734
Entamoeba histolytica and Entamoeba dispar Infection in Children in Bangladesh
Rashidul Haque, A. S. G. Faruque, Pauline Hahn,
David M. Lyerly, and William A. Petri, Jr.
International Centre for Diarrhoeal Disease Research, Dhaka,
Bangladesh; TechLab, Inc.. Blacksburg, and University of Virginia,
Charlottesville, Virginia
The prevalence of infection by the invasive parasite Entamoeba histolytica and the noninvasive
parasite Entamoeba dispar was determined in 2000 children in Bangladesh. Antigen detection
identified more cases of E. histolytica-E. dispar infection than did culture or microscopy.Microscopic
identification of E. histolytica-E, dispar complex infection in stool did not equate with the diagnosis
of amebic dysentery because most amebic infections in this population were due to E. dispar: Urban
children with diarrhea had a 4.2% prevalence of E. histolytica infection and a 6.5% prevalence of
E. dispar infection; rural asymptomatic children had a 1.0% prevalence of E. histolytica infection
and a 7.0% prevalence of E. dispar infection. Shigella dysenteriae and Shigella flexneri infections
were more frequent in children who also had Entamoeba infection, a potentially important consideration for the empiric treatment of dysentery in this population.
Materials and Methods
Study areas. Stool samples from children without diarrhea
were obtained from 987 children enrolled from a village in Mirzapur upazila (subdistrict) ~60 km north of Dhaka. Stool samples
from children were obtained from 1049 consecutive children en-
Received 5 August 1996; revised 24 September 1996.
This study was approved by the Research Review Committee of the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B) and
the Human Investigation Committee of the University of Virginia.
The University of Virginia has a license agreement with TechLab, Inc. for
antigen-detection tests for amebiasis.
Financial support: NIH (AI-36125); WHO; Center for Innovative Technology of the Commonwealth of Virginia; TechLab; ICDDR,B.
Reprints or correspondence: Dr. W. Petri, Room 2115 MR4 Bldg., University
of Virginia Health Sciences Center, Charlottesville, VA 22908.
The Journal ofinfectious Diseases 1997;175:734-6
© 1997 by The Universityof Chicago. All rights reserved.
0022-1899/97/7503-0039$01.00
tered into the surveillance program (4% systematic sampling) of
children with diarrhea treated at the International Centre for Diarrhoeal Disease Research. Samples were collected from January
1994 to September 1995 from both populations.
Microscopy, culture, and isoenzyme analysis of stool samples
for E. histolytica and E. dispar. Fresh stool samples were examined for the presence of blood visible to the naked eye, and a
smear of feces in 0.9% saline was examined microscopically for
the presence of red blood cells and E. histolytica- E. dispar complex cysts and trophozoites. In addition, stool samples from asymptomatic children were concentrated using a formyl-ether technique
for identification of cysts. Fresh stool. samples were cultured for
Entamoeba species in Robinson's medium within 6 h of collection,
and trophozoites were harvested when the number of organisms
was ~5 X 104 [10]. Identification of E. histolytica and E. dispar
was by isoenzyme electrophoresis of the cultured trophozoites,
according to the method described by Sargeaunt et al. [2].
Antigen detection. The Entamoeba test (designed to detect but
not differentiate the antigens of E. histolytica and E. dispar in
stool specimens) and the Entamoeba histolytica test (designed to
detect specifically E. histolytica in stool) were performed according to the manufacturer's instructions (TechLab).
Results
Children living in the rural area of Mirzapur had an overall
prevalence of asymptomatic colonization with E. histolyticaE. dispar complex of 8.0% as determined by antigen detection,
and 3.5% and 4.2% as determined by microscopy and culture,
respectively (table 1). There was no significant difference in
prevalence of infection in the different age groups, although
infection in girls was higher than infection in boys as determined by all three techniques (female-to-male ratios of E. histolytica-E. dispar complex infection: 1.3, 2.4, and 2.3 as determined by microscopy, culture, and antigen detection,
respectively).
Isoenzyme analysis of the amebae cultured from asymptomatic children was successfully accomplished in 29 cases, of
which 8 were identified as E. histolytica (zymodeme II only)
Downloaded from http://jid.oxfordjournals.org/ by guest on September 9, 2014
Entamoeba histolytica is now known to consist of 2 genetically distinct species, the invasive parasite E. histolytica, which
is the etiologic agent of amebic colitis and liver abscess, and
the noninvasive parasite Entamoeba dispar [1-4]. Microscopy
cannot differentiate E. histolytica from E. dispar [4, 5]. Culture
is more sensitive than microscopy, and isoenzyme analysis of
the cultured amebae enables differentiation of E. histolytica
from E. dispar; however, amebic cultures and isoenzyme analysis require a week to complete and show negative results for
many microscopy-positive samples [6].
A new approach to the detection of E. histolytica and E.
dispar is based on antigen detection in stool [6-9]. Antigenic
differences in the lectin of E. histolytica and E. dispar amebae
enable specific identification of the disease-causing amebae E.
histolytica [6, 8, 9]. In this study, we used the three wellcharacterized diagnostic tests for amebiasis-microscopy, culture, and antigen detection-to examine the prevalence of E.
histolytica and E. dispar infection in children in Bangladesh.
Concise Communications
JID 1997; 175 (March)
735
Table 1. Age-specific prevalence of E. histolytica-E. dispar complex infection in asymptomatic
children in a village in the Mirzapur upazila in Bangladesh and children with diarrhea at the International
Centre for Diarrhoeal Disease Research in Dhaka.
Infection detected by
Age (years)
113
190
684
987
569
206
274
1049
Culture"
E. histolyticaE. dispar antigen!
E. histolytica
antigen!
(2.7)
(4.7)
(3.4)
(3.5)
5 (4.4)
10 (5.3)
26 (3.8)
41 (4.2)
9 (8.0)
17 (8.9)
53 (7.7)
79 (8.0)
2 (1.8)
3 (1.6)
5 (0.7)
10 (1.0)
-{(0.5)
6 (2.9)
7 (2.6)
16 (1.5)
1 (0.2)
15 (7.3)
11 (4.0)
27 (2.6)
38 (6.7)
32 (15.5)
42 (15.3)
112 (10.7)
Microscopy*
3
9
23
35
8
14
22
44
(1.4)
(6.8)
(8.0)
(4.2)
NOTE. Data are no. (%).
* Fecal smears were examined microscopically for E. histolytica-E. dispar cysts and trophozoites.
t Fecal specimens were cultured in Robinson's medium.
+ E. histolytica-E. dispar antigen was detected using Entamoeba test (TechLab).
§ E. histolytica antigen was detected using E. histolytica test (TechLab).
and 21 identified as E. dispar (zymodeme I). When compared
with isoenzyme analysis, the antigen-detection test specific for
E. histolytica correctly identified 28 of the 29 cases and exhibited a sensitivity and specificity of 87.5% and 100%, respectively. Antigen detection identifiedE. histolytica in 10 of the 79
asymptomatic children with E. histolytica-E. dispar complex
infection, with a female-to-male ratio of infection of 0.8 and
a prevalence of 1%. The prevalence of E. dispar infection in
asymptomatic children was estimated to be 7% and 3% by
antigen detection and culture, respectively.
Children with diarrhea seen at the International Centre for
Diarrhoeal Disease Research hospital in Dhaka had a prevalence of E. histolytica-E. dispar complex infection of 10.7%
as determined by antigen detection and 1.5% and 2.6% as
determined by microscopy and culture, respectively (table 1).
E. histolytica-E. dispar complex infection was less common
in children < 3 years old. Prevalence of infection was similar
in both sexes as determined by all three techniques (femaleto-male ratios of E. histolytica-E. dispar complex infection of
0.7, 1.0, and 0.9 as determined by microscopy, culture, and
antigen detection, respectively).
Isoenzyme analysis of the amebae cultured from children
with diarrhea was successfully accomplished in 16 cases, of
which 13 were identified as E. histolytica (6 zymodeme II and
7 zymodeme XIV) and 3 were identified as E. dispar (zymodeme I). Compared with isoenzyme analysis, the antigen-detection test specific for E. histolytica correctly identified 13 of 16
cases and exhibited a sensitivity and specificity of 83.3% and
100%, respectively. The overall correlation between isoenzyme
analysis and antigen detection was 88.9%. Infection with the
pathogenic species E. histolytica was detected in 44 of the
1049 children with diarrhea by antigen detection, with a femaleto-male ratio of infection of 0.8 and a prevalence of 4.2%.
Overall, E. histolytica infection was found in 4.2% and 1.8%
of children with diarrhea by antigen detection and culture,
respectively; the prevalence of E.dispar infection was estimated to be 6.5% and 0.8% by antigen detection and culture,
respectively.
Microscopy was 37% sensitive and 99% specific, the E.
histolytica-E. dispar antigen-detection test was 94% sensitive
and 94% specific, and the E. histolytica antigen detection test
was 86% sensitive and 98% specific, compared with culture.
Children with diarrhea and E. histolytica or E. dispar infection had a significantly higher rate of Shigella dysenteriae and
Shigella fiexneri infection than did children without E. histolytica - E. dispar infection, as determined by antigen detection
(table 2). An association of Shigella infection with E. histolyt-
Table 2. Enteropathogens identified in children with diarrhea who
were negative or positive for antigen to E. histolytica-E. dispar or
E. histolytica alone.
E. histolytica/E. dispar
Enteropathogens
Negative
(n = 937)
Positive
(n = 112)
Vibrio cholerae
Shigella dysenteriae
Shigella flexneri
Salmonella species
235
29
46
20
35 (31)
19 (17)*
10 (9)*
2 (2)
(25)
(3)
(5)
(2)
E. histolytica
Negative
(n = 1005)
255
40
52
21
NOTE. Date are no. (%) of children.
* p < .01 compared with antigen-negative samples.
(25)
(4)
(5)
(2)
Positive
(n = 44)
15 (34)
8 (18)*
4 (9)*
1 (2)
Downloaded from http://jid.oxfordjournals.org/ by guest on September 9, 2014
Asymptomatic
children
1-2
3-5
6-14
All ages
Children with diarrhea
1-2
3-5
6-14
All ages
n
736
Concise Communications
ica-E. dispar infection was not observed when E. histolyticaE. dispar infection was determined by microscopy or culture
(only 1/16 E. histolytica-E; dispar microscopy-positive samples and 1/27 E. histolytica-E. dispar culture-positive samples
were also positive for Shigella species). Because of the high
prevalence of Shigella infections in stool samples that were
positive for E. histolytica-E; dispar antigen, we evaluated the
reactivity of pure cultures of S. dysenteriae, S. flexneri, and
Shigella sonnei in each of the antigen-detection tests. For this
analysis, brain-heart infusion cultures of these species, diluted
to contain 106 , 107 , or 108 cfu/mL, were prepared and tested
in each ELISA. None of the Shigella species at any of the
concentrations tested reacted in either of the antigen-detection
tests (data not shown).
Microscopy is inadequate for the diagnosis of amebic colitis.
Microscopic identification of E. histolytica - E. dispar complex
infection in stool did not equate with the diagnosis of amebic
dysentery; most parasites identified by microscopy in children
with dysentery were the noninvasive species E. dispar. In this
population ofchildren, the differentiation of amebic from bacillary dysentery requires species-specific identification of E. histolytica.
E. histolytica infection was present in 8% of grade school-
aged children in Dhaka with diarrhea. There was no evidence
of acquired immunity to colonization or infection with the E.
histolytica-E. dispar complex; the prevalence of infection was
not lower in older children in either the rural or urban populations. A similar age distribution of amebic colonization has
been observed in Africa [11, 12]. Colonization with E. dispar
was 3- 7 times more frequent than with E. histolytica in asymptomatic children from the rural area ofMirzapur. Other community-based studies have also demonstrated that asymptomatic
"cyst passers" are most likely to be infected with E. dispar
[11, 13, 14].
Both microscopy and culture underestimated the prevalence
of E. histolytica and E. dispar infection. Despite the lack of a
fourth well-characterized stool diagnostic test (such as polymerase chain reaction) to resolve discrepant results, our data
indicate that many of the additional cases of infection detected
by the antigen test were true-positive results. First, E. histolytica-E. dispar antigen detection and E. histolytica-specific antigen detection were highly specific compared with identification by culture plus isoenzyme analysis (the reference standard
for identification of E. histolytica in research laboratories). Second, the age- and sex-specific rates of infection determined by
the three tests were similar, which would not have been ex-
pected with a large number of false-positive results. Third,
neither of the antigen-detection tests cross-reacted with specimens culture-positive for other enteric pathogens such as Shigella, Salmonella, or Campylobacter species. Finally, neither
of the antigen tests exhibited false-positive problems with specimens from nonendemic areas.
We were surprised to find an association of E. histolyticaE. dispar and Shigella species infection. The high prevalence
of coinfection may simply reflect common routes of fecal-oral
exposure. The co-occurrence of these two agents of dysentery
may need to be taken into consideration in this population
when dysentery is empirically treated.
References
1. Petri WA Jr. Recent advances in amebiasis. Crit Rev Clin Lab Sci 1996;
33:1-37.
2. Sargeaunt PG, Williams JE, Grene JD. The differentiation of invasive
and non-invasive Entamoeba histolytica by isoenzyme electrophoresis.
Trans R Soc Trop Med Hyg 1978;72:519-21.
3. Tannich E, Horstmann RD, Knobloch J, Arnold HH. Genomic differences
between pathogenic and nonpathogenic Entamoeba histolytica. Proc
Natl Acad Sci USA 1989;86:5118-22.
4. Diamond LS, Clark CG. A redescription of Entamoeba histolytica Shaudinn 1903 (amended Walker 1911) separating it from Entamoeba dispar
(Brumpt 1925). J Eukaryot MicrobioI1993;40:340-4.
5. Gonzalez-Ruiz A, Haque R, Aguirre A, et al. Value of microscopy in the
diagnosis of dysentery associated with invasive Entamoeba histolytica.
J Clin Pathol 1994;47:236-9.
6. Haque R, Neville LM, Hahn P, Petri WA Jr. Rapid diagnosis of Entamoeba
infection using the Entamoeba and Entamoeba histolytica stool antigen
detection kits. J Clin MicrobioI1995;33:2558-61.
7. Gonzalez-Ruiz A, Haque R, Rehman T, et al. Diagnosis of amebic dysentery by detection of Entamoeba histolytica fecal antigen by an invasive
strain-specific, monoclonal antibody-based enzyme-linked imrnunosorbent assay. J Clin Microbiol 1994;32:964-70.
8. Haque R, Lyerly D, Wood S, Petri WA Jr. Detection of Entamoeba histolytica and Entamoeba dispar directly in stool. Am J Trop Med Hyg
1994; 50:595-6.
9. Haque R, Kress K, Wood S, et al. Diagnosis of pathogenic Entamoeba
histolytica infection using a stool ELISA based on monoclonal antibodies to the galactose-specific adhesin. J Infect Dis 1993; 167:247-9.
10. Robinson GL. The laboratory diagnosis of human parasitic amoebae. Trans
R Soc Trop Med Hyg 1968; 62:285-94.
11. Sargeaunt PG, Williams JE, Jackson TFHG, Simjee AE. A zymodeme
study of Entamoeba histolytica in a group of South African schoolchildren. Trans R Soc Trop Med Hyg 1982;76:401-2.
12. Bray RS, Harris WG. The epidemiology of infection with Entamoeba
histolytica in The Gambia, West Africa. Trans R Soc Trop Med Hyg
1977;71:401-7.
13. Gathiram V, Jackson TFHG. Frequency distribution of Entamoeba histolytica zymodemes in a rural South African population. Lancet 1985; 1:
719-21.
14. Jackson TFHG, Gathiram V, Simjee AE. Seroepidemiological study of
antibody responses to the zymodemes of Entamoeba histolytica. Lancet
1985; 1:716-9.
Downloaded from http://jid.oxfordjournals.org/ by guest on September 9, 2014
Discussion
JID 1997; 175 (March)