AIDS PATIENT CARE and STDs

Volume 19, Number 10, 2005

© Mary Ann Liebert, Inc.

Cerebral Toxoplasmosis in HIV-Positive Patients in

Brazil: Clinical Features and Predictors of Treatment
Response in the HAART Era

JOSE E. VIDAL, M.D.,1 ADRIAN V. HERNANDEZ, M.D.,6

AUGUSTO C. PENALVA DE OLIVEIRA, Ph.D.,2,5 RAFI F. DAUAR, M.D.,3
SILAS PEREIRA BARBOSA Jr., Ph.D.,4 and ROBERTO FOCACCIA, Ph.D.1

ABSTRACT

A prospective study of 55 confirmed or presumptive cases of cerebral toxoplasmosis in HIV-

positive patients in Brazil was performed to describe clinical characteristics and to identify
predictive factors for clinical response to the anti-Toxoplasma treatment. Cerebral toxoplas-
mosis led to the diagnosis of HIV infection in 19 (35%) patients, whereas it was the AIDS-
defining disease in 41 (75%) patients. Of these, 22 (54%) patients were previously know to be
HIV-positive. At diagnosis of cerebral toxoplasmosis, only 4 (7%) patients were on highly ac-
tive antiretroviral therapy (HAART), and 6 (11%) were receiving primary cerebral toxoplas-
mosis prophylaxis. The mean CD4! cell count was 64.2 (" 69.1) cells per microliter. Forty-nine
patients (78%) showed alterations consistent with toxoplasmosis on brain computed tomogra-
phy. At 6 weeks of treatment, 23 (42%) patients had complete clinical response, 25 (46%) par-
tial response, and 7 (13%) died. Alteration of consciousness, Karnofsky score less than 70, psy-
chomotor slowing, hemoglobin less than 12 mg/dL, mental confusion, Glasgow Coma Scale
less than 12 were the main predictors of partial clinical response. All patients were placed on
HAART within the first 4 weeks of diagnosis of cerebral toxoplasmosis. One year after the
diagnosis, all available patients were on HAART and toxoplasmosis prophylaxis, and only 2
patients had relapse of cerebral toxoplasmosis. In Brazilian patients with AIDS, cerebral toxo-
plasmosis mainly occurs as an AIDS-defining disease, and causes significant morbidity and
mortality. Signs of neurologic deterioration predict an unfavorable response to the treatment.
Early start of HAART seems to be related to better survival and less relapses.

INTRODUCTION been obtained in Brazil, a developing country

with a free and universal access program to

T HE INTRODUCTION of highly active antiretro-

viral therapy (HAART) has significantly
decreased the mortality in patients with AIDS
in developed countries.1,2 Similar results have
HAART since 1996.3
The epidemiology of the central nervous
system (CNS) opportunistic diseases has also
changed after the introduction of the HAART,

Departaments of 1Infectology, 2Neurology, 3Neurosurgery, and 4Radiology, Institute of Infectious Diseases Emilio
Ribas, Sao Paulo, Brazil.
5Clinical Research Unit in Human Retrovirology, University of Campinas, Sao Paulo, Brazil.
6Center for Clinical Decision Sciences, Department of Public Health, Erasmus University Medical Center, Rotter-

dam, The Netherlands.

626
CEREBRAL TOXOPLASMOSIS IN PATIENTS WITH AIDS
leading to the decline of the incidence of the
main neurologic complications.4 However,
central nervous system (CNS) opportunistic in-
fections are still a main cause of death and dis-
ability in patients with AIDS,5,6 especially in
countries where HAART is not available.7,8
It has been suggested that cerebral toxoplas-
mosis may be just as common as primary CNS
lymphoma in the HAART era.9 However, de-
spite the decline in incidence and mortality of
cerebral toxoplasmosis,10,11 recent studies con-
firmed that it remains the most frequent cause
of expansive brain lesion and a highly preva-
lent disorder of the CNS, even in the late
HAART era.12–14
Cerebral toxoplasmosis is a common com-
plication in Brazil, despite its decrease in inci-
dence after the introduction of HAART.15 Cur-
rently, it is the third AIDS-defining disease,
after tuberculosis and Pneumocystis jirovecii
(formerly Pneumocystis carinii) pneumonia.16
We sought to describe the main characteris-
tics of a cohort of Brazilian patients with AIDS
with cerebral toxoplasmosis, and to identify
predictive factors of partial clinical response.
MATERIALS AND METHODS
Setting
This study was performed at the Institute of
Infectious Diseases Emílio Ribas, Sao Paulo,
Brazil. This center is a tertiary teaching hospi-
tal with 200 beds and it serves as a reference
center for AIDS patients in the State of Sao
Paulo.
Study design
We carried out a prospective longitudinal co-
hort from June 2001 to June 2003. We included
patients with HIV infection, a first episode of
cerebral toxoplasmosis, and without concomi-
tant CNS disorders. The protocol was ap-
proved by the hospital Ethics Committee and
the patients signed an informed consent.
Definitions
The definition for a cerebral toxoplasmosis
case was adapted from the Centers for Disease
627
Control and Prevention.17 A confirmed case re-
quired identification of T. gondii tachyzoites in
brain samples obtained through biopsy or
necropsy. The presumptive case required: (1)
recent start of the neurologic abnormalities
consistent with intracranial disease or alter-
ation on the level of consciousness, (2) evidence
in the imaging (computed tomography [CT] or
magnetic resonance imaging [MRI]) of a brain
lesion with mass effect or with highlight after
injection of contrast substance or both findings,
and (3) response to anti-Toxoplasma-specific
treatment. The treatment included sulfadi-
azine, pyrimethamine, and folinic acid at rec-
ommended doses.18
Evaluation and follow-up
During hospital admission, we collected de-
mographic, epidemiologic, and clinical infor-
mation. Later, one specialized physician per-
formed the clinical evaluations on the 14th and
42nd days of treatment, and 1 year after the di-
agnosis. A brain CT and/or MRI was per-
formed on admission and on the 14th day of
treatment. The second imaging study was eval-
uated by one specialist who was unaware of any
prior patient information. On admission, we re-
quested an complete blood count, serum im-
munoglobulin (IgM) and IgG for T. gondii (en-
zyme-linked immunosorbent assay [ELISA]),
and CD4! T-cell count. In addition, we com-
pared the serologic results of cerebral toxoplas-
mosis patients with the results of other group
of HIV-positive patients without cerebral toxo-
plasmosis and with a similar CD4! T-cell count:
30 patients with cryptococcal meningitis (posi-
tive culture for Cryptococcus neoformans on cere-
brospinal fluid [CSF]), and 28 patients with tu-
berculous meningitis (positive culture for
Mycobacterium tuberculosis on CSF).
We gave corticosteroids for expansive le-
sions with important mass effect (risk of cere-
bral herniation, compression of the ventricular
spaces or shift of the midline), and also for the
cases of diffuse cerebral edema. All patients re-
ceived HAART before the fourth week of anti-
Toxoplasma treatment, and secondary prophy-
laxis for cerebral toxoplasmosis after the sixth
week of anti-Toxoplasma treatment.
628
Outcomes
Complete clinical response. Remission of all
neurologic abnormalities detected at baseline
by the end of therapy (6 weeks).
Partial clinical response. Definite improve-
ment in any neurologic abnormalities, but not
complete resolution of all, and no deterioration
or persistent new findings after the first week
of therapy.19
Predictive factors
We evaluated predictive factors for partial
clinical response at the end of the sixth week
of treatment. The data used in the univariate
analysis included demographic variables (age,
gender), epidemiologic (risk behavior for ac-
quiring HIV infection, time of HIV infection,
cerebral toxoplasmosis as HIV-defining infec-
tion, cerebral toxoplasmosis as AIDS-defining
disease); clinical (duration of neurologic symp-
toms, use of antiretroviral therapy, use of
trimethoprim-sulfametoxazol, headache, men-
tal confusion, focal motor deficit, nausea/vom-
iting, visual alteration, meningism, cerebellar
syndrome, hemiparesis, ataxia, alterations of
the cranial nerves, sensitive alteration, aphasia,
fever, consciousness alteration, psychomotor
slowing, behavior alteration, Karnofsky Per-
formance Scale, Glasgow Coma Scale, hospital
infection); laboratory (hemoglobin, lympho-
cytes count, platelets count, CD4! T-cell count,
serum IgM for T. gondii, and serum qualitative
and quantitative IgG for T. gondii); and radio-
logical (category of tomographic alteration,
typical or atypical pattern).
Statistical analysis
The discrete variables were expressed as per-
centage and compared with !2 tests or Fisher’s
exact test. The continuous variables were ex-
pressed as mean " SD, and compared with the
Student’s t test. The values of p # 0.05 were
considered statistically significant. The uni-
variate analysis evaluated the association be-
tween individual patient variables and the
sixth week treatment outcome. The strength of
association was expressed as odds ratios (OR)
and 95% confidence intervals (95% CI).
VIDAL ET AL.
RESULTS
Epidemiology
During the study period, 1138 HIV-infected
patients were admitted to the hospital. From
these patients, 115 (10%) were diagnosed as
cerebral toxoplasmosis. Sixty patients were ex-
cluded because of history of cerebral toxoplas-
mosis (n $ 47), coinfections of the central
nervous system (5 tuberculosis and 2 crypto-
coccosis), lost to follow-up (n $ 4), and lack
of authorization to participate in the study
(n $ 2).
Fifty-five patients were included (33 males,
mean age, 36 years). The HIV exposure cate-
gories were: heterosexual contact, 32 (58%);
homosexual, 15 (27%); bisexual, 3 (5%); and
injection drug use, 8 (15%). Cerebral toxoplas-
mosis led to the diagnosis of HIV infection in
19 (35%) patients, and it defined AIDS in 41
(75%). Among patients who knew their HIV-
positive status, the mean time of infection was
4 years (2 months to 12 years).
Previous exposure to antiretroviral therapy
was found in 12 (22%) of the patients. At cere-
bral toxoplasmosis diagnosis, 4 (7%) of these
were on HAART, and 6 (11%) were receiving
cerebral toxoplasmosis prophylaxis.
Cerebral toxoplasmosis diagnosis
From the 55 patients, 2 (4%) showed defini-
tive diagnosis through necropsy studies, and
53 (96%) showed presumptive diagnosis. Pa-
tients with definitive diagnosis showed clinical
and radiologic diffuse encephalitis, without
visible focal lesions.
Clinical characteristics
The mean duration of neurological symp-
toms was 17 " 12 days. The main signs and
symptoms are shown in Table 1. The Glasgow
Coma Scale and Karnofsky Scale means were
13 and 60, respectively.
Laboratorial characteristics
Nineteen (35%) patients showed low platelet
counts. The mean CD4! T-cell count was 64 "
69 cells per microliter, and 42 (76%) patients
CEREBRAL TOXOPLASMOSIS IN PATIENTS WITH AIDS 629

TABLE 1. MAIN SIGNS AND SYMPTOMS IN FIFTY-FIVE
PATIENTS WITH CEREBRAL TOXOPLASMOSIS AND AIDS
Signs and symptoms
Headache
Psychomotor slowing
Hemiparesis
Confusion
Fever
Alteration of consciousness
Convulsions
Alteration of cranial nerve
Visual alteration
Ataxia
Meningism
Nausea, vomits
Involuntary movements
showed values less than 100 cells per micro-
liter. The mean hemoglobin value was 11 " 5
g/dL. Twenty-two (40%) patients showed he-
moglobin values less than 10 g/dL. Fifty-four
(98%) patients showed anti-Toxoplasma IgG an-
tibodies, and 51 (93%) had high titles (%150 UI).
When compared to the results of the patients
with other neurologic diseases, patients with
cerebral toxoplasmosis presented more cases
with anti-Toxoplasma IgG antibodies (p # 0.001)
and more patients with high titles of antibod-
ies (p # 0.001) (Table 2).
Radiologic findings
On admission, the CT identified alterations
in 53 (96%) cases. The CT findings were classi-
fied in 5 categories (Table 3). Categories 1 and
2 were classified as typical pattern: 49 (78%);
categories 3, 4, and 5 were classified as atypi-
cal pattern: 14 (22%).
From the 53 patients with abnormal CT, 38
(72%) showed multiple lesions, and 15 (28%)
single lesions. A control CT at 2 weeks showed
complete response in 8 (15%) patients, and par-
n (%) tial response in 43 (78%). From the 4 patients
without a control CT, 3 died, and, the serious-
40 (73) ness of the clinical condition of the fourth did
40 (73)
40 (73) not allow imaging.
28 (51) Among the patients who only showed one
25 (46) category of tomographic alteration (n $ 47), 37
25 (46)
14 (26)
had lesions classified as typical and 10 as atyp-
11 (20) ical. No significant difference in duration of
8 (15) neurologic symptoms was found between typ-
5 (9) ical and atypical patterns. However, 5 of the 9
4 (7)
3 (5) patients with atypical alterations started anti-
2 (4) Toxoplasma treatment with an average of 3 days
after hospital admission. All patients with typ-
ical alterations initiated their treatment the day
of admission.
Outcome findings
At 2 weeks of treatment, 3 (6%) patients died,
36 (66%) patients had partial clinical response,
and 16 (29%) complete clinical response. Two
deaths were secondary to neurologic worsen-
ing, whereas 1 death was the result of nosoco-
mial pneumonia. At 6 weeks of treatment, 7
(13%) patients died, 25 (46%) patients had par-
tial clinical response, and 23 (42%) complete
clinical response. Four patients died as a result
of nosocomial infections (3 nosocomial pneu-
monia and 1 catheter-associated infection).
At 1 year of follow-up, 9 (16%) patients died,
21 (38%) had partial clinical responses, and 25
(45%) complete clinical responses (Fig. 1). Two
deaths that occurred up to 6 weeks were caused
by diseases not related to the cerebral toxo-
plasmosis (invasive pneumococcal disease and
P. jirovecii pneumonia). From the 46 patients
evaluated 1 year after the diagnosis, only 2 (4%)

TABLE 2. SEROLOGIC CHARACTERISTICS OF FIFTY-FIVE PATIENTS WITH CEREBRAL TOXOPLASMOSIS

AND FIFTY-EIGHT PATIENTS WITH OTHER AIDS-DEFINING NEUROLOGIC DISEASES

Cerebral Other AIDS-defining

toxoplasmosis neurologic diseasesa
(n $ 55) (n $ 58) p value

Age, mean years " SD 36 " 7.9 34 " 6.9 0.5

Male, n (%) 33 (60) 35 (60) 0.9
CD4!, mean " SD 64 " 69. 60 " 65. 0.6
ELISA (!), n (%) 54 (98) 34 (59) #0.001
ELISA % 150 UI, n (%) 51 (93) 12 (21) #0.001
aCryptococcal meningitis or tuberculous meningitis.
SD, standard deviation; ELISA, enzyme-linked immunosorbent assay.
630 VIDAL ET AL.

TABLE 3. CATEGORIES OF THE TOMOGRAPHIC FINDINGS IN FIFTY-FIVE

PATIENTS WITH CEREBRAL TOXOPLASMOSIS AND AIDSa

Categories of alterations in the CT n (%)

1. Hypodense lesions with ring-enhancing contrast and perilesional edema 28 (44)

2. Hypodense lesions with nodular enhancing and perilesional edema 21 (33)
3. Hypodense lesion without contrast enhancing and with expansive effect 10 (16)
4. CT without alterations and MRI demonstrating focal lesions 2 (3)
5. Diffuse cerebral edema, without visible focal lesions 2 (3)

Eight patients showed two categories of alterations.

CT, computed tomography; MRI, magnetic resonance imaging.

showed a relapse associated with a history of at cerebral toxoplasmosis diagnosis (OR $ 0.3,
lack of adherence to the secondary prophylaxis. 95% CI 0.03–2.9).

Predictive factors of partial clinical response

Univariate analysis identified 9 predictive
factors of partial clinical response to treatment:
alteration of consciousness, Karnofsky scale
less than 70, psychomotor slowing, hemoglo-
bin less than 12 mg/dL, mental confusion,
Glasgow Coma Scale value less than 12, dura-
tion of neurologic symptoms, seizures, and
atypical CT pattern (Table 4). For each day in
the presence of neurologic symptoms, the risk
of partial response incremented in 9% (OR $
1.1, 95% CI 1.0–1.2). Although not significant,
the risk of partial response to treatment was re-
duced by the previous use of HAART (OR $
0.4, 95% CI 0.09–1.4), and the use of HAART
DISCUSSION
This study shows that confirmed or pre-
sumptive cerebral toxoplasmosis is a common
HIV opportunistic infection, and explains 10%
of admissions in a Brazilian AIDS reference
center. Cerebral toxoplasmosis allowed the
identification of the HIV infection in 35% of the
patients, and it was the AIDS-defining disease
in 75% of cases. We found that the predictive
factors of partial response to treatment in-
volved severe neurologic damage. We also ob-
served an important benefit in survival and re-
lapses of cerebral toxoplasmosis cases after of
early initiation of HAART.

FIG 1. Clinical evolution of 55 patients with cerebral toxoplasmosis and AIDS. CT, cerebral toxoplasmosis; NI, noso-
comial infection; PCP, Pneumocystis jirovecii pneumonia; IPD, invasive pneumococcal disease.
CEREBRAL TOXOPLASMOSIS IN PATIENTS WITH AIDS 631

TABLE 4. PREDICTIVE FACTORS OF CLINICAL RESPONSE IN PATIENTS WITH CEREBRAL TOXOPLASMOSIS AND AIDS

Variable Definition n (%) OR (95% CI) p value

Alteration of consciousness Present 19 (40) 22.3 (4.2–119.3) #0.001

Absent 29 (60)
Karnofsky Scale &70 24 (50) 16.8 (3.8–74.9) #0.001
%70 24 (50)
Psychomotor slowing Present 33 (69) 14.9 (2.8–78.8) 0.001
Absent 15 (31)
Hemoglobin (g/dL) &12 12 (25) 8.9 (1.7–46.7) 0.010
%12 36 (75)
Mental confusion Present 24 (50) 4.9 (1.4–16.5) 0.011
Absent 24 (50)
Glasgow Coma Scale &12 10 (21) 12.4 (1.4–107.7) 0.023
%12 38 (79)
Time of For each additional 1.1 (1.0–1.2) 0.030
neurologic day of neurologic
symptoms symptoms
Seizures Present 11 (23) 5.9 (1.1–31.2) 0.037
Absent 37 (77)
Tomographic pattern Atypical 9 (22) 5.8 (1.0–32.8) 0.045
Typical 32 (78)

OR, odds ratio; CI, confidence interval.

More than a third of our patients did not
know their HIV status. Therefore, these pa-
tients are prone to acquire severe infections as
the first manifestation of immunosuppression.
Approximately three quarters of our patients
showed cerebral toxoplasmosis as AIDS-defin-
ing disease. In the pre-HAART era, these num-
bers varied between 22% and 53%.19,21–25 This
diference shows the loss of opportunity to use
HAART, as it was previously described.26
The progressive dysfunction of the immune
system allows the reactivation of a latent Tox-
oplasma infection, which justifies the presence
of IgG antibodies against T. gondii. Less than
6% of the patients with cerebral toxoplasmosis
showed negative tests.27 Although it is not rou-
tinely used, our results are similar to studies
that suggested that high titles of antibodies
against T. gondii could be useful to identify pa-
tients with cerebral toxoplasmosis or at risk to
develop it.28,29 However, other authors dis-
agree.30,31
Classically, 80%–90% of patients with cere-
bral toxoplasmosis showed enhancement (ring
or nodular) after the injection of contrast.32 This
is a typical finding, however, the spectrum of
radiologic alterations is broad. Between 6%
and 20% of the cerebral toxoplasmosis cases
showed expansive lesions without contrast en-
hancement.24,25 Nine (16%) of our patients
demonstrated this alteration. Possible explana-
tions for this finding include the use of insuf-
ficient quantities of contrast or severe levels
of immunosuppression.32,33 The diagnosis of
cerebral toxoplasmosis with normal CT and al-
tered MRI is difficult. It was reported in 4% of
other cases,34 and it was found in 3% of our pa-
tients. We considered that special attention
should be given to the atypical alterations, with
focus on an early administration of anti-Toxo-
plasma treatment.
The percentages of clinical response to the
anti-Toxoplasma treatment range between 65%
and 95%.19,21,23,27,32 However, there is little in-
formation on the type of clinical response. Con-
sequently, some authors reported complete re-
sponse rates between 29% and 50%.19,22,23 Our
results showed that 29% and 42% of the pa-
tients had complete response at the second and
sixth week of treatment, respectively. This sug-
gests that cerebral toxoplasmosis continues to
cause significant functional incapacity in a high
proportion of patients with AIDS.
Most of our patients had presumptive diag-
nosis, and therefore, we could have included
patients with other diagnosis. This may explain
in part the high proportion of patients with par-
cial response to the treatment. We did not per-
632
form cerebral biopsy to any of these patients.
However, the probability to have chosen pa-
tients with other diagnosis is remote, because
of the strict clinical assessment and follow-up,
and the exclusion of other neurologic diseases.
On the other hand, the adherence to anti-Tox-
oplasma treatment was closely verified. Thus, it
is very unlikely that undertreatment may have
influenced the outcome of the patients.
Studies during the pre-HAART era reported
that 11%–13% of patients with cerebral toxo-
plasmosis died, despite the use of anti-Toxo-
plasma therapy.22,24 In our series, at the end of
the sixth week, 7 (13%) patients had died,
which demonstrates that cerebral toxoplasmo-
sis continues causing high lethality. From
these, 5 deaths were the result of nosocomial
infections. The high number of deaths related
to nosocomial infections has been described in
the pre-HAART era studies, and should be con-
sidered in the clinical evolution of the patients
with cerebral toxoplasmosis.21,23 A recent
study performed in an outpatient clinic in Sao
Paulo reported that 12 of 206 HIV-positive pa-
tients died during the follow-up. No cases of
cerebral toxoplasmosis were reported as cause
of death.35 In contrast to these patients, our pa-
tients (1) knew their HIV status but were not
on HAART or prophylaxis for T. gondii or (2)
did not know their HIV status and were not on
any pharmacologic treatment. Our patients be-
longed to a subgroup that has high risk to de-
velop cerebral toxoplasmosis or any other
AIDS-defining condition.
Associated factors related to poor outcome
of cerebral toxoplasmosis included: alteration
of the level of consciousness, fever, multiple le-
sions, history of Karposi’s sarcoma or P.
jirovecii pneumonia as AIDS-defining diseases,
use of anticonvulsivants, lymphocytes count
less than 24%, and time of diagnosis of
AIDS.14,19,23–25,35 We identified other predictive
factors of treatment response: Karnofsky Scale
less than 70, psychomotor slowing, hemoglo-
bin less than 12 mg/dL, mental confusion,
Glasgow Coma Scale less than 12, seizures,
atypical CT alterations, and time of the neuro-
logic symptoms. Most of these factors seem to
be related with a more severe neurologic sta-
tus. However, some variables (e.g., anemia and
low Karnofsky Scale) seem to be associated
VIDAL ET AL.
with the extended and severe immunodefi-
ciency.37 The finding of atypical tomographic
alterations associated with the partial clinical
response suggest extensive periods of the dis-
ease without treatment, probably due to diffi-
culties during the initial diagnosis approach.
The survival of patients with cerebral toxo-
plasmosis in the pre-HAART era was estimated
to be between 4 and 13 months.24,36,38 Our re-
sults confirm the benefit observed in recent
years.39 In addition, the proportion of relapses
without secondary prophylaxis was between
50% and 80%, and with prophylaxis between
20% and 30%.40 In our cohort, the adequate use
of both HAART and prophylaxis justified the
low rate of recurrences. This finding suggests
that the adherence to HAART and prophylaxis
could change the natural history of cerebral
toxoplasmosis in developing countries.
There are controversies about the appropri-
ate timing for initiation of HAART in patients
with cerebral toxoplasmosis. A recent study
showed a threefold reduction in the risk of de-
veloping a new AIDS-defining event or death
in patients who started HAART within the first
2 months after cerebral toxoplasmosis diagno-
sis, suggesting an important survival benefit.14
In concordance with this, all our patients started
HAART within the first month after diagnosis.
The use of primary prophylaxis presents high
efficacy in preventing cerebral toxoplasmosis,
and studies in the HAART era confirm that re-
main a main factor in reducing disease occur-
rence. A study reported that only 34% of anti-
retroviral-experienced patients with a CD4!
T-cell count less than 100 cells per microliter re-
ceived prophylaxis at cerebral toxoplasmosis di-
agnosis.14 Similarly, only 6 of 12 our patients
who had experienced antiretroviral therapy,
had CD4! T-cell count less than 100 cells per
microliter, and had positive serology for T.
gondii were using prophylaxis at the time of neu-
rologic diagnosis. These data suggest the failure
to use this simple and relatively inexpensive
strategy to avoid a potentially fatal disease.
Our study has some limitations. Our results
do not represent the full spectrum of Brazilian
patients. A broader source of the study popu-
lation (e.g., other centers and regions) would
improve the generalizability of the findings.
Furthermore, it could be argued that would be
CEREBRAL TOXOPLASMOSIS IN PATIENTS WITH AIDS
necessary to define predictors with multivari-
ate analysis. However, we decided to perform
only a univariate analysis because of the low
number of events. In this context, a multivari-
ate analysis may have been misleading because
it could have included nonpredictive variables
or have excluded true predictive variables.
In conclusion, cerebral toxoplasmosis is still
a common neurological complication in Brazil-
ian patients with AIDS, despite the availability
of HAART and effective prophylaxis. This fact
reflects a loss of diagnostic and therapeutic op-
portunities, and it highlights the need to opti-
mize current control and prevention programs.
Clinical, tomographic, and laboratory charac-
teristics, especially those related to the severity
of neurologic compromise, seem to be associ-
ated to a partial clinical response. Early intro-
duction of HAART seems to improve survival
and reduce relapses of cerebral toxoplasmosis.
ACKNOWLEDGMENTS
José E. Vidal was supported by the Coorde-
nação de Aperfeiçoamento de Pessoal de Nível
Superior (CAPES) do Ministério da Educação
do Brazil (001/0700/000186/2003). Adrián V.
Hernández was supported by the Netherlands
Organization for Scientific Research (ZON/
MW 908-02-117).
Preliminary results of this study were pre-
sented at the 5th International Symposium on
Neurovirology and HIV Molecular, Baltimore,
Maryland, September 2–6, 2003.
REFERENCES
1. Palella FJ Jr, Delaney KM, Moorman AC, et al. De-
clining morbidity and mortality among patients with
advanced human immunodeficiency virus infection.
HIV Outpatient Study Investigators. N Engl J Med
1998;338:853–860.
2. Mocroft A, Ledergerber B, Katlama C, et al. Decline
in the AIDS and deaths rates in the EuroSIDA study:
An observational study. Lancet 2003;362:22–29.
3. Marins JR, Jamal LF, Chen SY, et al. Dramatic im-
provement in survival among adult Brazilian AIDS
patients. AIDS 2003;17:1675–1682.
4. Sacktor N. The epidemiology of human immunode-
ficiency virus-associated neurological disease in the
era of highly active antiretroviral therapy. J Neurovi-
rol 2002;8(Suppl 2):115–121.
633
5. Maschke M, Kastrup O, Esser S, Ross B, Hengge U,
Hufnagel A. Incidence and prevalence of neurologi-
cal disorders associated with HIV since the introduc-
tion of highly active antiretroviral therapy (HAART).
J Neurol Neurosurg Psychiatry 2000;69:376–380.
6. Sacktor N, Lyles RH, Skolasky R, et al. HIV-associ-
ated neurologic disease incidence changes: Multi-
center AIDS Cohort Study, 1990–1998. Neurology
2001;56:257–260.
7. Mehari E, Haimanot R. Patterns of neuroAIDS in
Africa. J Neurovirol 2003;60(Suppl 1):14.
8. Shankar SK, Satishchandra P, Mahadevan A, et al. Neu-
rological disorders associated with HIV/AIDS Asian
perspective. J Neurovirol 2003;60(Suppl 1):13–14.
9. Olantinwo TF, Herbowy MT, Hewwitt RG. Toxo-
plasmic encephalitis and primary lymphoma of the
brain—The shift in epidemiology: A case series and
review of the literature. AIDS Read 2001;11:444–449.
10. Abgrall S, Rabaud C, Costagliola D, et al. Incidence
and risk factors for toxoplasmic encephalitis in hu-
man imunodeficiency virus-infected patients before
and during the highly active antiretroviral therapy.
Clin Infect Dis 2001;53:1747–1755.
11. Jones JL, Sehgal M, Maguire JH. Toxoplasmosis-asso-
ciated deaths among immunodeficiency virus-in-
fected persons in the United States, 1992–1998. Clin
Infect Dis 2002;34:1161.
12. Oliveira J, Greco D, Oliveira G, et al. HIV-related neu-
rological disorders in Brazil in the era of the highly
active antiretroviral therapy [Abstract ThPeB7371]. In:
14th International AIDS Conference. Barcelona: July
7–12, 2002.
13. Antinori A, Lorenzini P, Larussa D, et al. HIV-asso-
ciated neurological diseases: Trends of prevalence in
the late-HAART era. Data from the Italian Registry
Investigative NeuroAIDS. J Neurovirol 2004;10(Suppl
3):60–61.
14. Antinori A, Larussa D, Congolani A, et al. Prevalence,
associated factors, and prognostic determinants of
AIDS-related toxoplasmic encephalitis in the era of
advanced highly active antiretroviral therapy. Clin In-
fec Dis 2004;39:1681–1691.
15. Guimaraes M. Temporal trends in AIDS-associated
opportunistic infections in Brazil, 1980–1999. Cad
Saude Publica 2000;31:393–398.
16. State of Sao Paulo Health Department. Reference
and Trainee Center-STD/AIDS-SP. Epidemiol Bull
2003;XXII:1.
17. Centers for Disease Control and Prevention. 1993 re-
vised classification system for HIV infection and ex-
panded surveillance case definition for AIDS among
adolescents and adults. MMWR Morb Mortal Wkly
Rep 1992;41:1–19.
18. Berger JR, Simpson DM. Neurologic complications of
AIDS. In: Scheld WM, Whitley RJ, Durack DT, eds.
Infections of the Central Nervous System. Philadel-
phia: Lippincott-Raven Publishers, 1997;2386–2393.
19. Luft BJ, Hafner R, Korzun AH, et al. Toxoplasmic en-
cephalitis in patients with the acquired immunodefi-
ciency syndrome. N Engl J Med 1993;329:995–1000.
634
20. Ministry of Health, Brazil. AIDS Epidemiol Bull
2001;XIV:1.
21. Navia BA, Petito CK, Gold JW, Cho ES, Jordan BD,
Price RW. Cerebral toxoplasmosis complicating the
acquired immune deficiency síndrome: Clinical and
neuropathological findings in 27 patients. Ann Neu-
rol 1986;19:224–238.
22. Leport C, Raffi F, Matheron S, et al. Treatment of cen-
tral nervous system oxoplasmosis with pyrimeta-
mine/sulfadizine combination in 35 patients with the
acquired immunodeficiency syndrome. Am J Med
1988;84:94–100.
23. Dannemann B, McCutchan JA, Israelski D, et al. Treat-
ment of toxoplasmic encephalitis in patients with
AIDS. A randomized trial comparing pyrimethamine
plus clindamycin to pyrimethamine plus sulfadi-
azine. Ann Intern Med 1992;116:33–43.
24. Renold C, Sugar A, Chave JP, et al. Toxplasma en-
cephalitis in patients with the acquired immunodefi-
cency syndrome. Medicine 1992;71:224–239.
25. Porter SB, Sande MA. Toxoplasmosis of the central
nervous system in the acquired immunodeficiency
syndrome. N Engl J Med 1992;327:1643–1648.
26. Manfredi R, Chiodo CF. Lack of change in the distri-
bution of AIDS-defining opportunistic diseases and
the related degree of immunodeficiency during the
periods before and after the introduction of highly ac-
tive antiretroviral therapy. Eur J Microbiol Infect Dis
2001;20:410–413.
27. Skiest DJ. Focal neurological disease in patients with
acquired immunodeficiency syndrome. Clin Infect
Dis 2002;34:103–115.
28. Hellerbrand C, Goebel FD, Disko R. High predictive
value of Toxoplasma gondii IgG antibody levels in HIV-
infected patients for diagnosis of cerebral toxoplas-
mosis. Eur J Clin Microb Infect Dis 1996;15:869–872.
29. Derouin F, Leport C, Pueyo S, et al. Predictive value
of Toxoplasma gondii antibody titres on the ocurrence
of toxoplasmic encephalitis in HIV-infected patients.
AIDS 1996;10:1521–1527.
30. Raffi F, Franck J, Pelloux H, et al. Specific anti-Toxo-
plasma IgG antibody immunoblot profiles in patients
with AIDS-associated Toxoplasma encephalitis. Diag
Microbiol Infect Dis 1999;34:51–56.
31. Leport C, Franck J, Chene G, et al. Immunoblot pro-
file as preditor of toxoplasmic encephalitis in patients
infected with immunodeficiency virus. Clin Diagn
Lab Immunol 2001;8:579–584.
VIDAL ET AL.
32. Cohen BA. Neurological manifestations of toxoplas-
mosis in AIDS. Semin Neurol 1999;19:201–211.
33. Walot I, Miller BL, Chang L, Mehringer CM. Neu-
roimaging finding in patients with AIDS. Clin Infect
Dis 1996;22:906–919.
34. Knobel H, Guelar A, Graus F, Miro JM, Padro S, Mer-
cader JM. Toxoplasmic encephalitis with normal CT
scan and pathologic MRI. Am J Med 1995;99:220–221.
35. Casseb J, Marcondes Fonseca LA, Veiga AP, et al.
AIDS incidence and mortality in a hospital-based co-
hort of HIV-1–seropositive patients receiving highly
active antiretroviral therapy in Sao Paulo, Brazil.
AIDS Patients Care STDs 2003;17:447–452.
36. Cohn JA, McMeeking A, Cohen W, Jacobs J, Holzman
RS. Evaluation of the policy of empiric treatment of
suspected Toxoplasma encephalitis in patients with the
acquired immunodeficiency syndrome. Am J Med
1989;86:521–527.
37. Mocroft A, Kirk O, Barton SE, et al. Anaemia is an in-
dependent predictive marker for clinical prognosis
in HIV infected patients from across Europe. AIDS
1999;13:943–950.
38. Haverkos HW, the TE Study Group. Assessment of
therapy for toxoplasma encephalitis. Am J Med
1987;82:907–914.
39. Guzman MT, Gaspar G, Castilla V, et al. Survival after
first episodes of HIV-opportunistic diseases and its
changes over the decade 1990–2000 in the COMESEM
cohort study [Abstract TuPeC4732]. Presented at: XIV
International AIDS Conference. Barcelona: 2002.
40. Merigan TC Jr, Bartlett JG, Bolognesi D. Neurologic
complications of HIV-1 infection and AIDS. In: Meri-
gan TC Jr, Bartlett JG, Bolognesi D, eds. Textbook of
AIDS Medicine. Philadelphia: William & Wilkins,
1999;477–497.
Address reprint requests to:
Dr. José E. Vidal
Rua Capote Valente 668
Apto. 78
Jardim América
CEP 05409-002
Sao Paulo
Brazil
E-mail: jbermudez@emilioribas.sp.gov.br