Acoustically evoked event-related potentials in HIV-associated dementia

Electroencephalography and clinical Neurophysiology, 86 (1993) 152-160 © 1993 Elsevier Scientific Publishers Ireland, Ltd. 0013-4649/93/$06.00

152

EEG92510

Acoustically evoked event-related potentials in HIV-associated dementia G. Arendt, H. Hefler and H. Jablonowski

a

Depts. o f Neurology and ~ Medicine, University o f Duesseldorf (FRG)

(Accepted for publication: 17 September 1992)

Summary Cognitiveevent-related potentials (ERPs) were recorded in 33 HIV-positive individuals with clinical signs of dementia (group C) using an auditory oddball paradigm, in comparison to a healthy control group of comparable education and social environment (group A) and to a cohort of clinically asymptomatic HIV-positive individuals (group B). Parameters evaluated were: N1, P2, N2 and P3 latencies and N1-P2, P2-N2 and N2-P3 amplitudes. Results of group B and C were correlated with clinical findings, test psychometricperformance and standard EEG records of both groups. The demented patients revealed significant prolongations of N2 and P3 peak latencies or complete abolition of these "endogenous," not stimulus-related, ERP peaks. Clinical bradykinetic symptoms and time-dependent psychometric abilities correlated with N2 and P3 latency prolongations and with the general slowing of the alpha rhythm, indicating an involvementof cortical and subcortical structures in HIV-related brain disease. Key words: Event-related potentials; Dementia; HIV-related dementia; Cognitive deficits

Soon after the acquired immunodeficiency syndrome was defined as a new infectious entity, discussion on cerebral involvement arose (Beresford 1987; Levy and Bredesen 1989; Bottomley et al. 1990). Several studies (Arendt et al. 1989, 1990a; Goodin et al. 1990; Wilkie et al. 1990; Lunn et al. 1991; O11o et al. 1991; Poutainen et al. 1991) described early motor (slowing of fast alternating movements and of rapid single index finger extensions (Arendt et al. 1989, 1990a)) or cognitive (N2-P3 amplitude reductions of event-related potentials (Arendt et al. 1990b; O11o et al. 1991), pathological performance in different neuropsychological test batteries (Lunn et al. 1991; Poutainen et al. 1991)) abnormalities in clinically asymptomatic HIV-positive patients. O t h e r groups (Goethe et al. 1989; Janssen et al. 1989; McArthur et al. 1989b; Seines et al. 1990; Franzblau et al. 1991) found no difference in neuropsychological test performance between normals and clinically asymptomatic HIV-positive individuals. The question whether there is early brain involvement progressing to clinically manifest dementia is very important. The subclinical cognitive impairment and its time course should therefore be tested in detail. Whereas there are many studies on asymptomatic HIV-positive individuals, clinical studies

Correspondence to: G. Arendt, Dept. of Neurology, Univ. of Duesseldorf, Moorenstr. 5, 4 Duesseldorf (FRG).

describing HIV-related dementia and comparing its clinical and electrophysiological characteristics with other dementing disorders are scarce. Thus, this study describes cognitive deficits in demented HIV-positive patients, classified according to the A A N criteria (American Academy of Neurology 1991); it correlates the results with clinical findings, psychometric tests and conventional E E G records and compares the findings in the demented patients with findings in clinically asymptomatic patients.

Methods Patients

In 33 HIV-positive patients with clinical signs of dementia (subjective complaint of memory loss, general mental slowing, memory dysfunction, affective disturbances) neurological abnormalities were scored. The clinical score ranged from 0 -- no, 1 = mild, 2 = moderate to 3 = severe symptoms. Patients with i.v. or oral drug or alcohol abuse, fever and proven C M V retinitis were excluded. In addition to clinical examination, patients underwent psychometric tests, conventional E E G recording and acoustically evoked event-related potential measurements (ERP). Results of the demented cohort (C) •were compared to those of a control group of HIVnegative individuals of comparable education and so-

ERPs IN HIV-ASSOCIATED DEMENTIA

153

cial environment (A) and to the performance of a much larger cohort of clinically asymptomatic HIVpositive individuals (B); group C patients had a slightly, nevertheless statistically significant, higher mean age than group A and B patients (Table I). This, however, was taken into account for evaluation of the P3 parameters (Figs. 3 and 4). Table I gives information on age, sex, CDC stage (CDC 1986), and risk behaviour of the 3 groups. In the control group a higher percentage of female subjects was tested in comparison to the other two groups. Group B patients had no virostatic medication; in group C, 9 patients were under current AZT medication, 8 had discontinued, the remaining had never taken AZT. Motor test results of groups A and B have been presented in separate studies (Arendt et al. 1989, 1990a).

Psychometric tests On the same day when ERP records (see below) were obtained, patients were tested psychometrically using: (1) the multiple choice vocabulary test form B (MWT-b) to assess the level of premorbid verbal intelligence, (2) the German version of the Raven standard progressive matrices, testing actual non-verbal intelligence, (3) the syndrome short test (SKT) measuring

short-term memory, attention and concentration, (4) the AMDP psychic and somatic findings to evaluate physical and psychological condition of the patient groups according to the American DSM III system. A 10-point difference MWT-b > Raven was considered as a test psychological threshold criterion of dementia. Depression was rated with the AMDP rating scale. For further methodological details see Arendt et al. (1990b).

ERP recording All individuals of the 3 groups underwent acoustically evoked event-related potential recording, using an auditory oddball paradigm. Abnormalities in the afferent auditory pathway were excluded. One group B patient had a higher hearing threshold than others, but was able to distinguish the higher pitched tones correctly. Patients were seated in a reclining chair and told to close the eyes for optimal concentration and in order to avoid eye blink artefacts, automatically rejected by the system using a threshold criterion for the frontal electrodes. They were also instructed that they would hear a series of 2 tone pips differing in pitch, with the lower pitched tone (1000 Hz, 90 dB) occurring more frequently (overall probability 80%) than a higher pitched tone (2000 Hz, 90 dB, probability 20%) the latter being randomly embedded in the frequently presented tones. Individuals had to

TABLE I Demographical data. Controls (A)

Asymptomatic HIV patients (B)

Demented HIV patients (C)

Age (years)

35.5_+ 10.4

37.1 _+8.4

42.7 + 9.8

Sex

13 female 30 male

8 female 92 male

4 female 29 male Before showing clinical signs of dementia

CDC stage

II III IV A IV C1 IV C2 IV D

25 36 9 17 6 7

1

28 2 2

Risk group

Homosexual men Bisexual men Heterosexuals Hemophiliacs Pat. with WillebrandJuergens syndrome Pat. with blood transfusion Central Africans Pat. with drug abusing partner No evident risk group

15 28

83 3 1 4

23 3 3 1

3

154

count the high-frequency rare tone (target) and to ignore the lower tone (non-target). Two complete independent runs were carried out in each subject. In every run, stimuli were presented until the recording system had collected 64 infrequent and 256 frequent trials. Patients were informed before the examination that they would be asked for the correct n u m b e r of " t a r g e t " tones after each run. The intraindividual reliability of the paradigm has been shown in other studies (Arendt et al. 1990b). A Cadwell 8400 16-channel system was used. AgAgC1 electrodes were attached to the scalp according to the conventional 10/20 system. Ear-linked electrodes served as reference. This is a standard paradigm for P300 recording and has been used in earlier studies ( H o e m b e r g et al. 1986; Arendt et al. 1990b). For group comparisons, the following parameters were evaluated over Cz: N1, P2, N2 and P3 latencies, N1-P2, P2-N2 and N2-P3 amplitudes. Normal ranges were defined on the data base of the control group.

EEG recording Routine E E G recording was performed with the same montage of the electrodes using a 21-channel Siemens E E G recorder (Mingograf). MRI scanning In groups B and C M R I scans with a D I A sonic scanner (0.35 T, T E = 60 sec, T R = 2500-3000 msec, axial scanning) were performed. The results of group B have been reported previously (Arendt et al. 1989). Statistics For statistical comparison of parameters across groups, Student's t test was used. For correlations, the rank coefficient was determined. The normal, age-dependent range for an evaluated p a r a m e t e r X was defined in the following way: first the regression line X = a * A + b (regression line between age A and the p a r a m e t e r X) was calculated. Then for each individual patient with an age A i the difference D~ between the measured value X i and the value on the regression line (a * A i + b) was determined. The standard deviation MD of the differences D i measures the variation of tlae individual values X i around the regression line. The area of 2 * M D width around the age dependence regression line was defined as the normal range. The results of a patient were judged to be abnormal if they lay outside this normal range.

Results

ERP parameters 79% of the group C patients showed abnormal results in at least one p a r a m e t e r of the E R P s in contrast

G. ARENDT ET AL. female control, 26 years

|

,^x - ~ ~ P2 ~ N I"

~--/ l

~P2

/

....

,.F /

NI'--''-

~

_ - first run Cz second run

/ I j /

N

first run Cz second run tna°;~tasrgets

[

-75 0

1

I

I

[

I

t

h

I

300

I

675 ms

Fig. 1. Normal ERPs o f a 26-year-old scronegative female with correct responses to target (solid line) and non-target (dashed line)

tones over Cz.

to 28% in group B patients without central nervous symptoms. All of the patients counted the higher pitched tones with an error of less than 3 / 6 4 ( < 5%). 42% of the group C patients had abolished "endogenous," not stimulus-related E R P peaks, 31% had latency prolongations and 6% an additional N2-P3 amplitude reduction or an amplitude reduction alone. Fig. 1 shows the normal E R P of a 26-year-old seronegative female with adequate responses to target (solid line) and non-target (dashed line) tones over Cz. All of the latency peaks can be clearly distinguished in the target tone trials. In the non-target tones only reproducible N1 and P2 components are found. Fig. 2 presents the ERPs of 3 HIV-positive patients with characteristic electrophysiological pathology. In the upper part the E R P performance of a 44-year-old patient in C D C stage II shows normal responses to the target and non-target tones. The middle part reveals the E R P results of a 41-year-old slightly demented female hemophiliac with a clear N2 and a slight P3 latency prolongation. The lower part shows the results of a severely demented 24-year-old homosexual with abolished E R P peaks; the N1 component - as the only reproducible peak - is very late; the patient counted 61 target tones, but this slight mistake cannot account for the severe E R P abnormalities. Table II compares the group mean values of the 7 E R P parameters tested across the 3 groups. The demented patients revealed a significant prolongation of the N2 and P3 latencies in comparison to group A and B patients on a group statistical level. The N2-P3 amplitudes were significantly reduced in group B and to a lesser extent in group C patients in comparison to group A on a group statistical level as well. These findings are illustrated in Figs. 3 and 4.

ERPs IN HIV-ASSOCIATED DEMENTIA

155

Clinical examination In group C basal ganglia dysfunction was found in addition to the clinical signs of dementia, but no other focal neurological deficits, whereas in group B no clinical CNS symptoms were present. The basal ganglia symptoms in group C were: bradykinesia (n = 32), tremor (n = 16), dystonia (n = 23) and parkinsonian gait disturbance (n = 18).

_

--1__.

CDC II, 44 yr (3

""

" ~

.

.

.

.

~- . . . .

N2 " ~

first run

f

~

Cz second run

-

"

~

first run Cz

P 2 ~ / ~

"~

secOnd run

I ,,v I

-75 0

I

1

1

I

300

i

i

t

f

J

675 ms

CDC IV B, 41 yr <2

N1 ,,,,,'"~. -'--,,. .

........ first run ~s~ Cz / ~ - ' - ' N , . f . . . . . . . - second run

/

N1

-

---"N1

N2

~

first run Cz

second run

I

-75 0

I

I

I

I

300

I

I

I

I

]

675 ms

CDC IV B, C1 24 yr d

--~_~-,,

.,

,

~-~

first run

-~

Cz secondrun

....--

first

non-tarqets targets t

i

t

f

I

I

I

run

Cz second run

14 t~V l

I

I

-75 0 300 675 ms Fig. 2. ERPs of a CDC stage II patient (upper part) with normal latencies and amplitudes, a mildly demented female patient (middle part) with N2 and slightly prolonged P3 latency and the ERPs of a severely demented male patient (lower part) showing merely an N1 component of low reproducibility.

Table III demonstrates the results of the clinical score. Psychometric testing Table IV presents the results of psychometric testing of patients in groups B and C in comparison to the controls. The table reveals a dementing disorder in group C with a marked decline in time-dependent cognitive abilities (see SKT results). It must be emphasized that the Raven test was performed only in 20 of the 33 clinically demented patients, because the cognitive impairment was so severe in the other 13 patients that they were unable to carry out the test (the Raven test is not time-limited and thus may require concentration over a long period of time; the 13 patients interrupted the test because they were not able to work on it for hours and not because they were unable to understand the instructions). In the demented patients correlation of psychometric parameters with E R P amplitudes and latencies did not yield significant results. Only the SKT correlated with P3 latency prolongation ( P < 0 . 0 5 , one-sided testing). The depression score also did not correlate with E R P parameters and was not significantly different from the depression scores of the other two groups. None of the psychometric or E R P parameters correlated with age (see Table V) in demented patients, whereas in group A and B patients there was a correlation of N2-P3 amplitude reduction with age, pathologically accelerated in group B (see below). Furthermore, one-sided correlation testing revealed a significant correlation of the SKT with the clinical score ( P < 0.05). M R I scans MRI scans of group B were normal. The MRI scans in group C revealed cortical atrophy in 37% and echointensive zones in subcortical structures in 63%, most often in the globus pallidus, the putamen and the thalamus. None of the MRI scans gave any hint of opportunistic brain infection. E E G records Conventional E E G recording showed slowing of the alpha rhythm in 9 group B patients and in 22 group C patients, diffuse theta activity in 22 group B patients and in 15 group C individuals; in 4 group B and in 9 group C patients there was a theta focus. Furthermore, theta paroxysms were observed in 4 group B patients. There were no slow biphasic complexes (Beaumanoir et ai. 1988) in group C patients or any form of epileptic activity. The decreased frequency of the alpha rhythm as the marked and characteristic E E G finding in the demented patients correlated significantly with the clini-

156

G. A R E N D T ET AL.

TABLE II Latency (msec) (x 4- S.D.)

Group A

Group B

GroupC

N1

P2

n = 43 103.6 -+ 10.3 n = 100 106.2 -+14.1 n=33 107.9 -+ 15.3

n = 43 181.2 -+ 21.5 n 91 180.6 -+18.9 n=19 180.4 -+23.1

Amplitude (#V) (x _4-S.D.)

N2 n = 43 225.1 r[-+ 26.0 [[ n = 91 [[ 235.5 [[ []- -+27.6 [[ [[ n = 1 9 u---u- 253.4 * * +27.5

=

* Significant P < 0.05

I

I

** Significant P < 0.01

I

I.

P3

N1-P2

P2-N2

N2-P3

n = 43

n = 43

n = 43

n = 43

336.2

10.4

4.3

15.9

+3.7 n = 91 11.1 -+4.5

+3.0 n =91 5.7 -+3.6

+20.5 n = 91 337.2 +31.7

*--~1

n = 19

n = 19

n = 19

378.3 ** -+21.2

12.0 _+4.9

8.6 +4.0

~

+4.7 n = 91 12.1 ** -+6.6 n = 19 13.1 * -+4.4

* * * Significant P < 0.002 ~ - - - - J .

cal score (0.5228, P < 0 . 0 1 two-sided testing) and showed a weaker correlation with the SKT (0.321, P < 0.05 one-sided testing), but no correlation with the P3 parameters. Fourteen demented patients with diffuse slowing of the alpha rhythm showed cortical atrophy in MRI scans. In summary, the functional tests seemed to be more sensitive in detecting CNS deficits in HIV-positive patients than the imaging techniques. Thus, the main results of the presented study are: (1) the ERP results in clinically asymptomatic HIVpositive patients (N2-P3 amplitude reduction only) differ from those of demented HIV-positive individuals (N2 and P3 latency prolongation or total peak abolition); (2) the electrophysiological abnormalities (EEG and ERP) correlate with time-dependent psychometric

testing and with the clinical findings, revealing mainly subcortical symptoms.

Discussion The question whether electrophysiologically detectable subclinical deficits have any relevance and are precursors of clinical dementia in HIV-infected patients is under debate. To settle this point, follow-up examinations of immunologically and neurologically asymptomatic CDC II patients are necessary. Studies on already manifest HIV-related dementia, however, are required to improve knowledge about the characteristics of the disease. P

P300-Latencies H I V - i n f e c t e d patients

~30C-Latencres

300

-

Latenc~es

~r~

demented HIV-pos. pat.

Controls

500

450 450 [ !

425

425

460

o o

400 400

L i

~

375 ~

~

~

~ --

•

~

--

375

32a ~

-

•

-,,p-•

I

~g •

~ ~ ~ -e " ~ -

~ ~

o

o

oo

o

300

o

~

o

o~_--o

o

~

oo

oo ~ o a2s

o .... z---

o.9-- ~ - " o °

_

o

34O

300

2 7 5 r-

• 25C

250

15 15

Q

"8 o

So

;j

oo

.... o ' % °o oO

~ ~ ~ "

8

o

25

35

45

55

65

25

Oi

i

,

35

45

55

a g e (years)

i

65

25

35

45 age

55

1

65

5

{y~s)

aQe ( y e a r s )

Fig. 3. Correlation of P3 latency with age in normals (left), in clinically asymptomatic (middle) and d e m e n t e d (right) HIV-positive individuals with latency prolongations in the latter group. The normal range (see Methods) is indicated by dashed lines (note the difference in scale in the demented patient group).

ERPs IN H I V - A S S O C I A T E D D E M E N T I A

157 TABLE IV

TABLE III Group C

Brady-

patients

kinesia

Tremor

Dystonia

1

3

0

0

0

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

3 3 3 3 3 3 3 3 3 3 3 3 3 1 2 0 3 2 3

2 2 0 2 0 2 2 0 0 2 0 2 0 0 0 0 2 0 0

3 1 2 2 0 2 3 1 2 2 1 0 0 2 0 0 1 0 0

3 3 3 3 0 3 3 0 0 3 3 0 0 0 0 0 3 0 0

21

2

0

0

0

22 23 24 25 26 27 28 29 30 31 32 33

1 3 3 3 1 3 2 3 3 3 3 3

0 2 2 0

1 2 3 1 0 2 1 2 2 3 3 1

0 3 3 3 0 3 0 3 3 3 3 0

0 2 1

0 2 3 3

0

86/33

31/33

43/33

Group

Parkinsonian gait disturb.

P3 lat. N2-P3 P3 lat. N2-P3

51/33

20

20

o

o

~

(B) ampl. (B)

D

0.297 n.s. 0.167 n.s. 0.1543 n.s. 0.2122 *

0.416 * 0.054 n.s. 0.06 n.s. -0.025 n.s.

-0.28 n.s. 0.013 n.s. 0.035 n.s. 0.020 n.s.

•

•

15

o°

Amplitudes H[V-po&

demented

in

pat.

25 [

-.......

o

00% o

P 300

oatients

~

00°o o o

4

ampl. (C)

SKT

P300-Amplitudes

25

•

(C)

Age

Neuropathological studies in HIV-infected subjects (Moskowitz et al. 1984; Navia et al. 1986; Gyorkey et al. 1987) have revealed a variety of abnormalities. In early HIV-related brain disease astrocytosis of white matter, mild pallor of myelin staining in the absence of inflammation, multinucleated giant cells and cortical atrophy have been described (McArthur et al. 1989a) whereas in patients with dementing symptoms late in the course of HIV infection post mortem findings were mainly multinucleated giant cells, multifocal rarification of the white matter with diffuse pallor associated with perivascular lymphocytic infiltrates and brown pigmented macrophages (Navia et al. 1986). Conse-

25 ..': " i ............... •

113.5_+18.9 101.8_+ 19.6 10.8_+ 5.3

• P ~<0.05. SKT = syndrome short test; D = depression score.

HIV-infeoted

•

C

112.7_+ 7.01 111.9_+10.5 4.3_+ 3.4

TABLE V

Controls

•

B

112.9+10.1 113.1_+ 9.8 3.5_+ 0.1

MWT-b = multiple choice vocabulary test - form B; SKT = syndrome short test.

P300-Amplitudes

•

A IQ MWT-b IQ Raven SKT

i

~~

2' t

0

oo

~

. . . .

"

I

•

•

!

•

10

oo "~----°~o.

% 15

i

i

i

i

25

35

45

55

a g e (years)

o 8 ° ~ e 8o~ 20

0o0

35

45

~

~

0

~

•

•

+-. . . . .

•• i

65

15

25

55

65

25

35

45

55

65

J

75

a g e (yeers)

Fig. 4. Correlation of N2-P3 amplitude with age in normals (left) and in clinically asymptomatic HIV-positive individuals (middle) with a clearer amplitude reduction with aging in the patient group (B) and no such correlation in the demented HIV positives (right). The normal range (see Methods) is indicated by dashed lines (note the difference in scale in the demented patient group).

158 quently, a recent neuropathological consensus conference (Budka et al. 1991) has attempted to differentiate between " HI V encephalitis" and "HIV-related leucoencephalopathy," the first with multinucleated giant cells and microglial nodules, the latter with diffuse demyelination. Furthermore, neuropathological studies have provided data emphasizing either a subcortical (Kure et al. 1990) or a cortical site (Ketzler et al. 1990) of HIV-related brain pathology. Whereas Kure et al. reported a very high expression of the HIV protein gp41 in basal ganglia and thalamus, Ketzler et al. found diffuse neuronal loss and a reduction of nerve cell density in the frontal cortex. As to clinical manifestations there is evidence that signs detected earliest, such as especially bradykinesia, are related to subcortical pathology and more specifically basal ganglia dysfunction. This is corroborated by motor testing in HIV-positive individuals (Arendt et al. 1989, 1990a). ERPs are commonly considered to have a subcortical generator (Sutton et al. 1965; Donchin and Cohen 1967; Donchin 1979, 1988; Halgren et al. 1980; Duncan-Johnson and Donchin 1982). Thus, the latency prolongations or, even more dramatically, complete abolition of late P3 peaks observed in HIV-related dementia (group C) correspond to the above-mentioned clinical findings in that both supported the predominant role of subcortical pathology in determining clinical manifestations. Interestingly, HIV-negative demented populations present similar ERP abnormalities (Goodin et al. 1978; Goodin and Aminoff 1986). In contrast, amplitude reduction of ERPs has been observed in psychiatric disorders not accompagnied by dementia, such as schizophrenia and endogenous depression, and was shown not to correlate with ratings on psychiatric scales (Pfefferbaum et al. 1984b). This resembles results of an earlier study (Arendt et al. 1990b) in which amplitude reductions in HIV-positive patients asymptomatic for CNS deficits were found, which also were not correlated with depression scores. As underlined by Kovner et al. (1989) these findings cannot be explained by lack of attention or personality factors. These results are suggestive of similar cognitive deficits in acute psychiatric illness and clinically asymptomatic HIV-positive patients, the amplitude reductions not correlating with emotional factors, whereas results in demented individuals are characterized by latency prolongations or complete abolition of "endogenous" ERP peaks (N2, P3). Thus, it must be concluded that different mechanisms account for the different ERP abnormalities in group B and C patients, which is underlined by the fact that only the age dependence of N2-P3 amplitudes in normals (Pfefferbaum et al. 1984a) is accelerated in the

G. ARENDT ET AL. asymptomatic patients. This may be suggestive of an additional process taking place in the demented individuals. The latency prolongation of the late ERP components in demented patients not only correlated with time-dependent psychometric testing - indicating a slower stimulus processing - but also significantly with the clinical score. ERP and EEG abnormalities, testing different cerebral pathway functioning, did not intercorrelate. We believe that there is cortical (EEG slowing) and subcortical (ERP abnormalities) dysfunction in demented HIV patients with the subcortical impairment causing earliest marked clinical symptoms. Additionally, both dementia components can be diagnosed by electrophysiological methods. The EEG slowing as well as the ERP abnormalities correlated independently with the clinical score and therefore both contribute to "dementia." Future studies will have to clarify whether the amplitude reduction in early HIV infection possibly characterizes a reversible process, as suggested by therapy studies (Schmitt et al. 1988), and whether latency prolongation indicates irreversible demyelination. As to therapeutic effects it may be noted that half of the demented patients in this study never underwent therapy and were those to reveal the most pathological ERP results in comparison with treated demented patients. Thus, it may be speculated whether early treatment could prevent dementia or protract its clinical manifestation. In conclusion, it is suggested that HIV-related dementia ia based on subcortical as well as cortical dysfunction. Electrophysiological methods provide access to the evaluation of these different components, thus allowing further differentiation of the clinical syndrome. It must, however, be stressed that neither refined clinical testing nor sophisticated electrophysiological techniques are capable of determining specifically to what extent functional deficits are due to circumscript structural lesions or rather to irritation in physiological circuits, as for example, conceivable in the case of toxic effects. Follow-up studies in asymptomatic patients in whom clinical and electrophysiological parameters are already well-documented are required for further insight into mechanisms underlying HIV-associated dementia. Furthermore, these studies will benefit from incorporating data collected by imaging techniques and neuropathological examination and from closely monitoring therapeutic effects.

References

American Academy of Neurology,the Working Group. Nomenclature and researchcase definitionsfor neurologicmanifestationof

ERPs IN HIV-ASSOCIATED DEMENTIA human immunodeficiency virus-type 1 (HIV-1) infection. Neurology, 1991, 41: 778-785. Arendt, G., Hefter, H., Elsing, C., Neuen-Jakob, E., Strohmeyer, G. and Freund, H.-J. Neue elektrophysio!ogische Befunde zur Hhufigkeit der Gehirnbeteiligung bei klinisch-neurologisch asymptomatischen HIV-Infizierten. Z. EEG-EMG, 1989, 20: 280-287. Arendt, G., Hefter, H., Elsing, C., Strohmeyer, G. and Freund, H.-J. Motor dysfunction in HIV-infected patients without clinically detectable central-nervous deficit. J. Neurol., 1990a, 237: 362-368. Arendt, G., Hefter, H., Hoemberg, V., Nelles, H.-W., Elsing, C. and Freund, H.-J. Early abnormalities of cognitive event-related potentials in HIV-infected patients without clinically evident CNS deficits. Electroenceph. clin. Neurophysiol., 1990b, Suppl. 41: $370-$380. Beaumanoir, A., Burkhard, P., Gauthier, G., Le Floch-Rohr, J., Ochsner, F. and Waldvogel, F. EEG dans 19 cas de SIDA avec atteinte de l'enc~phale. Rev. EEG Neurophysiol., 1988, 18: 313322. Beresford, Ch. Dementia in HIV infection. NZ Med. J., 1987, 1: 32. Bottomley, P.-A., Hardy, C.J., Cousins, J.P. et al. AIDS dementia complex: brain high-energy phosphate metabolite deficits. Radiology, 1990, 176: 407-411. Budka, H., Wiley, C.A., Kleihues, P., Artigas, J., Asbury, A.K., Cho, E.S., Cornblath, D.R., Dal Canto, M.C., DeGirolami, U., Dickson, D., Epstein, L.G., Esiri, M.M., Giangaspero, F., Gosztonyi, G., Gray, F., Griffin, J.W., H~nin, D., Iwasaki, Y., Janssen, R.S., Johnson, R.T., Lantos, P.L., Lyman, W.D., McArthur, J.C., Nagashima, K., Peress, N., Petito, C.K., Price, R.W., Rhodes, R.H., Rosenblum, M., Said, G., Scaravilli, F., Sharer, L.R. and Vinters, H.V. HIV-associated disease of the nervous system: review of nomenclature and proposal for neuropathology-based terminology. Brain Pathol., 1991, 1: 143-152. CDC. Classification system for human T-lymphotropic virus type III lymphadenopathy. Associated virus infectious morbidity and mortality. Weekly Rep., 1986, 35: 334-339. Donchin, M. Event-related brain potentials: a tool in the study of human information processing. In: H. Begleiter (Ed.), Evoked Brain Potentials and Behaviour. Plenum Press, New York, 1979: 13-18. Donchin, M. Event-related potential: the tool in the study of human information processing. In: H. Begleiter (Ed.), Even Related Brain Potentials and Behaviour. Plenum Press, New York, 1988: 13. Donchin, E. and Cohen, L. Averaged evoked potentials and intramodality selective attention. Electroenceph. clin. Neurophysiol., 1967, 22: 537-546. Duncan-Johnson, C.C. and Donchin, E. The P300 component of the event-related brain potential as an index of information processing. Biol. Psychol., 1982, 14: 1-52. Franzblau, A., Letz, R., Hershman, D., Mason, P., Wallace, J.I. and Bekesi, J.G. Quantitative neurologic and neurobehavioral testing of persons infected with human immunodeficiency virus type 1. Arch. Neurol., 1991, 48: 263-268. Goethe, K.E., Mitchell, J.E., Marshall, D.W., Brey, R.L., Cahill, W.T., Leger, G.D., Hoy, L.J. and Boswell, R.N. Neuropsychological and neurological function of human immunodeficiency virus seropositive asymptomatic individuals. Arch. Neurol., 1989, 46: 129-133. Goodin, D.S. and Aminoff, M.J. Electrophysiological differences between subtypes of dementia. Brain, 1986, 109: 1103-1113. Goodin, D.S., Squires, K. and Starr, A. Long latency event related components of the auditory evoked potential in dementia. Brain, 1978, 101: 635-648.

159 Goodin, D.S., Aminoff, M.J., Chernoff, D.N. and Hollander, H. Long latency event-related potentials in patients infected with human immunodeficiency virus. Ann. Neurol., 1990, 27: 414-419. Gyorkey, F., Melnick, J.L. and Gyorkey, P. Human immunodeficiency virus in brain biopsies of patients with AIDS and progressive encephalopathy. J. Infect. Dis., 1987, 9: 261. Halgren, E., Squires, N.K., Wilson, C.L., Rohrbaugh, J.W., Babb, T.L. and Crandall, P.H. Endogenous potentials generated in the human hippocampal formation and amygdala by infrequent events. Science, 1980, 210: 803-805. Hoemberg, V., Hefter, H., Granseyer, G., Strauss, W., Lange, H. and Hennerici, M. Event-related potentials in patients with Hunting: ton's disease and relatives at risk in relation to detailed psychometry. Electroenceph. clin. Neurophysiol., 1986, 63: 552-569. Janssen, R.S., Saykin, A.J., Cannon, L. et al. Neurological and neuropsychological manifestation of HIV-1 infection: association with AIDS-related complex but not asymptomatic HIV-1 infection. Ann. Neurol., 1989, 26: 592-600. Ketzler, S., Weis, S., Haug, H. and Budka, H. Loss of neurons in the frontal cortex in AIDS brains. Acta Neuropathol., 1990, 80: 92-94. Kovner, R., Perecman, E., Lazar, W., Hainline, B., Kaplan, M.H., Lesser, M. and Beresford, R. Relation to personality and attentional factors to cognitive deficits in human immunodeficiency virus infected subjects. Arch. Neurol., 1989, 46: 274-277. Kure, K., Weidenheim, K.M., Lyman, W.D. and Dickson, D.W. Morphology and distribution of HIV-1 gp41-positive microglia in subacute AIDS encephalitis. Acta Neuropathol., 1990, 80: 393400. Levy, R.M. and Bredesen, D.E. Controversies in HIV-related central nervous system disease: neuropsychological aspects of HIV-1 infection. AIDS Clin. Rev., 1989, 151-191. Lunn, S., Skydsberg, M., Schulsinger, H. et al. A preliminary report on the neuropsychologic sequelae of human immunodeficiency virus. Arch. Gen. Psychiat. 1991, 48: 139-142. McArthur, J.C., Becker, P.S., Parisi, J.E., Trapp, B., Seines, O.A., Cornblath, D.R., Balakrishnan, J., Griffin, J. and Price, D. Neuropathological changes in early HIV-I dementia. Ann. Neurol., 1989a, 26: 681-684. McArthur, J.C., Cohen, B.A., Seines, O.A. and Kainar, A.J. Low prevalence of neurological and neuropsychological abnormalities in otherwise healthy HIV-1 infected individuals: results from the multicenter AIDS cohort study. Ann. Neurol., 1989b, 26:601-611. Moskowitz, L.B., Hensley, G.T., Chan, J.C., Conley, F.K., DonovanPost, M.J. and Gonzalez-Arias, S.M. Brain biopsies in patients with acquired immunodeficiency syndrome. Arch. Pathol. Lab. Med., 1984, 108: 368-371. Navia, B.A., Jordan, B.D. and Price, R.W. The AIDS-dementia complex. II. Neuropathology. Ann. Neurol., 1986, 19: 517-524. Ollo, C., Johnson, R. and Grafman, J. Signs of cognitive change in HIV disease: an event-related brain potential study. Neurology, 1991, 1: 193-199. Pfefferbaum, A., Ford, J.M., Wenegrat, B.G., Roth, W.T. and Kopell, B. Clinical application of the P3 component of event-related potentials. I. Normal aging. Electroenceph. clin. Neurophysiol., 1984a, 59: 85-103. Pfefferbaum, A., Ford, J.M., Wenegrat, B.G., Roth, W.T. and Kopell, B.S. Clinical application of the P3 component of event-related potentials. II. Dementia, depression, schizophrenia. Electroenceph. clin. Neurophysiol., 1984b, 59: 104-124. Poutainen, E., Haltia, M., Elovaara, I., Laehdevirta, J. and Iivanainen, M. Dementia associated with human immunodeficiency virus: subcortical or cortical? Acta Psychiat. Scand., 1991, 83: 297-301. Schmitt, F.A., Bigley, J.W., McKinnis, R. et al. Neuropsychological

160 outcome of Zidovudine (AZT). Treatment of patients with AIDS and AIDS related complex. New Engl. J. Med., 1988, 319: 15731578. Seines, O.A., Miller, E., McArthur, J.C., Gordon, B., Munoz, A., Sheridan, K., Fox, R., Saah, A.J. and the Multicenter AIDS Cohort Study. HIV-1 infection: no evidence of cognitive decline during the asymptomatic stages. Neurology, 1990, 40: 204-208.

G. ARENDT ET AL. Sutton, S., Braren, M., Zubin, J. and John, E.R. Evoked potential correlates of stimulus uncertainty. Science, 1965, 150: 1187-1188. Wilkie, F.L., Eisdorfer, C., Morgan, R., Loewenstein, D.A. and Szapcznik, J. Cognition in early human immunodeficiency virus infection. Arch. Neurol., 1990, 47: 433-440.