Motor analysis predicts progression in HIV-associated brain disease

JOURNAL OF THE

NEUROLOGICAL SCIENCES

ELSEVIER

Journal of the Neurological Sciences 123 (1994) 18/)-185

Motor analysis predicts progression in HIV-associated brain disease G. Arendt ~"*, H. Hefter ", F. Hilperath a, H.-J. von Giesen ", G. Strohmeyer ~, H-J. Freund ~' Departments ~[ " Neurology and ~' Medicine, Heinrich lteine Unil etwiO', Moorenstra/3e 5, D-4000 Dusseldor]i Germany

(Received 26 January_ 1993; revised 14 October 1993; accepted 8 November 1993)

Abstract

One hundred HIV-positive individuals without clinically evident central nervous system (CNS) deficits entered this follow-up study and were examined clinically and with a well-defined motor test battery every 3 months over 2 years or until they deceased. They underwent magnetic resonance tomography once a year. None received any form of therapy at onset of the study. Three groups were analyzed: (A) patients without electrophysiologically detectable motor impairment (n = 23), (B) patients with electrophysiologically detectable motor impairment but no virostatic medication (n = 33), and (C) patients with motor deficits undergoing AZT treatment (n = 44) after study onset. Group A patients, although slightly deteriorating over time, had the best clinical and electrophysiological outcome compared to the other groups, whereas group B patients deteriorated markedly in both clinical and electrophysiological tests, even though the majority did not develop cerebral complications during the observation period. Those group C patients belonging to early CDC stages (II and III) improved electrophysiologically under AZT therapy, while 76% of the patients in more advanced stages (CDC IVA-D) died of cercbral AIDS manifestations. Four patients of this group, being alive at the end of the study, were completely demented. It is suggested that early detectable motor impairment predicts future cerebral involvement in AIDS. Late onset of virostatic treatment did not influence the clinical outcome. Key words: HIV infection; Motor slowing; T-helper cell count; Clinical course

1. Introduction

In research on A I D S - r e l a t e d brain disease it became obvious that many HIV-positive individuals sooner or later reveal m o t o r ( A r e n d t et al. 1989) a n d / o r cognitive (O11o et al. 1991) deficits correlating with CDC-stages. The A m e r i c a n A c a d e m y of Neurology 1991 agreed on a n o m e n c l a t u r e for these disabilities defining the mild form as " H I V - l - a s s o c i a t e d minor c o g n i t i v e / m o t o r disorder" and the severe form as " H I V - l - a s s o c i a t e d dementia complex". However, to date, the significance of the " m i n o r c o g n i t i v e / m o t o r disorder" - comprising also subclinical deficits detectable with sensitive test procedures for the course of A I D S - r e l a t e d brain involvement remains unclear. It is unknown w h e t h e r these " m i n o r deficits" are first signs of a cortical a n d / o r subcortical dementive process developing later ( A A N 1991) or whether they are unspecifically related to the course of

* Corresponding author. Tel.: (+ 49-2) 311 8469. 0022-510X/94/$(17.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0{122-510X(93)E0271-A

the disease and, thus, only reflect deterioration of the patients' general health. A recent study (Murray et al. 1992) on SIV-infected monkeys revealed early m o t o r and cognitive disabilities as the most indicative sign of virus-induced brain affection suggesting that m o t o r and cognitive abnormalities in clinically asymptomatic HIV-infected individuals probably are highly relevant for the course of the disease. The study presented here was designed as a clinical and electrophysiological follow-up of a group of HlVpositives with electrophysiologicaIly detectable motor deficits. The initial m o t o r results have been published a few years ago ( A r e n d t et al. 1989). Furthermore, the influence of virostatic medication, which has been proposed to improve neuropsychological p e r f o r m a n c e (Fischl et al. 1987; Schmitt et al. 1988; A r e n d t et al. 1992), on m o t o r deficits and immune status was analyzed. Following-up the clinical and electrophysiological course of " m i n o r neuropsychological deficits" is especially important for optimal timing of antiviral medication early or late in the course of HIV-infection which

(i. Arendt et al. /Journal o# the Neurological Sciences 123 (1994) 180-185

- to date - is based on testing the immunological status alone.

2. Methods

Patients One hundred HIV-positive patients without clinically detectable CNS deficits were followed-up over a 2-year period using a well-defined motor test battery (Arendt et al. 1989). Intraindividual in comparison to interindividual variability has been tested previously (Hefter 1991), e.g., the coefficient of variability of five consecutive measurements of times to peak every half year over 2 years was less than 6.5% in contrast to an interindividual coefficient of 19.7% in another control population of 29 normal subjects or 15.1% in the present control group (Hefter 1991). Thus, this test battery is an appropriate tool for follow-up studies. Exclusion criteria were alcohol and drug abuse, fever, opportunistic cerebral infections, cytomegalovirus retinitis and seizures. None of the patients received any form of virostatic medication at the beginning of observation. 86 patients were right-handed, 9 ambidextrous and 5 left-handed (self-addressed hand preference). There were 92 male and 8 female individuals (mean age 37.1 + 8.1 years). 41 individuals of comparable age (35.5 + 10.4 years) and sex distribution (33 males, 8 females) served as controls. IQ results of the patients (MWT-b: 110.8 + 14.5, Raven: 110.7 + 12.4) and controis (MWT-b: 111.2 + 13.9, Raven: 111.1 + 10.1) were also comparable. Distribution to the risk groups revealed 83 homosexual men, 3 bisexual men, 1 heterosexual man, 5 hemophiliacs, 1 patient with blood transfusion, 1 Central African, 3 individuals with drug abusing partners, and 3 belonging to none of the defined risk groups. The initial distribution to the CDC stages was: 33 individuals in CDC stage II, 26 CDC III, 6 CDC IVA, 23 CDC IVC1, 2 CDC IVC2, 10 CDC IVD. All patients underwent clinical examination by a neurologist at the beginning of the study and every three months from then on including tests for motor performance and extrapyramidal function. After the initial examination the patient cohort was separated into three groups. Group A: patients without electrophysiological motor deficits (n = 23; 14 CDC II, 6 CDC III, 1 CDC IVC2, 2 CDC IVD). Group B: patients with pathological results in the motor tests who refused virostatic treatment (n = 33; 17 CDC II, 8 CDC III, 1 CDC IVA, 3 CDC IV C1, 1 CDC IVC2, 3 CDC IVD). Group C: patients with pathological motor performance and AZT treatment (n = 44; 2 CDC II, 12 CDC III, 5 CDC IVA, 20 CDC IVC1, 5 CDC IVD). In any case, treatment was started at a T-helper cell-count of less than 400/tzl or if the patient himself

18 |

desired treatment. None of the patients was lost to follow-up, but 5 died in group A, 3 in group B and 25 in group C after 2 years (see Statistics), Thus, after 2 years, in group A there were 10 individuals in stage CDC II, 3 CDC III, 1 CDC IVA, 1 CDC IVB, 1 CDC 1VC1 and 2 CDC IVC2 (n = 18), in group B 16 CDC II, 9 CDC III, 1 CDC IVA, 1 CDC IV C1, 1 CDC IVC2 and 2 CDC IVD (n = 30) a n d i n group C 2 CDC II, 11 CDC III, 2 CDC IVA and 4 CDC IVB (n = 19).

Imaging In all patients MRI scans were performed at study onset (DIA-Sonic, 0.35 tesla, axial scanning, TE = 60 msec, TR = 2500-3000 msec) and repeated once a year.

Motor testing Motor tests comprised the determination of tremor peak frequency (TPF, normal range: 8-12 H z ) , the maximal rate of rapid alternating movements (MRAM, normal range: 6-8 Hz), and the times to peak of most rapid contractions (MRC, normal range 120-140 msec). For methodological details of motor testing see Arendt et al. (1989).

Statistical analysis Differences of mean values between groups A, B and C were analyzed using Student's t-test for uncorrelated means. Significance of the differences between initial results and those after 24 months were analyzed by the t-test for dependent mean values. Mean values (Table 1) and statistics rely only on patients being alive at the end of the study. To compensate multiple comparisons, a Bonferroni adjustment was made for the p-values.

3. Results

After a 2-year period, 12 out of 18 (66.6%) of group A patients revealed no or only minor deterioration in motor performance; 5 patients had died, 2 of Kaposi sarcoma. 1 of lung tuberculosis, and 2 of subarachnoidal hemorrhage, 6 showed a marked deterioration in motor performance. 28 out of 33 of group B patients (84.9%) had deteriorated in motor performance after 2 years; 2 had died of Kaposi sarcoma, 1 of cerebral toxoplasmosis: 2 had stable motor values. 56.8% (n = 25) of the group C patients, belonging to CDC-stage IVC1 and D, had died after 2 years. 76% (n = 19) of these 25 patients had died from cerebral AIDS manifestations (n = 10 of HIV encephalopathy, 7 of HIVencephalopathy and an additional opportunistic brain infection, 2 of an opportunistic infection alone, diagnosis confirmed by autopsy in 57%} and 6 died from extracerebral AIDS manifestations. The latter 6 indi-

G. Arendt et al./Journal q( the Neurological Sciences 123 (1994) 180 185

82

E E

+1

+1 0

E +F tt~

(s

o;

cA +, ¢J

# 0

Z~

viduals showed only slightly abnormal motor performance at the beginning of the study. 19 group C patients were alive after 2 years: 4 of them with markedly pathological motor results in the beginning of the study were completely demented, 15 with moderately abnormal motor performance showed improved results after A Z T therapy. The distribution of the H1V patient groups A - C to CDC stages at study onset (see Methods) and after 2 years shows nearly no shift in stages for groups A and B; those group A and B patients who died during the observation period were CDC IV (AIDS) patients (see above) except for the 2 group A patients who died from cerebral bleeding. All of the group C patients, belonging to CDC stages IVCI, died within the observation period. Table 1 shows group statistical results of motor testing in controls and patient groups as well as the immune status at the beginning of the study and after 2 years. Group A patients revealed significant changes of TPF over the observation period ( p <0.05) and a significant decline in MRAM ( p < 0 . 0 1 ) . MRC and T-helper cell counts did not change significantly. TPF baseline values did not differ significantly from those of the other groups at the beginning of the study, whereas the other values (MRAM: A / B : p < 0.05, A / C : p < 0.05, MRC: A / B : p < 0.01, A / C : p < 0.001, T-helper cell count: A / C : p < 0.001) were significantly better at the beginning of the study in comparison to the other groups. These differences reflect the selection criterion. Group B patients showed a significant decline over time in all motor and immune variables (TPF: p < 0.01, MRAM: p < 0.001, MRC: p < 0.001, T4 helper cell count: p <0.001). The deterioration over time was more pronounced in comparison to group A. Group C patients showed no significant deterioration over time in TPF, MRC and T-helper cell count, but in MRAM there was a significant decline ( p < 0.001). In comparison to group A, MRAM baseline values were significantly lower ( p < 0.05) and deteriorated significantly over time (p <0.01). In contrast, no significant deterioration in MRAM was seen in comparison to group B patients. In group C patients, there was a temporary improvement under therapy in MRC- and T-helper cell counts lasting 9 months (MRC) and 6 months (T-helper cell count), respectively, followed by a deterioration reaching the initial values after the observation period. MRC was the only motor variable reflecting a transient influence of therapy, whereas MRAM declined continuously in all of the 3 patient groups. Fig. 1 gives typical examples of MRC recordings, the most sensitive motor value, of a group A and a group B patient and of 2 group C individuals in early and late infection stages. It can be derived from the figure that the group A patient deteriorated only slightly over

(;. Arendt et ul. / Journal of the Neurological Sciences 123 (1994) 180-185

time, whereas the group B patient showed markedly prolonged MRCs after 2 years. Motor improvement of group C patients under therapy in early stages in comparison to unchanged motor performance of treated full-blown AIDS patients is clearly demonstrated. However, response to therapy was not that clear-cut in all patients.

at the beginning of the study

after two years

I~3

Fig. 2a demonstrates the temporal development of TPF (upper part) and MRAM (lower part) group mean values, Fig. 2b that of MRC (upper part) and T-helper cell count (lower part) group mean values in the three groups. Group A patients slightly deteriorated in all of the variables, an effect which is more accentuated in group B. AZT treatment (group C patients) induced temporary recovery of TPF, MRC and T4 cells, therefore group statistical results in this group were not significantly deteriorating compared to the baseline values. Once more, we want to point out that the mean values were only calculated for those patients who could be followed up over the full observation period.

4. Discussion

itOOm

,

~100i

clin. and electrophysiol, asymptomatic patient (CDC II)

t--.---~ 100 i

,.o 100 mm

non-treatedp~Jent(CDCill)

tlO0 nm

,,

~ 100 lain

AZT-treated patient(CDC III)

~

- -

1~ ms

lOOms

AZT-treated patient (CDC IVC1) Examples o f most rapid contractions of group A, B a n d C patients at the beginning of the study and after two years. Fig. 1. M R C s o f p a t i e n t s in d i f f e r e n t C D C s t a g e s at t h e b e g i n n i n g o f the study and after 2 years.

Incidence and prevalence of AIDS-related brain disease are not yet clear. Clinicians are still debating (Janssen et al. 1989) significance of subclinical cerebral symptoms detected by neuropsyehologicat and electrophysiological methods for the course of HIV-related CNS disease. The question addressed in this study was whether sensitive motor tests could serve as predictors for cerebral disease progression. This study shows that over a 2-year period there is slight deterioration of patients with normal motor performance accompanied by decreasing T-helper cell counts. In contrast, irrespective of treatment, many patients in CDC stage IV C1 with initially pathological motor performance die within a 2-year period of cerebral AIDS manifestations, i,e., of opportunistic infections as well as of AIDS associated dementia. It is, of course, interesting whether patients with opportunistic cerebral infections will show a development of electrophysiologicat variables differing from that in patients with HIV-1 associated dementia. Resolving this issue, however, requires further studies relying on more patients. What can be derived from the actual data is that a high percentage of patients with pathological motor performance dies from cerebral AIDS manifestations. The patients with abnormal motor performance belonging at study onset to CDC stage IVC1 defined by pneumocystis carinii pneumonia or other AIDS defin. ing extracerebral infections are more likely to die from cerebral manifestations than patients in earlier stages. This could express global immune ineom~tence but it has to be figured out why these patients do not die from extracerebral AIDS manifestations. The group C patients with full-blown AIDS and only slightly abnormal motor performance at the beginning of the study did not show cerebral manifestations in our follow-up. Immunologically asymptomatic patients (CDC II and III) with pathological motor performance did not deteriorate markedly over time. Though motor testing and

G. Arendt et al. /Journal o( the Neurological Sciences 123 (1994) 180 185

184

,r,.,p,

• • (;lllll}+ II [] r;r{ml, f

~5 (a)

200

(b)

\1, , , ~

190

14

\

Ii

180

13

170 12 £ I

11 o

L,. 13. F-

10

160 150 140 130

9

120

8

110 100 BASE BASE

3

6

9

12

15

18

21

3

6

9 Time

Time

in

12

15

tB

24

in m o n t h s

monfhs • •

21

24

(;toup A G r o n p ~t

0 C;zoup I'

• • 1000

Group A Group B Croup C

Means

: SD

800

600

6 <~

AO0 5

200

0 BASE

3

6

I 12

9 Time

in

15

d 18

I 21

i 24

months

BASE

3

6

9 Time

12

15

18

21

24

in m o n t h s

Fig. 2. a: TPF development over time in asymptomatic HIV patients (A), symptomatic non-treated (B) and treated individuals (C), upper part. MRAMs development over time in asymptomatic HIV patients (A), symptomatic non-treated (B) and treated individuals (C), lower part. b: MRCs development over time in asymptomaticHIV patients (A), symptomatic non-treated (B) and treated individuals (C), upper part. T-helper cell count development over time in asymptomatic HIV patients (A), symptomatic non-treated (B) and treated individuals (C), lower part.

immune status are correlated, they do not always show the same temporal development. Therefore, both test procedures provide different information on the course of the disease. Careful evaluation of motor values could thus be a method for selecting patients requiring primary prophylaxis for cerebral diseases corresponding to pentacarinate in the case of pneumocystis carinii pneumonia. It will be crucial to follow-up C D C II patients with pathological motor performance over a longer period to find out whether they develop HIV-1 associated dementia earlier than patients with normal motor performance. In the present study, the time course of decline in T-helper cell count and motor test values was parallel in most of the patients. This observation needs further consideration. Perhaps, virus effects on T-helper cells and CNS function are similar. Other studies underlining positive therapy effects of zidovudine on T-helper cells and neuropsychological test performance of H1Vpatients (Bornstein et al. 1991) support this notion. Conversely, one could presume that better motor performance may result from stabilized physical health, either due to A Z T treatment or spontaneously. Both,

neuropathological and immunological studies are required to confirm parallel temporal development of motor performance and T-helper cell count (Arendt et al. 1992; Bornstein et al. 1991), recently shown in animal experiments (Murray et al. 1992). Another point of discussion revealed by this study, is the beginning of virostatic treatment in HIV-positive individuals. Nothing is known about the optimal time for treatment onset. To date, patients are treated when T-helper cell counts fall below 4 0 0 / u l or if a patient himself requires treatment. The question is whether early signs of cerebral involvement should be a reason for starting therapy. The data presented here show that patients without early signs of brain involvement deteriorate only slightly in electrophysiological variables over the observation period whereas individuals with initially pathological motor results deteriorate markedly over time. Obviously, the observation period and the number of patients enrolled in the study are not yet sufficient to judge on the sense of virostatic treatment of immunologically and neurologically asymptomatic patients. Thus, it must be clarified in future studies

(3. Arendt et aL /Journal of the Neurological Sciences 123 (1994) 180-185

whether asymptomatic patients will sooner or later reveal signs of brain involvement and whether patients treated early will show such symptoms to a lesser extent than non-treated individuals.

Acknowledgment The authors thank Dr. Andreas Kleinschmidt for carefully reading the manuscript and for his helpful comments.

References American Academy of Neurology, Working group, AIDS Task Force. (1991) Nomenclature and research case definitions for neurologic manifestations of human immunodeficiency virus-type 1 (HIV-I) infection. Neurology, 41:778-785 Arendt, G., Hefter, H., Elsing, C., Neuen-Jacob, E., Strohmeyer, G.. Freund, H.-J. (1989) Neue elektrophysiologische Befunde zur H~iufigkeit der Gehirnbeteiligung bei klinisch-neurologisch asymptomatischen HIV-Infizierten. Z. EEG-EMG, 20:280-287 Arendt, G., Hefter, H., Buescher, L., Hilperath, F., Elsing, C., Freund, H.-J. (1992) Improvement of motor performance of

185

HIV-positive patients under AZT therapy. Neurology, 42: 891896 Bornstein, R.A., NasraUah, H.A., Para, M,F.. Fass, R.J., Whitacrc, C.C., Rice, R.R., Jr. (1991) Rate of CD4 decline and neuropsychological performance in HIV infection. Arch. Neurol.. 48(7): 704-707 Fischl, M.A., Richman, D.D., Grieco, M.H. (1987) The efficacy o[ acidothymidine (AZT) in the treatmen! of patients with AIDS and AIDS-related complex. New Engt..1. Med., 317:!85-191 Hefter, H. (1991) Untersuchungen zur Zeitstruktur witlkiirlicher Bewegungen von Normalpersonen und Patienten mit motorischen St6rungen. MD Thesis, University of DiJsseldorL Diisseldorf. Janssen, R.S., Saykin, A.J., Cannon, L. (1989) Neurological and neuropsychological manifestations of HIV-1 Infection: association with AIDS-related complex but no! asymptomatic HIV-I infection. Ann. Neurol., 26:592-600 Murray, E.A., Rausch, D.M., Lendvay, J., Sharer, L.R., Eiden, L.E. (1992) Cognitive and motor impairments associated with SIV infection in rhesus monkeys. Science, 255:1246-1249 O11o. C.. Johnson, R., Grafman, J. (1991) Signs of cognitive change in HIV disease: an event-related brain potential study. Neurology, 4l: 209-215 Schmitt, F.A., Bigley, J.W., McKinnis, R., Logue, P.E., Evans, R.W., Drucker, J,L and the AZT Collaborative Working Group (1988) Neuropsychological outcome of zidovudine (AZT). Treatment of patients with AIDS and AIDS related complex. New Engl..1. Med., 319:1573-1578