ORIGINAL ARTICLE Scand J Infect Dis 28: 71 -73, 1996

Risk of Toxoplasmic Encephalitis in AIDS Patients: Indications for Prophylaxis OONA DUNLOP', VIBEKE ROOTWELT'g2, METTE SANNES', ANNE K. GOPLEN3, MICHAEL ABDELNOOR2, KJELL SKAUG4, KARE BAKLIEN', ANNE SKAR4, KJETIL MELBY4, BJ0RN MYRVANG' and JOHAN N. BRUUN' From the 'Department of Infectious Diseases, 'Clinical Research Unit, 3Department of Pathology, 4Department of Microbiology and the 'Blood Bank and Department of Immunology, Ullevdl University Hospital, Oslo, Norway

To establish the indications for primary prophylaxis against toxoplasmic encephalitis in the Norwegian HIV-positive population, we estimated the incidence of toxoplasmic encephalitis, and related the degree of immunodeficiency and the presence of IgC antibodies against Toxoplasma gondii (T. gondii) to the development of toxoplasmic encephalitis. This retrospective study includes all HIV-positive patients at our hospital from April 1983 to October 1994 (n = 705). A total of 238 patients had AIDS, which represents almost 90% of all AIDS patients in Oslo. Autopsy was done in over 70% of the patients who died during this period; 19 patients developed toxoplasmic encephalitis (8.0Y0). The median CD4 cell count was 75 x lo6 cell/l (range 0-280) at the time of diagnosis of toxoplasmic encephalitis. T. gondii serology was studied in 698 (99.0%) of the patients, and was found positive for 17.8%. Of the patients with toxoplasmic encephalitis 18/19 had IgC antibodies against T. gondii and of the 40 AIDS patients who bad anti-T. gondii IgC, 18 (45%) developed toxoplasmic encephalitis. We conclude that there is indication for prophylactic treatment of HIV positive patients who have IgC antibodies against T. gondii and who have fewer than 200 x lo6 CD4 cells/l.

0. Bunlop, MD, Department of Infectious Diseases, Medical Clinic, Ullevll University Hospital, N-0407 Oslo, Norway

INTRODUCTION

Toxoplasmic encephalitis is a common opportunistic infec- tion in HIV-positive patients with an incidence ranging from 3 to 50% (1). The infection is usually due to reactiva- tion of Toxoplasma gondii (T. gondii) and is associated with considerable morbidity and mortality, even with im- proved diagnostic procedures and adequate treatment ( 2 ) .

Several therapeutic regimes have a prophylactic effect. There is at present no consensus internationally about the indications for prophylaxis.

By estimating the incidence of toxoplasmic encephalitis, the association with IgG anti-T. gondii antibodies and the level of CD4-positive T cells at the time of diagnosis of toxoplasmic encephalitis, we wished to determine the risk of developing toxoplasmic encephalitis. Based on these results, the purpose of this study was to establish the indications for prophylactic treatment for toxoplasmic encephalitis in the Norwegian AIDS population.

PATIENTS AND METHODS Incidence The Department of Infectious Diseases at UllevPl Hospital is responsible for the treatment of all adult HIV symptomatic pa- tients (except hemophiliacs) in Oslo. From the beginning of the epidemic in 1983 to October 1994 we have treated 705 HIV-posi- tive patients. This represents over 85% of AIDS patients from Oslo and over half of all HIV positive patients in Norway (3). Through- out the epidemic we have had an autopsy rate of 73%.

All patient charts have been reviewed retrospectively by a clini- cian (O.D. or J.N.B.). The diagnosis of AIDS was based on the CDC criteria, usually after a plenary discussion in the department,

and adjusted according to the criteria of 1993 (4). The presumptive diagnosis of toxoplasmic encephalitis was based on the combina- tion of clinical signs of focal central nervous system abnormality or reduced level of consciousness, focal lesions shown by computized tomography (CT) or nuclear magnetic resonance imaging, and response to treatment within 1 month. Biopsy was not done. A definite diagnosis was made in 7 cases by neuroautopsy.

Basic patient data such as HIV-related and AIDS-defining com- plications, CD4-positive T-cell levels and serology were registered in a computer database together with patient demographics. All medication was registered and categorized as either antiretroviral, antitoxoplasmic (including medication with only anecdotal effect) and other medication.

Immunodeficiency

The level of CDCpositive T cells was used as a measure of immunodeficiency. CD4 cells are measured on a routine basis for all patients at regular intervals (every 3 months for patients with symptoms or CD4 values below 400 x lo6 cells/l and every 6 months for asymptomatics), together with a clinical evaluation.

The CD4 analysis was done by flow cytometry (FACSCAN Flowcytometer, Becton-Dickinson). For the last 5 years we have used a two-colour immunofluorescence whole-blood lysis method. Absolute lymphocyte subset numbers were calculated, based on total and differential blood counts. The reference range was 400- 1,500 x lo6 cells/l.

T. gondii serology

T. gondii IgG serological testing was performed as part of routine investigations in HIV-positive patients. A dye test was used until 1989; from April 1989 we used Toxoscreen DA (direct agglutina- tion), (BioMerieux); from April 1991 Toxo IgG Micro EIA 2 (ELISA) (BioMerieux) and from September 1993 Platelia Toxo IgG (ELISA) (Sanofi, Diagnostics Pasteur). For patients for whom serological data were missing, stored frozen sera (routine storage) were used (n = 135).

0 1996 Scandinavian University Press. ISSN 0036-5548

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cMas

ter

Uni

vers

ity o

n 11

/12/

14Fo

r pe

rson

al u

se o

nly.

12 0. Dunlop et al. Scand J Infect Dis 28

Patients

T. gondii serological investigations were carried out in 698 of 705 (99.0%) registered HIV-positive patients (Fig. 2). IgG antibodies to T. gondii was found in 124 individuals (18%).

Of the patients with toxoplasmic encephalitis, 18 of 19 had positive T. gondii serological findings either at the time of diagnosis or previously. T. gondii IgG was available for 17 of 19 patients 4 months or more before diagnosis (IgG antibodies to T. gondii, present for 16/17) and for 18 at the time of diagnosis. One patient was T. gondii IgG negative repeatedly both 3 years before and at diagnosis; T. gondii IgM was also negative in this patient. He had clinical signs of CNS effects and a focal non-enhancing lesion was shown on CT. He responded slowly and only partly to treatment. At autopsy 3 months later there was cytomegalovirus en-

Fig. I . CDCcell counts at time of diagnosis of toxoplasmic en- cephalitis, for each of the patients. In 1 patient CD4 cells were not analysed.

RESULTS

We identified 19 (8.0%) cases of toxoplasmic encephalitis among our 238 AIDS patients. This was the first AIDS- defining event in 8 of the 19 patients (42.1%). Of these 19 patients, 17 were treated for toxoplasmic encephalitis, 6 of the latter had a dramatic improvement of treatment and 10 improved but had sequelae. Three patients died from their initial attack of toxoplasmic encephalitis, and an additional 7 had recurrence after initial improvement. Of the 17 patients followed to death, all died with severe central nervous system symptoms, and toxoplasmic encephalitis was found in 7 of the 10 patients who underwent autopsy. In 2 patients the diagnosis of toxoplasmic encephalitis had not been made before autopsy. Toxoplasmic pneumonia was found at autopsy in 1 additional patient without encephalitis.

Most patients had a low CD4 count at the time of diagnosis of toxoplasmic encephalitis (Fig. I), with a me- dian of 75 x lo6 cells/l (range 0-280).

cephalitis in the same area as the suspected toxoplasmic lesion. The patient however fulfils our diagnostic criteria for toxoplasmic encephalitis.

The prognostic importance of a positive IgG anti-T. gondii was addressed by evaluating all IgG-positive AIDS patients not lost to follow up (n =40). We found that 18 (45%) developed toxoplasmic encephalitis (Fig. 2). One additional patient developed toxoplasmic pneumonia (diag- nosis made at autopsy). Most patients (n = 12) had used some possible prophylactic medication, but only few (n = 2) had been on systematic prophylaxis. There was no signifi- cant difference in medication with possible prophylactic treatment between the group of IgG anti-T. gondii-positive AIDS patients who developed toxoplasmic encephalitis compared with those who did not. In 11 patients who did not develop toxoplasmic encephalitis and had CDCcell counts <200 x lo6 cells/l, the mean time without any pro- phylactic treatment was > 7 months. No patients developed toxoplasmic encephalitis while on trimethoprim-sulpha- methoxazole medication (treatment or prophylaxis).

DISCUSSION

We found that 8.0% of the AIDS patients developed toxo- plasmic encephalitis. The risk of developing toxoplasmic

HIV positive patients

IgG positive IgG negative

0 A:;2\

L F t - AIDS

Dead Dead R 4 \

Fig. 2. Flow chart showing the distri- bution of HIV-positive patients ac- cording to T. gondii IgG, AIDS and toxoplasmic encephalitis. IgC. T. 0 36 \

I

gondii IgG; TOXO, toxoplasmic en- TOXO Non-toxo Non-toxo TOXO Non-toxo TOXO cephalitis. 16 20 2 2 147 1

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cMas

ter

Uni

vers

ity o

n 11

/12/

14Fo

r pe

rson

al u

se o

nly.

Scand J Infect Dis 28 Toxoalasmic encephalitis in HIV 73

encephalitis in an HIV-positive population varies substan- tially from country to country. In the United States this complication develops in 3-10% of patients with AIDS, while in Europe and Africa up to 50% of AIDS patients will develop this complication (1). The difference is proba- bly due to several factors. The most important is the prevalence of T. gondii infection in the population. Epi- demiological studies have shown that 13% of the pop- ulation in Oslo have IgG antibodies to T. gondii (5, and Pi1 Jenum, personal communication). This prevalence is much lower than found in many other European coun- tries such as France (70%), but similar to most of the USA (1).

In USA about 30% of the AIDS patients who are IgG anti-T. gondii-positive eventually will develop toxoplasmic encephalitis, whereas in France about 50% of these patients develop this complication. This difference is still unex- plained but has been attributed to different strains of T. gondii or a larger parasite load in the French population. Although the Norwegian AIDS population has a low inci- dence of IgG anti-T. gondii positivity, similar to the USA incidence (1), there seems to be a European risk of develop- ing toxoplasmic encephalitis, as 45% of T. gondii IgG-posi- tive patients eventually developed this complication in our study. This may give support to the theory of different T. gondii strains rather than a difference in parasite load. Strains with different virulences are well known from labo- ratory studies of T. gondii (6). As medication with possible prophylactic effect has only been used systematically in recent years this only seems to explain some of the differ- ence between the USA and Europe.

The incidence of positive serological findings for T. gondii in this AIDS population is similar to that found in a previous Norwegian study (7), and the incidence of toxoplasmic encephalitis is very similar to reports from a Danish AIDS population (8). All except 1 of our patients with toxoplasmic encephalitis had positive IgG anti-T. gondii either at the time of diagnosis or previously.

It is possible that there were patients who had toxoplas- mic encephalitis without fulfilling the diagnostica criteria of this study. However autopsy was performed in over 75% of cases. Treatment for presumed toxoplasmic encephalitis was initiated in an additional 14 patients who did not respond, and at autopsy most of these patients had central nervous system lymphoma ( n = 11) or other brain lesions (unpublished data). It is therefore unlikely that the study

severely underestimates the true incidence ot toxoplasmic encephalitis in this population.

Most studies of toxoplasmic encephalitis in AIDS have found median CD4-cell numbers varying between 30 and 50 at the first appearance of this complication. Although in our study most of toxoplasmic encephalitis occurred when there were low CD4 levels, 5 of 19 patients had CD4 cell numbers over 100. Thus, we recommend that prophylaxis should be considered even for cell numbers at about 200.

Given the high rate of morbidity and also the substantial risk of mortality, we conclude that HIV positive patients who are also IgG anti-T. gondii-positive and develop im- munodeficiency are at high risk of developing toxoplasmic encephalitis (almost 50%) and should be given primary prophylactic treatment against this complication. This con- clusion is supported by previous studies and reviews (9, 10).

We now use trimethoprim-sulfamethoxazole, 2 single strength doses, 3 times weekly, which also is our standard prophylaxis against Pneumocystis carinii pneumonia.

REFERENCES 1 . Luft BJ, Remington JS. Toxoplasmic Encephalitis in AIDS.

Clin Infect Dis 15: 211-222, 1992. 2. Porter SB, Sande MA. Toxoplasmosis of the central nervous

system in the acquired immunodeficiency syndrome. N Engl J Med 327: 1643-1648, 1992.

3. Nilsen a, Hasseltvedt V, Lystad A. AIDS-situasjonen i Norge pr 31. desember 1994. Meldesystem for infeksjonsykdommer (MSIS) 23(4): 1995.

4. 1993 revised classification system for HIV infection and ex- panded surveillance case definition for AIDS among adoles- cents and adults. MMWR 41: 1-19, 1993.

5. Stray-Pedersen B, Lorentzen-Styr A-M. The prevalence of Toxoplasma antibodies among 11,736 pregnant women in Norway. Scand J Infect Dis 11: 159-165, 1979.

6. Reikvam A, Lorentzen-Styr AM. Virulence of different strains of Toxoplasma gondii and host response in mice. Nature 261 : 508-509, 1976.

7. Fle RW, Nilsen A, Voltersvik P, Haukenes G. Serum antibod- ies to viral pathogens and Toxoplasma gondii in HIV-infected individuals. APMIS 101: 946-952, 1993.

8. Smith E, Pers C, Aschow C, Mathiesen L. Cerebral toxoplas- mosis in Danish AIDS patients. Scand J Infect Dis 23: 703- 709, 1991.

9. Carr A, Tindall B, Brew BJ, Marriott DJ, Harkness JL. Penny R, Cooper DA. Low-dose trimethoprim-sulfamethoxazole pro- phylaxis for toxoplasmic encephalitis in patients with AIDS. Ann Intern Med 1 17: 106- 1 11, 1992.

10. Gallant JE, Moore RD, Chaisson RE. Prophylaxis for oppor- tunistic infections in patients with HIV Infection. Ann Intern Med 120: 932-944. 1994.

Submitted July 10, 1995; accepted October 16, 1995

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cMas

ter

Uni

vers

ity o

n 11

/12/

14Fo

r pe

rson

al u

se o

nly.