae 106 95 107 n n n n n7 Rubella 122 116 118 n n n n n8 Rubella 103 106 101 n n n n n9 Arbovirus 130 112 150 n n n n n10 Arbovirus 113 132 112 n n n n n11 Adenovirus 77 82 83 n + n + +12 Adenovirus 107 105 84 + n + + n13 Adenovirust 110 106 81 + + + n n14 Rotavirus 123 106 106 n n n + n15 Rotavirus 116 116 88 + + n n n16 Cryptococcus 85 91 77 + + n n n17 Epidemic nephropathy 110 103 98 n + n + +

*The defined cut off point was 2SD below the mean of the controls. n = Normal range; tInfection in conjunction with borreliosis; + = Abnormal finding in any ofthe test measuring the function.

481



j.com/

J Neurol N


ovember 1996. D

ownloaded from



Table 6 Literature concerning adult cases of non-herpetic encephalitis

Aetiology (reference) Observations on cognition Usual outcome

Influenza57, 0 Impaired memory, Varies from fatal to completepersisting confusion recovery

M pneumoniae".s Confusion, psychotic Varies from fatal to completeepisodes recovery

Rubella", Confusion, aphasia, Varies from fatal to completeemotional liability recovery

Tick borne encephalitis"" 7 Confusion, speech defects, Complete recovery(Kumlinge virus) persistent fatigability,

one case of dementiaAdenovirus'1 5i Confusion, agitation, Good recovery

period of amnesiaRotavirus No reports on adultsCryptocuccosus 7' Confusion, personality Varies from fatal to complete

change, psychotic episodes, recoverymemory defect, dementia

Epidemic nephropathy77 77 No details Good recovery

HSVE, n = 44) and the patients with unspec-

ified encephalitis (n = 33) analysed either bygroup means or frequencies of deficits.

DiscussionThe cognitive deficits in patients with HSVEwere more frequent as well as more severe

than in those with non-HSVE. Intellectualfunctions and logical thinking were not statis-tically more impaired in patients with HSVE,but the HSVE group had more deficits inmemory functions and in verbal-semantic andvisuoperceptual functions. In the postacutestage, the risk for cognitive defects in HSVEwas twofold to fourfold compared with a non-

HSVE group. However, the variability was

high, as the 95% confidence intervals of therisk ratios ranged from 1 to 9. Some patientswith HSVE were cognitively intact. In a recent

study,'3 dense amnesia was found in six of 10patients with HSVE and a noticeable antero-

grade memory impairment in the four others.However, in that study memory impairmentwas an inclusion criteria, which explains theabundant findings. Only a few studies havedescribed mild forms of HSVE. 17 26

The neuropsychological assessment was

carried out about one month after the onset ofsymptoms. The cognitive status cannot beregarded as final at that stage, and thereforewe contined to follow up the patients to see

whether the patients later became employed or

retired. Almost half of the patients withHSVE, along with 89% of the non-HSVEpatients, were eventually able to return to

work. Although neuropsychological findingsand psychiatric symptoms may be moderate to

severe during the first weeks and months afterthe onset, improvement takes place within thefirst year and the outcome of HSVE may bemore favourable than has been reported previ-ously.The good outcome in HSVE is also

reflected in the fact that the radiological find-ings in our patients were mostly unilateral.This is by contrast with previous neuropatho-logical'0 as well as MRI'3 findings, which sug-

gested that the damage in HSVE is nearlyalways bilateral. Progression of the lesion fromunilateral to bilateral used to be common,27

but after the introduction of acyclovir treat-

ment such advanced cases have become rarer.

Unilateral findings, resulting in only mild neu-

ropsychological sequelae, have been reportedin other studies.'82829 The neuropsychologicalprofile in some of our patients suggested bilat-eral damage, which was not detected in neuro-radiological imaging. Repeated MRI was notobtained in all patients, and minor lesions mayhave gone unseen. However, pronouncedasymmetry was evident.We think that the good outcome in our

series is due to the fact that acyclovir medica-tion was initiated very early, during the firstfour days of the onset in 12 of 17 patients. Wedid not wait for the HSVE diagnosis to be con-firmed, but started medication at the meresuspicion of HSVE. Treatment latencies ofless than four days have been reported to cor-relate with a favourable outcome.3I I'

We used clinically available standard CSFand serum laboratory tests for establishing theaetiological diagnosis. The retrospective diag-nosis of HSVE was based on the rise in theantibody titres in the CSF and a pathologicalserum:CSF antibody ratio. Simultaneous risein serum antibody titres was considered con-firmatory. The same methods performed bythe same laboratory have been used in previ-ous epidemiological studies.92 3 These meth-ods have also been widely used by others.34 38

The problems in laboratory verification ofHSVE have been known for decades39 andbiopsy has been suggested as the only reliablemethod.' 4( As acyclovir treatment is safe andcontroversy over the risks and benefits ofbiopsy exist,41 42 biopsy samples were not takenin our patients. Recently, the PCR techniquehas been proved reliable in the diagnosis43'47and this method was used as soon as it becameavailable to us.

This study did not include all patientstreated for acute encephalitis in theDepartment of Neurology. It consisted ofthose referred for neuropsychological exami-nation. Patients who died in the early phasewere not included. On the other hand, a fewpatients without specific aetiological diagnosiswho were discharged rapidly due to mildcourse of disease were also left out. Therefore,the possible bias in the inclusion of cases isbidirectional and probably does not distort theresults. Also the fact that only specified aeti-ologies were included in the analyses mightproduce error. However, the patients with aspecified encephalitis had neither more norless cognitive defects than the patients withnon-specified encephalitis, and the percentageof patients later returning to work was similar inthese groups.

Table 6 gives the relevant medical literatureconcerning non-herpetic aetiologies. Reportson the cognitive status are scant and often lim-ited to the acute stage. Neuropsychologicalassessment has usually not been made.Outcome also varies, and whereas reportsfrom earlier decades often present fatal cases,more recent reports include total recoveries. Inour series almost all patients with influenza B,Mycoplasma pneumoniae, rubella, and TBEvirus encephalitis had no cognitive sequelae.One patient with influenza B virus and bothpatients with rotavirus encephalitis had resid-

482



j.com/

J Neurol N


ovember 1996. D

ownloaded from



ual neuropsychological problems at the posta-cute assessment, but they recovered andreturned to work. The patient with cryptococ-cosis was forced to retire because of cognitivedecline and epilepsy. Cognitive defects werealso found in one patient with encephalitisafter epidemic nephropathy, the Scandinavianform of haemorrhagic fever with renal syn-drome.

Infection of the CNS caused by members ofthe adenovirus group have been reportedmostly in children.48-50 Reports of adenovirusencephalitis in adult patients are scarce and allpatients have had concomitant diseases.51-53 Inone of them nuclear inclusions and focalnecrosis of grey matter was found at necropsy.We studied two patients with adenoviralencephalitis without concomitant diseases,and one patient with concurrent borreliosis.They all had a neuropsychological profile sug-gesting severe CNS dysfunction. Borreliaburgdorferi infection may in itself result in vari-ous CNS manifestations and cognitivedeficits.5F56 Although the bilateral hippocam-pal damage seen in MRI and the findings inneuropsychological assessment were not typi-cal of Lyme disease, we cannot exclude theimpact of borreliosis in this case. Memorydefect was clear in all three patients with aden-ovirus infection and two of them had anamnesic type impairment, in which delayedmemory was impaired. They also had moodchanges and irritability. They later returned towork (two studying, one to short term jobs)but have all had difficulties.

In conclusion, the risk for cognitive deficitswas twofold to fourfold in patients with HSVEcompared with non-HSVE patients but almosthalf of the HSVE patients were later able toreturn to work. The sequelae of adenovirusencephalitis were surprisingly severe and spe-cific, resembling the amnesia after HSVE.With rapid acyclovir treatment, the leastsevere cases of HSVE were equivalent to thenon-HSV encephalitis cases with good out-come, whereas the most severe cases of non-HSV encephalitis were equivalent to the caseswith HSVE with poor outcome,

We thank Dr Esko Matikainen at the Institute of OccupationalHealth for the invaluable help in providing the control group.

1 Hanley DF, Johnson RT, Whitley RJ. Yes, brain biopsyshould be a prerequisite for herpes simplex encephalitistreatment. Arch Neurol 1987;44:1289-90.

2 Palo J, Jokelainen M, Kaste M, Teravainen H, Waltimo 0.Neurological diseases: infections. In: Neurology. Porvoo:WSOY, 1984:371-92.

3 Meyer HM, Johnson RT, Crawford IP, Dascomb HE,Rogers NG. Central nervous system syndromes of "viral"etiology. A study of 713 cases. Am J Med 1960;29:334-47.

4 Buttner T, Dorndorf W. Viral encephalitis. Experienceswith 53 patients in Middle Hessia. Fortschr NeurolPsychiatr 1988;56:315-25.

5 Wood M, Anderson M. Acute encephalitis. In: Neurologicalinfections. London: WB Saunders, 1988:381-486.

6 Skoldenberg B, Forsgren M, Alestig K, et al. Acyclovir ver-sus vidarabine in herpes simplex encephalitis.Randomised multicentre study in consecutive Swedishpatients. Lancet 1984;ii:707-11.

7 Miller JD, Ross CA. Encephalitis. A four-year survey.Lancet 1968;i:1121-6.

8 Beghi E, Nicolosi A, Kurland LT, Mulder DW, HauserWA, Shuster L. Encephalitis and aseptic meningitis,Olmsted County, Minnesota, 1950-1981: I. Epide-miology. Ann Neurol 1984;16:283-94.

9 Kennard C, Swash M. Acute viral encephalitis: its diagnosisand outcome. Brain 1981;104:129-48.

10 Hierons R, Janota I, Corsellis JA. The late effects of necro-

tizing encephalitis of the temporal lobes and limbic areas:a clinico-pathological study of 10 cases. Psychol Med1978;8:21-42.

11 Yoneda Y, Mori E, Yamashita H, Yamadori A. MRI vol-umetry of medial temporal lobe structures in amnesia fol-lowing herpes simplex encephalitis. Eur Neurol 1994;34:243-52.

12 Rose FC, Symonds CP. Persistent memory defect follow-ing encephalitis. Brain 1960;83: 195-212.

13 Kapur N, Barker S, Burrows EH, et al. Herpes simplexencephalitis: long term magnetic resonance imaging andneuropsychological profile. Jf Neurol Neurosurg Psychiatry1994;57: 1334-42.

14 Oxbury JM, MacCallum FO. Herpes simplex virusencephalitis: clinical features and residual damage.PostgradMed Jf 1973;49:387-9.

15 Kennedy PG. A retrospective analysis of forty-six cases ofherpes simplex encephalitis seen in Glasgow between1962 and 1985. QJ7Med 1988;68:533-40.

16 Gordon B, Selnes OA, Hart JJ, Hanley DF, Whitley RJ.Long-term cognitive sequelae of acyclovir-treated herpessimplex encephalitis. Arch Neurol 1990;47:646-7.

17 Counsell CE, Taylor R, Whittle IR. Focal necrotising herpessimplex encephalitis: a report of two cases with good clin-ical and neuropsychological outcomes. JNeurol NeurosurgPsychiatry 1994;57:1115-7.

18 Laurent B, Allegri RF, Thomas-Anterion C, Foyatier N,Naegele-Faure B, Pellat J. Long-term neuropsychologicalfollow-up in patients with herpes simplex encephalitisand predominantly left-sided lesions. BehaviouralNeurology 1991;4:211-24.

19 Wechsler D. Wechsler adult intelligence scale. New York: ThePsychological Corporation, 1955.

20 Wechsler D. Wechsler adult intelligence scale. WAIS manual.Helsinki: Psykologien Kustannus OY, 1970.

21 Wechsler D. Standardized memory scale for clinical use. J7Psychol 1954;19:87-95.

22 Benton A. The revised visual retention test. New York: ThePsychological Corporation, 1974.

23 Luria AR. The investigation of speech processes.Comprehension of grammatical structure. In: Traumaticaphasia. Its syndromes, psychology and treatment. TheHague: Mouton, 1970:314-322.

24 Christensen AL. Luria's Neuropsychological Investigation.Text. Copenhagen: Munksgaard, 1975.

25 Beck AT, Beck RW. Screening depressed patients in familypractise, a rapid technique. Postgrad Med J 1972;December:81-5.

26 Klapper PE, Cleator GM, Longson M. Mild forms ofherpesencephalitis. J Neurol Neurosurg Psychiatry 1984;47:1247-50.

27 Koskiniemi M, Ketonen L. Herpes simplex virusencephalitis: progression of lesion shown by CT. J Neurol1981;225:9-13.

28 Eslinger PJ, Damasio H, Damasio AR, Butters N.Nonverbal amnesia and asymmetric cerebral lesions fol-lowing encephalitis. Brain Cogn 1993;21:140-52.

29 Laurent B, Allegri RF, Michel D, et al. Primarily unilateralherpes encephalitis. Long-term neuropsychological studyof 9 cases. Rev Neurol (Paris) 1990;146:671-81.

30 Whitley RJ, Alford CA, Hirsch MS, et al. Vidarabine versusacyclovir therapy in herpes simplex encephalitis. N Engl JfMed 1986;314:144-9.

31 Whitley RJ, Alford CA, Hirsch MS, et al. Factors indicativeof outcome in a comparative trial of acyclovir and vidara-bine for biopsy-proven herpes simplex encephalitis.Infection 1987;15 (suppl 1):S3-8.

32 Koskiniemi M, Rautonen J, Lehtokoski Lehtiniemi E,Vaheri A. Epidemiology of encephalitis in children: a 20-year survey. Ann Neurol 1991;29:492-7.

33 Farkkila M, Koskiniemi M, Vaheri A. Clinical spectrum ofneurological herpes simplex infection. Acta Neurol Scand1993;87:325-8.

34 Klapper PE, Laing I, Longson M. Rapid non-invasive diag-nosis of herpes encephalitis. Lancet 1981;ii:607-9.

35 Nahmias AJ, Whitley RJ, Visintine AN, Takei Y, AlfordCA. Herpes simplex virus encephalitis: laboratory evalua-tions and their diagnostic significance. J Infect Dis1982;145:829-36.

36 Kahion J, Chatterjee S, Lakeman FD, Lee F, Nahmias AJ,Whitley RJ. Detection of antibodies to herpes simplexvirus in the cerebrospinal fluid of patients with herpessimplex encephalitis. Jf Infect Dis 1987;155:38-44.

37 Lakeman FD, Koga J, Whitley RJ. Detection of antigen toherpes simplex virus in cerebrospinal fluid from patientswith herpes simplex encephalitis. Y Infect Dis 1987;155:1172-8.

38 Aurelius E. Herpes simplex encephalitis. Early diagnosisand immune activation in the acute stage and duringlong-term follow-up. Scand J Infect Dis Suppl 1993;89:3-62.

39 Johnson RT, Olson LC, Buescher EL. Herpes simplexvirus infections of the nervous system. Problems in labo-ratory diagnosis. Arch Neurol 1968;18:260-4.

40 Morawetz RB, Whitley RJ, Murphy DM. Experience withbrain biopsy for suspected herpes encephalitis: a reviewof forty consecutive cases. Neurosurgery 1983;12:654-7.

41 Anderson NE, Willoughby EW, Synek BJ, Croxson MC,Glasgow GL. Brain biopsy in the management of focalencephalitis. J Neurol Neurosurg Psychiatry 1991;54:1001-3.

42 Fishman RA. No, brain biopsy need not be done in everypatient suspected of having herpes simplex encephalitis.Arch Neurol 1987;44:1291-2.

483



j.com/

J Neurol N


ovember 1996. D

ownloaded from



43 Rowley AH, Whitley RJ, Lakeman FD, Wolinsky SM.Rapid detection of herpes-simplex-virus DNA in cere-brospinal fluid of patients with herpes simplex encephali-tis. Lancet 1990;335:440-1.

44 Aurelius E, Johansson B, Skoldenberg B, Staland A,Forsgren M. Rapid diagnosis of herpes simplexencephalitis by nested polymerase chain reaction assay ofcerebrospinal fluid. Lancet 1991 ;337: 189-92.

45 Anderson NE, Powell KF, Croxson MC. A polymerasechain reaction assay of cerebrospinal fluid in patientswith suspected herpes simplex encephalitis. J NeurolNeurosurg Psychiatry 1993;56:520-5.

46 Tyler KL. Polymerase chain reaction and the diagnosis ofviral central nervous system diseases. Ann Neurol 1994;36:809-11.

47 Lakeman FD, Whitley RJ. Diagnosis of herpes simplexencephalitis: application of polymerase chain reaction tocerebrospinal fluid from brain-biopsied patients and cor-relation with disease. National Institute of Allergy andInfectious Diseases Collaborative Antiviral Study Group._Infect Dis 1995;171:857-63.

48 Gabrielson MO, Joseph C, Hsiung GD. Encephalitis asso-ciated with adenovirus type 7 occurring in a family out-break. J Pediatr 1966;68:142-4.

49 Huttunen L. Adenovirus type 7-associated encephalitis.Scandjl Infect Dis 1970;2:151-3.

50 Kelsey DS. Adenovirus meningoencephalitis. Pediatrics1978;61 :291-3.

51 Chou SM, Roos R, Burrell R, Gutmann L, Harley JB.Subacute focal adenovirus encephalitis. J NeuropatholExp Neurol 1973;32:34-50.

52 Kelsey DS, McLean WT. Adenoviral meningoencephalitisin a patient with lead toxicity. Arch Neurol 1979;36:384-5.

53 West TE, Papasian CJ, Park BH, Parker SW. Adenovirustype 2 encephalitis and concurrent Epstein-Barr virusinfection in an adult man. Arch Neurol 1985;42:815-7.

54 Pachner AR, Duray P, Steere AC. Central nervous systemmanifestations of Lyme disease. Arch Neurol 1989;46:790-5.

55 Krupp LB, Masur D, Schwartz J, et al. Cognitive function-ing in late Lyme borreliosis. Arch Neurol 1991;48:1125-9.

56 Halperin JJ, Luft BJ, Anand AK, et al. Lyme neuroborrelio-sis: central nervous system manifestations. Neurology1989;39:753-9.

57 Flewett TH, Hoult JG. Influenzal encephalopathy andpostinfluenzal encephalitis. Lancet 1958;ii:1 1-15.

58 McConkey B, Oxon BM, Daws RA. Neurological disordersassociated with Asian influenza. Lancet 1958; 15-17.

59 Sulkava R, Rissanen A, Pyhala R. Post-influenzalencephalitis during the influenza A outbreak in1979/1980. J Neurol Neurosurg Psychiatry 1981;44:161-3.

60 Hawkins SA, Lyttle JA, Connolly JH. Two cases ofinfluenza B encephalitis [letter]. J Neurol Neurosurg

Psychiatry 1987;50: 1236-7.61 Ponka A. The occurrence and clinical picture of serologi-

cally verified Mycoplasma pneumoniae infections withemphasis on central nervous system, cardiac and jointmanifestations. Annals of Clinical Research 1979;1 1 (suppl24): 1-60.

62 Decaux G, Szyper M, Ectors M, Cornil A, Franken L.Central nervous system complications of Mycoplasmapneumoniae. JT Neurol Neurosurg Psychiatry 1980;43:883-7.

63 Fisher RS, Clark AW, Wolinsky JS, Parhad IM, Moses H,Mardiney MR. Postinfectious leukoencephalitis compli-cating Mycoplasma pneumoniae infection. Arch Neurol1983;40:109-13.

64 Abramovitz P, Schvartzman P, Harel D, Lis I, Naot Y.Direct invasion of the central nervous system byMycoplasma pneumoniae: a report of two cases. _7 InfectDis 1987;155:482-7.

65 Weinblatt ME, Caplan ES. Fatal Mycoplasma pneumoniaeencephalitis in an adult. Arch Neurol 1980;37:321

66 Miller HG, Stanton JB, Gibbons JL. Para-infectiousencephalomyelitis and related syndromes: a criticalreview of the neurological complications of certain spe-cific fevers. Q _

Med 1956;25:427-505.67 Ikemura Y, Akena H, Okada A. Postinfectious rubella

encephalitis. Nervenarzt 1984;55:83-5.68 Ziebart-Schroth A. Tick-borne meningoencephalitis: clini-

cal picture and unusual features. Wien Klin Wochenschr1972;84:778-81.

69 Ackermann R, Rehse Kupper B. Central Europeanencephalitis in the Federal Republic of Germany. FortschrNeurol Psychiatr Grenzgeb 1979;47:103-22.

70 Conrads R, Plassmann E. Meningo encephalitis (FSME).Fortschr Medl982;100:799-801.

71 Wahlberg P, Saikku P, Brummer Korvenkontio M. Tick-borne viral encephalitis in Finland. The clinical featuresof Kumlinge disease during 1959-1987. .7 Intern Med1989;225: 173-7.

72 Lewis JL, Rabinovich S. The wide spectrum of cryptococcalinfections. Am Med 1972;53:315-22.

73 Weenink HR, Bruyn GW. Cryptococcosis of the nervoussystem. In: Vinken PJ, et al, eds. Handbook of clinical neu-rology. Amsterdam: Elsevier/North-Holland BiomedicalPress, 1978:459-502.

74 Steiner I, Polacheck I, Melamed E. Dementia andmyoclonus in a case of cryptococcal encephalitis. ArchNeurol 1984;41:216-7.

75 Yu YL, Lau YN, Woo E, Wong KL, Tse B. Cryptococcalinfection of the nervous system. QJfMed 1988;66:87-96.

76 Lahdevirta J, Savola J, Brummer-Korvenkontio M, BerndtR, Illikainen R, Vaheri A. Clinical and serological diagno-sis of nephropatia epidemica, the mild type of haemor-rhagic fever with renal syndrome. .7 Infect 1984;9:230-8.

77 Launes J, Hautanen A. Nephropathia epidemica encephali-tis. Acta Neurol Scand 1988;78:234-5.

484



j.com/

J Neurol N


ovember 1996. D

ownloaded from


aetiologies - bmj · 5-18, p

Documents