DOI: 10.1111/ j.1468-1293.2007.00530.

x
HIV Medicine (2008), 9, 126–130 r 2008 British HIV Association

ORIGINAL RESEARCH

Isosporiasis in patients with HIV infection in the highly

active antiretroviral therapy era in France
M Lagrange-Xélot,1,2 R Porcher,2,3 C Sarfati,2,4 N de Castro,1,2 O Carel,1 J-D Magnier,5 V Delcey2,5 and J-M Molina1,2
1
Department of Infectious Diseases, Hôpital Saint Louis, Paris, France, 2University of Paris 7, Paris, France, 3Department of
Biostatistics, Hôpital Saint Louis, Paris, France, 4Laboratory of Parasitology-Mycology, Hôpital Saint Louis, Paris, France
and 5Internal Medicine Unit, Hôpital Lariboisière, Assistance Publique Hôpitaux de Paris, Paris, France

Background
Isosporiasis, a rare cause of diarrhoea among HIV-infected patients in the pre-highly active
antiretroviral therapy (HAART) era, seems to be re-emerging.
Methods
A retrospective study was carried out for the period 1995–2003 in two hospitals in Paris to describe
the prevalence, clinical characteristics and therapeutic outcome of isosporiasis in HIV-infected
patients, and to compare the findings with those for cryptosporidiosis and microsporidiosis.
Results
The prevalence of isosporiasis increased from 0.4 per 1000 patients in the pre-HAART era
(1995–1996) to 4.4 per 1000 patients in the HAART era (2001–2003), whereas the prevalence
of cryptosporidiosis and microsporidiosis decreased. Compared with patients with either
cryptosporidiosis (n 5 91) or microsporidiosis (n 5 58), patients with isosporiasis (n 5 28) more
frequently originated from sub-Saharan Africa (72%), were more frequently female and
heterosexual, and had a higher median CD4 count at diagnosis (142 cells/mL). All patients
with isosporiasis presented with diarrhoea, which was severe enough to lead to hospital admission
for 60% of them. Fever was uncommon (7%). All patients were treated for isosporiasis, 27 of them
with cotrimoxazole. Relapse of isosporiasis occurred in six of 16 patients (38%) despite maintenance
cotrimoxazole therapy and HAART.
Conclusion
Isosporiasis in France occurs mostly in patients emigrating from sub-Saharan Africa and can induce
severe diarrhoea. Relapse is common despite cotrimoxazole maintenance therapy.
Keywords: cotrimoxazole, diarrhoea, HIV, isosporiasis, relapse, sub-Saharan Africa
Received: 7 September 2007, accepted 2 November 2007

intermittent shedding of Isospora belli cysts in stools [10].

Introduction
Isosporiasis was a rare cause of chronic diarrhoea among
HIV-infected patients in France and the USA in the pre-
highly active antiretroviral therapy (HAART) era, account-
ing for only 2% of all causes of diarrhoea, compared with
10–19% in developing countries in Africa and South
America [1–9]. Its diagnosis relies on direct parasitological
stool examination, which has to be repeated because of the
Correspondence: Dr Marie Lagrange-Xélot, Service des Maladies
Infectieuses, Hôpital Saint Louis, 1, avenue Claude Vellefaux 75010 Paris,
France. Tel: 1 33 1 4249 4585; fax: 1 33 1 4249 4820; e-mail:
marie.lagrange@sls.aphp.fr
Treatment relies mainly on cotrimoxazole, which has
proven effective in randomized trials, with prompt resolu-
tion of the diarrhoea and clearance of the parasite from
stools. However, long-term maintenance therapy with
cotrimoxazole is needed to avoid relapses, which occur in
nearly 50% of patients following discontinuation of initial
therapy. Few data have been published on isosporiasis
since the use of HAART became widespread. Following
the recent occurrence of a few cases of isosporiasis in our
hospital in Paris, we wished to describe the epidemio-
logical, clinical and therapeutic outcomes of this
parasitic infection among HIV-infected patients in the
HAART era.

126

Methods
We retrospectively examined all cases of isosporiasis
occurring in HIV-infected patients using the records of
the Parasitology Laboratory of the Saint Louis Hospital in
Paris from May 1995 to September 2003. This laboratory
performs all parasitic stool examinations for HIV-infected
patients under care at both the Lariboisière and Saint Louis
Hospitals in Paris. The diagnosis of isosporiasis was made
on direct stool examination or intestinal biopsies after
Kinyoun coloration, as previously described [11]. The
medical records from all patients were reviewed by the
same investigator (MLX) and the following data were
analysed: age, sex, country of origin, HIV risk factors, date
of HIV infection, previous opportunistic infections, CD4 cell
count, plasma HIV RNA level, Centers for Disease Control
and Prevention clinical classification and the use of
HAART. These characteristics were also recorded for HIV-
infected patients diagnosed during the same period in the
Parasitology Laboratory with either microsporidiosis or
cryptosporidiosis, two other intestinal opportunistic infec-
tions [12]. Furthermore, for each patient for whom a
diagnosis of isosporiasis was made, the following data were
recorded from clinical charts: presence and duration of
diarrhoea (defined as more than three loose or liquid stools
per day for more than 48 h), presence of fever (temperature
higher than 37.5 1C), weight loss (in kg), presence of
alithiasic cholecystitis (defined as right hypochondrium
pain and thickening of the gall bladder wall on ultrasono-
graphy) and/or cholangitis (elevated alkaline phosphatase
and dilatation of intra- and extra-hepatic bile ducts on
ultrasonography), initial therapy for isosporiasis, use of
maintenance therapy and relapses (defined as the reap-
pearance of oocysts in stools after at least 7 days of
appropriate therapy, with recurrence of diarrhoea).
Statistical analysis
In order to compare the prevalences of the three
opportunistic intestinal infections (microsporidiosis, cryp-
tosporidiosis and isosporiasis) in the pre- and post-HAART
eras, we defined two periods: 1995–1996 and 2001–2003.
The numbers of patients followed in each period in the two
hospitals were obtained from our computerized database
(French Hospital Database) [13].
The baseline characteristics of patients with each of these
protozoal intestinal infections were first compared using
Fisher’s exact test for qualitative variables and the Kruskal–
Wallis rank-sum test for quantitative variables. In the case
of a significant difference in the three-group comparison,
post-hoc two-by-two group comparisons were performed
using either Fisher’s exact test or the Wilcoxon rank-sum
r 2008 British HIV Association HIV Medicine (2008) 9, 126–130
Isosporiasis in the HAART era 127
2.5
Prevalence (% of HIV-infected patients)
Isosporiasis
2
Cryptosporidiosis
Microsporidiosis
1.5
1
0.5
0
1995 1996 1997 1998 1999 2000 2001 2002 2003
Year
Fig. 1 Annual prevalence of isosporiasis, cryptosporidiosis and
microsporidiosis at Lariboisière and Saint Louis Hospitals from
1995 to 2003.
test, using Hochberg’s correction for multiple testing [14].
Similarly, we compared the characteristics of patients, with
a minimum of 12 months of follow-up, with or without a
relapse of isosporiasis, using Fisher’s exact test or the
Wilcoxon rank-sum test. The cumulated incidence rate of
the first relapse of isosporiasis was estimated by the
Kaplan–Meier method. All tests were two-sided at the
0.05 level.
Results
Prevalence and baseline characteristics
Between January 1995 and September 2003, 28 patients
were diagnosed with isosporiasis and included in our study.
Isosporiasis was diagnosed on parasitological stool exam-
ination in 28 patients, and duodenal biopsy in three
patients. I. belli oocysts were detected in the first stool
sample in 26 of 28 patients. In one patient, only the fourth
stool sample yielded the parasite. The annual prevalences
of isosporiasis, cryptosporidiosis and microsporidiosis
during the study period are shown in Fig. 1. Compared
with the pre-HAART era (prevalence of 0.4 per 1000
patients in 1995–1996), the prevalence of isosporiasis
increased in the post-HAART era (prevalence of 4.4 per
1000 patients in 2001–2003; P 5 0.001). The opposite was
true for both cryptosporidiosis and microsporidiosis, the
prevalences of which decreased from 19 and 12 per 1000
patients in the pre-HAART era to 2.6 and 1.3 per 1000
patients in the post-HAART era, respectively (both
Po0.0001).
Comparing the baseline characteristics of patients
with isosporiasis to those with cryptosporidiosis or
128 M Lagrange-Xélot et al.

Table 1 Baseline characteristics of patients with isosporiasis, cryptosporidiosis and microsporidiosis in Lariboisière and Saint Louis Hospitals
(1 January 1995 to 31 December 2003) and P-values for comparison of isosporiasis vs. cryptosporidiosis and microsporidiosis

Isosporiasis Cryptosporidiosis Microsporidiosis Total population P

Number of patients 28 91 58 6827

Sex ratio (male/female) 1.8 (18/10) 3.3 (70/21) 28 (56/2) 3.0 (5122/1705) 0.0001
Age (years) [mean (range)] 34 (24–50) 40 (21–66) 39 (25–62) 41 (17–87) 0.005
Geographical provenance [n (%)]*
Sub-Saharan Africa 20 (72%) 12 (13%) 9 (16%) 1625 (27%) o0.0001
North Africa 2 (7%) 3 (3%) 2 (3%) 464 (8%)
Latin America 2 (7%) 1 (1%) 0 282 (5%)
Asia 2 (7%) 3 (3%) 0 102 (2%)
Europe/North America 2 (7%) 72 (79%) 47 (81%) 3629 (59%)
Route of HIV infection [n (%)]
Heterosexual 20 (71%) 26 (29%) 15 (26%) 2761 (40%) o0.0001
Homosexual/bisexual 8 (29%) 52 (57%) 42 (72%) 2954 (43%)
Other 0 (0%) 13 (14%) 1 (2%) 1112 (16%)
CD4 count (cells/mL) [median (range)] 154 (17–655) 24 (0–720) 35 (0–431) 385 (0–2274) o0.0001

Percentages may differ from 100% because of rounding. * There were 725 patients with unknown geographical provenance.

microsporidiosis, it appears that patients with isosporiasis 1.0

were more frequently female and heterosexual, more
frequently originated from sub-Saharan Africa (72%), and
had a significantly higher median CD4 count at diagnosis
of 154 cells/mL (Table 1).
Isosporiasis revealed HIV infection in eight patients
(29%). The other patients had been diagnosed with HIV
infection for a median of only 1 year (1 month to 13 years),
and eight patients had previously developed AIDS-defining
events [tuberculosis (n 5 3), non-Hodgkin lymphoma
(n 5 2), oesophageal candidiasis (n 5 2) and Kaposi sarco-
ma (n 5 1)]. Also, only eight patients (29%) were on
HAART at diagnosis, and five (17%) were on cotrimoxazole
prophylaxis at a daily dose of 400 mg. All but one patient
presented with chronic diarrhoea of a median duration of
1 month range (0–60 months), and median weight loss was
6 kg (0–16 kg). The severity of diarrhoea was demonstrated
by the high proportion of patients requiring hospital
admission (60%). Of note, only two patients (7%) had fever
at admission, and four (14%) had cholangitis.
Treatment and outcome
One patient had oocysts detected in stools but the diarrhoea
resolved before he received any treatment. The other 27
patients who had chronic diarrhoea received treatment
with either cotrimoxazole (n 5 26, at a median dose of
3200 mg daily) or a combination of albendazole and
ornidazole (n 5 1). The median duration of treatment was
15 days range (8–176 days). Treatment tolerance was good,
as only one patient discontinued cotrimoxazole after 17
days because of neutropenia. Following initial therapy,
22 patients (78%) received cotrimoxazole maintenance
therapy at a median daily dose of 800 mg to prevent a
0.8
0.6
0.4
0.2
0.0
0 10 20 30 40
Months
Fig. 2 Cumulative incidence of a first relapse of isosporiasis
estimated using the Kaplan–Meier method.
relapse of this infection. In a median follow-up time of 17.5
months, nine patients (32%) experienced 27 episodes of
clinical and parasitological relapse of isosporiasis (median
of three episodes per patient), with a cumulative 2-year
incidence of 45% (95% confidence interval 29.7–64.5)
(Fig. 2). At the time of the first relapse, eight of these nine
patients were on HAART (median duration 5 months), with
a median CD4 count of 132 cells/mL, but only one had a
viral load that was undetectable. The median time to
relapse was 10 months, and six patients relapsed despite
receiving maintenance cotrimoxazole prophylaxis, all of
them also being on HAART therapy at the time of relapse.
Three of these patients presented with a chronic clinical
course with persistent shedding of I. belli oocysts in stools
that was poorly responsive to treatment. Baseline char-
acteristics of patients with or without relapses were not
significantly different (data not shown).

r 2008 British HIV Association HIV Medicine (2008) 9, 126–130


Discussion
Following the introduction of HAART, the prevalence of a
number of opportunistic infections has dramatically
decreased, and especially those of protozoal intestinal
infections [15, 16]. In this study we show that, in contrast
to cryptosporidiosis and microsporidiosis, the prevalence of
isosporiasis increased in recent years among HIV-infected
patients followed at two hospitals in Paris, despite the
availability of HAART. This paradoxical finding is, how-
ever, probably explained by the change over time in the
demographic characteristics of patients followed at these
two institutions. Indeed, we observed in France an increase
in the proportion of HIV-infected patients who originated
from sub-Saharan Africa from 3% in the pre-HAART era to
15% in the 1997–2003 period [17]. Similarly, most patients
with isosporiasis in our study originated from sub-Saharan
Africa, and were more frequently female and heterosexual
(Table 1). These characteristics strongly differ from those of
patients with either cryptosporidiosis or microsporidiosis,
who were mainly white homosexuals. Finally, as I. belli is a
protozoon usually found in tropical areas, it is likely that
HIV-infected patients from sub-Saharan Africa were
infected in their native country before emigrating to
France, chronic isosporiasis being a reactivation of a
previous latent infection, similar to what is known for
cryptosporidiosis and microsporidiosis [18]. The recent
emigration of these patients to France is further supported
by the facts that 29% of patients did not know they were
infected with HIV, and that the others had been aware of
their HIV infection for a median of only 1 year. Also, only
29% of patients were receiving HAART at the time of
isosporiasis diagnosis in this study, despite a low median
CD4 count of 154 cells/mL, explaining why the effect of
HAART could not really be assessed. Indeed, Guiguet et al.
[17] recently reported a 79% reduction of the incidence of
isosporiasis in France in the HAART era (1997–2003)
among patients with clinical follow-up. In conclusion, the
increase in the number of cases of isosporiasis among HIV-
infected patients followed at our institutions in the HAART
era is likely to be a result of the recent immigration of
patients from sub-Saharan Africa who present late for care.
Interestingly, patients with isosporiasis have a signifi-
cantly higher CD4 cell count at diagnosis than patients
with other protozoal intestinal parasites. Therefore, para-
sitic stool examinations should be performed in HIV-
infected patients with diarrhoea even if the CD4 count is
above 100 cells/mL. Also, as oocyst shedding in stools
is intermittent, parasitic stool examination should be
repeated every few days in these patients.
The clinical characteristics of patients with isosporiasis
in this study were very similar to those of previously
r 2008 British HIV Association HIV Medicine (2008) 9, 126–130
Isosporiasis in the HAART era 129
published studies. All but one patient complained of
chronic watery diarrhoea, with a median duration of
1 month. Diarrhoea was severe enough to justify hospital
admission in 60% of patients. Fever was very uncommon
among our patients, in contrast to most published studies
[3, 19]. Bile duct involvement, and particularly cholangitis
(14%), seemed quite frequent in our study.
All patients with chronic diarrhoea received treatment
active against I. belli oocysts, mostly cotrimoxazole.
Treatment was, however, usually for longer (median
15 days) and at a higher dose (median daily dose of
3200 mg) than recommended [20]. Also, although 22 of our
28 patients received maintenance therapy for isosporiasis,
a 2-year incidence rate of relapse of 45% was observed in
this study. Furthermore, relapses were documented while
patients were still receiving cotrimoxazole prophylaxis,
and three patients had a chronic clinical course with
persistent oocyst shedding in stools despite continuous
cotrimoxazole therapy. Although the adherence to therapy
could be questioned in some patients, it was excellent in
others who were under direct observed therapy. As
exogenic reinfestation is probably very rare in France,
endogenic reinfestation could be suspected in some of
these patients, because of the persistence of intra-
enterocyte merozoites, and potential extra-intestinal loca-
tions of the parasite on which anti-infectious treatment
could be less effective [21, 22]. Another explanation could
be the selection of resistance to cotrimoxazole, but this
hypothesis could not be tested as I. belli cannot be
cultivated in vitro and has not yet been sequenced. The
role of HAART was difficult to assess in this study, but it is
interesting to note that eight of the nine patients who
experienced a relapse of isosporiasis were on HAART at
the time of the relapse, with a relatively preserved CD4
count of 132 cells/mL. Long-term follow-up will be
needed to determine whether immune restoration under
HAART could prevent the occurrence of further relapses in
these patients.
In conclusion, the recent increase in the prevalence of
isosporiasis among HIV-infected patients in our hos-
pitals in Paris is likely to be a result of the increased
proportion of patients recently immigrating from
sub-Saharan Africa and who present late to care.
Isosporiasis presents as a severe chronic diarrhoea, the
diagnosis of which relies on a simple parasitic stool
examination. Treatment with cotrimoxazole is usually
rapidly effective but maintenance therapy is needed to
prevent relapses. In a few patients, however, chronic
isosporiasis persists despite appropriate therapy, raising
the possible roles of immune depression, extra-intestinal
localizations of the parasite and/or selection of resistance
to antibiotics.
130 M Lagrange-Xélot et al.
References
1 Cotte L, Rabodonirina M, Piens MA. Prevalence of intestinal
protozoans in French patients infected with HIV. JAIDS 1993;
6: 1024–1029.
2 Berlin OG, Conteas CN, Sowerby TM. Detection of Isospora in
the stools of AIDS patients using a new rapid autofluorescence
technique. AIDS 1996; 10: 442–443.
3 DeHovitz JA, Pape JW, Boncy M. Clinical manifestations and
therapy of Isospora belli infection in patients with the acquired
immunodeficiency syndrome. N Engl J Med 1986; 315: 87–90.
4 Certad G, Arenas-Pinto A, Pocaterra L. Isosporiasis in
Venezuelan adults infected with human immunodeficiency
virus: clinical characterization. Am J Trop Med Hyg 2003; 69:
217–222.
5 Sauda FC, Zamarioli LA, Ebner Filho W. Prevalence of
Cryptosporidium sp. and Isospora belli among AIDS patients
attending Santos reference center for AIDS, Sao Paulo, Brazil.
J Parasitol 1993; 79: 454–456.
6 Dieng T, Ndir O, Diallo S. Prevalence of Cryptosporidium sp and
Isospora belli in patients with acquired immunodeficiency
syndrome (AIDS) in Dakar (Senegal). Dakar Med 1994; 39:
121–124.
7 Henry MC, De Clercq D, Lokombe B. Parasitological
observations of chronic diarrhoea in suspected AIDS adult
patients in Kinshasa (Zaire). Trans R Soc Trop Med Hyg 1986;
80: 309–310.
8 Conlon CP, Pinching AJ, Perera CU, Moody A. HIV-related
enteropathy in Zambia: a clinical, microbiological, and
histological study. Am J Trop Med Hyg 1990; 42: 83–88.
9 Kumar SS, Ananthan S, Saravanan P. Role of coccidian
parasites in causation of diarrhoea in HIV infected patients in
Chennai. Indian J Med Res 2002; 116: 85–89.
10 Henderson HE, Gillespie GW, Kaplan P. The Human Isospora.
Am J Hyg 1963; 78: 302–309.
11 Ng E, Markell EK, Fleming RL, Fried M. Demonstration of
Isospora belli by acid-fast stain in a patient with acquired
immune deficiency syndrome. J Clin Microbiol 1984; 20:
384–386.
12 Mannheimer SB, Soave R. Protozoal infections in patients
with AIDS. Cryptosporidiosis, isosporiasis, cyclosporiasis,
and microsporidiosis. Infect Dis Clin North Am 1994; 8:
483–498.
13 Grabar S, Le Moing V, Goujard C et al. Clinical outcome of
patients with HIV-1 infection according to immunologic and
virologic response after 6 months of highly active
antiretroviral therapy. Ann Intern Med 2000; 133: 401–410.
14 Hochberg Y. A sharper Bonferroni procedure for multiple tests
of significance. Biometrika 1988; 75: 800–803.
15 Miao YM, Awad-El-Kariem FM, Franzen C et al. Eradication of
cryptosporidia and microsporidia following successful
antiretroviral therapy. JAIDS 2000; 25: 124–129.
16 Le Moing V, Bissuel F, Costagliola D et al. Decreased prevalence
of intestinal cryptosporidiosis in HIV-infected patients
concomitant to the widespread use of protease inhibitors. AIDS
1998; 12: 1395–1397.
17 Guiguet M, Furco A, Tattevin P. HIV-associated Isospora belli
infection: incidence and risk factors in the French hospital
database on HIV. HIV Med 2007; 8: 124–130.
18 Sorvillo F. Isospora belli and the acquired immunodeficiency
syndrome. N Engl J Med 1990; 322: 131–132.
19 Pape JW, Verdier RI, Johnson WD Jr. Treatment and
prophylaxis of Isospora belli infection in patients with the
acquired immunodeficiency syndrome. N Engl J Med 1989;
320: 1044–1047.
20 Verdier RI, Fitzgerald DW, Johnson WD Jr et al. Trimethoprim-
sulfamethoxazole compared with ciprofloxacin for treatment
and prophylaxis of Isospora belli and Cyclospora cayetanensis
infection in HIV-infected patients. A randomized, controlled
trial. Ann Intern Med 2000; 132: 885–888.
21 Michiels JF, Hofman P, Bernard E. Intestinal and extraintestinal
Isospora belli infection in an AIDS patient. A second case
report. Pathol Res Pract 1994; 190: 1089–1093; discussion
1094.
22 Frenkel JK, Silva MB, Saldanha J. Isospora belli infection:
observation of unicellular cysts in mesenteric lymphoid tissues
of a Brazilian patient with AIDS and animal inoculation.
J Eukaryot Microbiol 2003; 50 (Suppl): 682–684.
r 2008 British HIV Association HIV Medicine (2008) 9, 126–130