- Email: [email protected]
Detection of Cryptosporidium and Giardia intestinalis in bedouin children from Southern Israel
hlernalional
Journal/or Parasilology, Vol. 24, No. 3, pp. 4O!Lll I, 1994 Copynght 0 1994 Australian Society for Parasitology Elsevier Science Ltd Printed in Great Britain. All rights reserved 0020-7519/94 57.00+ 0.00
Pergamon
0020-7519(93)EOOO2-R
RESEARCH
NOTE
DETECTION OF CRYPTOSPORIDIUM AND GIARDIA INTESTINALIS IN BEDOUIN CHILDREN FROM SOUTHERN ISRAEL J.
EL-ON,*
R.
DAGAN,~
D.
FRASERI
and
R.
J.
DECKELBAUM~
* Department of Microbiology and Immunology; tpediatrics Infectious Disease Unit; SEpidemiology and Health Services Evaluation Unit; Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel @epartment of Pediatrics, Columbia University, New York, U.S.A. (Received 15 November 1993; accepted 3 December 1993) Abstract-EL-ON J., DAGAN R., FRASER D. and DECKELBAUM R. J. 1994. Detection of Cryptosporidium and Giardia intestinalis in Bedouin children from southern Israel. International Journalfor Parasitology 24: 40941 I. During an 18 month period, a total of 4796 stool specimens collected from 151 Bedouin children enrolled in a cohort study and followed from birth, were screened for Giurdia intestinalis, Cryptosporidiurn spp. and other intestinal parasites. Specimens were collected in phenol-alcohol-formalin (PAF) preservative and examined prior to, and after, formalinether concentration (FEC). During 6 months of the second year Giardiu intestinalis was observed in 17.6% of the specimens and Cryptosporidium in 0.9% as compared with 1.8% (Giardia intestinalis) and 1.6% (Cryptosporidium) observed during the first year. Giurdiu intestinalis was detected in 8.4% (407/4796) of all the samples examined and Cryptosporidium in 1.3% (63/4796). Other intestinal protozoan parasites and helminthic ova demonstrated in the stool specimens included: Entumoeba coli (0.1%); Enfamoeba histolyticu (< 0.1 X); Hymenolepis nana (0.1%); and Trichuris trichiura (< 0.1%). Mixed infection with 2 parasites was observed in 0.3% of the specimens. PAF fixation was found to he highly effective in preserving the integrity and antigenicity of both Cryptosporidium-oocysts and Giardia intestinalis-cysts. The detection rate of Giardia intestinalis and Cryptosporidium before FEC was not significantly different from that obtained after FEC, showing differences of only 1% and 3% for Giurdia intestinalis and Cryptosporidium, respectively. INDEX KEY WORDS: Giardia intestinalis; Cryptosporidium; phenol-alcohol-formalin formalinether concentration (FEC).
Giardia intestinalis and Cryptosporidium spp. are considered common protozoan parasites causing gastrointestinal infections in humans and in animals world wide. Rates of infections with either parasite are highly variable within and between populations. The frequency of Giardia intestinalis is generally high in children (Peterson, 1972) and that of Cryptosporidium is particularly high in children from developing countries (Mata, Bolanos, Pizarro & Vives, 1984) and in immunocompromised patients (Navin & Juranek, 1984). During an 18 month period (December 1989 to May 1991) as part of an ongoing cohort study, a total of 4796 phenol-alcohol-formalin (PAF) fixed fecal specimens (Beck & Davies, 1981) were collected from
*To whom
all correspondence
should
(PAF)
fixation;
examined for Giardia and other intestinal parasites. The infants had been followed from birth and the intake into the cohort had been dispersed over the entire period. Most of the samples (84%) had been routinely collected as monthly samples, while the others were diarrhea1 episodes. The study programme included the examination of two to three specimens collected daily from each child. The stool specimen was filtered through one layer of dry gauze of 1.35 mm2 mesh size. The filtrate was centrifuged at 500 g for 4 min at room temperature. The supernatant was discarded and the pellet was resuspended in 10 ml of 0.85% saline and centrifuged again as previously described. The pellet was resuspended in a few drops of saline and examined microscopically for Cryptosporidium-oocysts after modified cold Kinyon’s acid-fast staining (Ma & 151 Bedouin
intestinalis.
be addressed. 409
infants
and
Cryptosporidium
410
J.
EL-ON et al.
TABLE I-FREQUENCY OF GI, CR SPECIES AND OTHERINTESTINAL PARASITES IN STOOLSAMPLESFROMCHILDREN,COLLECTED BETWEENDECEMBER1989 AND MAY 1991 IN A COHORTSTUDY
Organism
Giardia intestinalis Cryptosporidium spp. Entamoeba coli Eniamoeba histolytica ~ymeno~ep~s nanig Trickuris frj~h~~ra
Mixed infection Total
December 89-November 90 1st year (2765 specimens) NO. % of all % of all positive specimens infestations
December *May 91 2nd year (203 1 qkcimens) % of all No. % of all positive specimens infestations
1st + 2nd year (4796 specimens) No. % of all % of all positive specimens infestations
50
1.8
50.0
351
17.6
89.5
407
8.4
44
44.0
19
0.9
4.8
63
2
1.6 CO.1
2.0
4
0.2
1
CO.1
0
0.0
0
6
0.3
1 3
CO.1 0.1
0 12
0 0.6
1 1s
1.3 0.1 co.1 0.1 CO.1 0.3
100
3.6
399
19.7
499
10.4
:::: 100
Soave, 1983). Other intestinal parasites were detected using wet smear preparations which were examined at low ( x 100) and high ( x 400) magnification. Formalin-ether ~on~ntration (FEC) (Ritchie, 1948) was used to concentrate parasites. All ~rypt~~p~ri~i~~ positive specimens were also confirmed by immunofluorescent assay (IFA), using a commercial kit (Meriflour Cryptosporidium, Meridian Diagnostic, Cincinnati, OH) according to the instructions of the manufacturer. The total rates of positive stools obtained during the two study periods are summarized in Table 1. The rate of detection was significantly higher in the second period (19.7%) as compared with the first (3.6%). Giardia intestinalis was found in only 1.8% of 2765 specimens examined during the first period but in as many as 17.6% (357/2031) during the second period, while Cryptosporidium was detected in 1.6% and 0.9% of stool samples during the first and the second period, respectively. Cryptosporidium and Giardia intestinalis were not detected in the stools of children up to 5 and 6 months of age, respectively. Over the 18 month period there was a total of 499 (10.4%) positive specimens. Of those 0.3% were multiple infections. The highest frequency of parasites was observed with Giardia intestinalis and Cryptosporidium. Other detected parasites (Entamaeba coli, Entamoeba histolytica, Hymenolepis nana and Trichuris trichiura) were less common, showing a prevalence of ~0.1%. The PAF fixation was found to interfere with the acid-fast staining of Cryptosporidiutn. However, washing the stool specimens once with 0.85% saline removed most of the fixative and allowed successful staining of the oocysts. The PAF fixative also interferes with the brightness and the identification of Giardia intestinalis cysts examined at low magnification (X 100). Such an effect was not observed at a higher ma~ification (X 400). PAF was not found
6
:I 1.5 0 3.0 100
1 6
81.6
12.6 1.2 0.2
1.2 0.2
3.0 100
effective for long term preservation of helminthic ova, and most of them lost their integrity within several months. However, fixed Cryptosporid~um-oocysts and G~ardiaintestinalis-cysts remained intact over a period of more than 2 years and their antigenicity was preserved as demonstrated by both IFA assay (for Cryptosporidium) and ELISA (for Giardia intestinalis). A 100% detection rate was obtained at a concentration of 2 3 x lo4 Giardia intestinalis cysts and 2 1.4 x lo4 Cryptosporidium oocysts per g of stool, respectively. Diagnosis of Giardia ~ntestjna~isand Cry~tospor~dium mainly depends on the demonstration of the parasites either in wet smears (for Giardia intest~naiis) or in acid-fast stained preparations (for Cryptosporidium). Neither technique is highly sensitive and either may fail to detect the parasites in samples containing I 3 x lo4 (Giardia intestinalis) and I 1.4 x lo4 Cryptosporidium per gram of stool specimen, as demonstrated in this study. These results indicate that the rate of infestations with G~ardia intestinaiis and Cryptosporidium may have been higher than those observed. Indeed, the IFA assay available for Cryptosporidium (Sterling & Arrowood, 1986) and the ELISA developed for Giardia intestinalis (Stibbs, 1989) have been shown to be more sensitive and specific than microscopy. Recently, the modified formalin ethyl acetate coprodiagnostic concentration technique was found incapable of detecting Cryptosporidium in many symptomatic and asymptomatic individuals mainly due to oocyst loss during the gauze filtration procedure (Perry, Matthews & Miller, 1990; Weber, Bryan, Bishop, Wahlquist, Sullivan & Juranek, 1991). A better recovery of Cryptosporidium from stools was obtained using this technique followed by flotation over a hypertonic sodium chloride solution (Weber et al., 1991). In the present study one layer of gauze of
Research Note 1.35 mm* mesh size was used for stool filtration. The detection rate for almost all the parasites examined was not affected by the gauze filtration and most of them have been demonstrated almost similarly before and after the gauze filtration procedure. The loss of Cryptosporidium oocysts by the gauze filtration however, might be caused by oocysts being trapped in the mucus in the stool that may attach to the gauze during the filtration procedure (Perry et al., 1990). Nevertheless, none of the tested Cryptosporidium positive patients were misdiagnosed following filtration. With regard to Giardia intestinalis, while 5% of the patients were not diagnosed as having giardiasis following the filtration, 8.8% would have been missed. In the present study the rate of Giardiu intestinalis and Cryptosporidium detection in the PAF fixed specimens before FEC was not found to be significantly different from that obtained after FEC. These results were consistent during the 2 periods of the study, in both low prevalence and high prevalence of parasites. This may possibly be due to alterations in the specific gravity of the Giardiu intestinalis-cysts and the Cryptosporidium-oocysts. We therefore conclude that in a field study where large numbers of specimens are collected’ in preservatives the time consuming effort of FEC to improve Giardia intestinalis and Cryptosporidium detection rate is probably ineffective and unjustified.
Acknowledgements-This work was supported in part by ICIDR grant No. 1 POl-AI 26497 and the USA-Israel CDR grant No. DHR-5544G SS906000. We thank Mrs Ruth
411
Sneier and Mrs Frederique Abensoor for their technical assistance. REFERENCES BECKJ. W. & DAVIESJ. E. 1981. Medical Parasitology. 3rd edition, p. 318. The Mosley, CC.V., St Louis, MO. MA P. & SOAVER. 1983. Three-step stool examination for cryptosporidiosis in ten homosexual men with protracted watery diarrhea. Journal of Infectious Diseases 147: 824 828.
MATA L., BOLANOSH, PIAZARROD. & VIVES M. 1984. Cryptosporidiosis in hospital patients with gastroenteritis. American Journal of Tropical Medicine and Hygiene 33: 24-29. NAVIN T. R. & JURANEK D. D. 1984. Cryptosporidium
clinical and epidemiologic, and parasitological
review.
Reviews of Infectious Diseases 6: 313-327.
PERRYJ. L., MATTHEWSJ. S. T., & MILLER G. R. 1990. Parasite detection by efficiences of five stool concentration systems. Journal of Clinical Microbiology 28: 1094-1097. PETERSONH. 1972. Giardiasis (lambliasis). Scandinavian Journal of Gastroenterology 7 (suppl.) 14: l-44. RITCHIEL. S. 1948. An ether sedimentation technique for routine stool examination. Bulletin of U. S. Army Medical Department 8: 326.
STERLINGC. R. & ARROWOCJD M. J. 1986. Detection of Cryptosporidium spp. infections using direct immunofluorescent assay. Pediatric Infections Disease Journal 5: 139-142.
STIBBSH. H. 1989. Monoclonal antibody based enzyme immunoassay for Giardia lamblia antigen in human stool. Journal of Clinical Microbiology 27: 2882-2888.
WEBERR., BRYANR. T., BISHOPH. S., WAHLQUIST.S. P. SULLIVANJ. J. & JURANEKD. D. 1991. Threshold of detection of Cryptosporidium oocysts in human stool specimens: evidence for low sensitivity of current diagnostic methods. Journal of Clinical Microbiology 29: 1323-1327.
Journal/or Parasilology, Vol. 24, No. 3, pp. 4O!Lll I, 1994 Copynght 0 1994 Australian Society for Parasitology Elsevier Science Ltd Printed in Great Britain. All rights reserved 0020-7519/94 57.00+ 0.00
Pergamon
0020-7519(93)EOOO2-R
RESEARCH
NOTE
DETECTION OF CRYPTOSPORIDIUM AND GIARDIA INTESTINALIS IN BEDOUIN CHILDREN FROM SOUTHERN ISRAEL J.
EL-ON,*
R.
DAGAN,~
D.
FRASERI
and
R.
J.
DECKELBAUM~
* Department of Microbiology and Immunology; tpediatrics Infectious Disease Unit; SEpidemiology and Health Services Evaluation Unit; Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel @epartment of Pediatrics, Columbia University, New York, U.S.A. (Received 15 November 1993; accepted 3 December 1993) Abstract-EL-ON J., DAGAN R., FRASER D. and DECKELBAUM R. J. 1994. Detection of Cryptosporidium and Giardia intestinalis in Bedouin children from southern Israel. International Journalfor Parasitology 24: 40941 I. During an 18 month period, a total of 4796 stool specimens collected from 151 Bedouin children enrolled in a cohort study and followed from birth, were screened for Giurdia intestinalis, Cryptosporidiurn spp. and other intestinal parasites. Specimens were collected in phenol-alcohol-formalin (PAF) preservative and examined prior to, and after, formalinether concentration (FEC). During 6 months of the second year Giardiu intestinalis was observed in 17.6% of the specimens and Cryptosporidium in 0.9% as compared with 1.8% (Giardia intestinalis) and 1.6% (Cryptosporidium) observed during the first year. Giurdiu intestinalis was detected in 8.4% (407/4796) of all the samples examined and Cryptosporidium in 1.3% (63/4796). Other intestinal protozoan parasites and helminthic ova demonstrated in the stool specimens included: Entumoeba coli (0.1%); Enfamoeba histolyticu (< 0.1 X); Hymenolepis nana (0.1%); and Trichuris trichiura (< 0.1%). Mixed infection with 2 parasites was observed in 0.3% of the specimens. PAF fixation was found to he highly effective in preserving the integrity and antigenicity of both Cryptosporidium-oocysts and Giardia intestinalis-cysts. The detection rate of Giardia intestinalis and Cryptosporidium before FEC was not significantly different from that obtained after FEC, showing differences of only 1% and 3% for Giurdia intestinalis and Cryptosporidium, respectively. INDEX KEY WORDS: Giardia intestinalis; Cryptosporidium; phenol-alcohol-formalin formalinether concentration (FEC).
Giardia intestinalis and Cryptosporidium spp. are considered common protozoan parasites causing gastrointestinal infections in humans and in animals world wide. Rates of infections with either parasite are highly variable within and between populations. The frequency of Giardia intestinalis is generally high in children (Peterson, 1972) and that of Cryptosporidium is particularly high in children from developing countries (Mata, Bolanos, Pizarro & Vives, 1984) and in immunocompromised patients (Navin & Juranek, 1984). During an 18 month period (December 1989 to May 1991) as part of an ongoing cohort study, a total of 4796 phenol-alcohol-formalin (PAF) fixed fecal specimens (Beck & Davies, 1981) were collected from
*To whom
all correspondence
should
(PAF)
fixation;
examined for Giardia and other intestinal parasites. The infants had been followed from birth and the intake into the cohort had been dispersed over the entire period. Most of the samples (84%) had been routinely collected as monthly samples, while the others were diarrhea1 episodes. The study programme included the examination of two to three specimens collected daily from each child. The stool specimen was filtered through one layer of dry gauze of 1.35 mm2 mesh size. The filtrate was centrifuged at 500 g for 4 min at room temperature. The supernatant was discarded and the pellet was resuspended in 10 ml of 0.85% saline and centrifuged again as previously described. The pellet was resuspended in a few drops of saline and examined microscopically for Cryptosporidium-oocysts after modified cold Kinyon’s acid-fast staining (Ma & 151 Bedouin
intestinalis.
be addressed. 409
infants
and
Cryptosporidium
410
J.
EL-ON et al.
TABLE I-FREQUENCY OF GI, CR SPECIES AND OTHERINTESTINAL PARASITES IN STOOLSAMPLESFROMCHILDREN,COLLECTED BETWEENDECEMBER1989 AND MAY 1991 IN A COHORTSTUDY
Organism
Giardia intestinalis Cryptosporidium spp. Entamoeba coli Eniamoeba histolytica ~ymeno~ep~s nanig Trickuris frj~h~~ra
Mixed infection Total
December 89-November 90 1st year (2765 specimens) NO. % of all % of all positive specimens infestations
December *May 91 2nd year (203 1 qkcimens) % of all No. % of all positive specimens infestations
1st + 2nd year (4796 specimens) No. % of all % of all positive specimens infestations
50
1.8
50.0
351
17.6
89.5
407
8.4
44
44.0
19
0.9
4.8
63
2
1.6 CO.1
2.0
4
0.2
1
CO.1
0
0.0
0
6
0.3
1 3
CO.1 0.1
0 12
0 0.6
1 1s
1.3 0.1 co.1 0.1 CO.1 0.3
100
3.6
399
19.7
499
10.4
:::: 100
Soave, 1983). Other intestinal parasites were detected using wet smear preparations which were examined at low ( x 100) and high ( x 400) magnification. Formalin-ether ~on~ntration (FEC) (Ritchie, 1948) was used to concentrate parasites. All ~rypt~~p~ri~i~~ positive specimens were also confirmed by immunofluorescent assay (IFA), using a commercial kit (Meriflour Cryptosporidium, Meridian Diagnostic, Cincinnati, OH) according to the instructions of the manufacturer. The total rates of positive stools obtained during the two study periods are summarized in Table 1. The rate of detection was significantly higher in the second period (19.7%) as compared with the first (3.6%). Giardia intestinalis was found in only 1.8% of 2765 specimens examined during the first period but in as many as 17.6% (357/2031) during the second period, while Cryptosporidium was detected in 1.6% and 0.9% of stool samples during the first and the second period, respectively. Cryptosporidium and Giardia intestinalis were not detected in the stools of children up to 5 and 6 months of age, respectively. Over the 18 month period there was a total of 499 (10.4%) positive specimens. Of those 0.3% were multiple infections. The highest frequency of parasites was observed with Giardia intestinalis and Cryptosporidium. Other detected parasites (Entamaeba coli, Entamoeba histolytica, Hymenolepis nana and Trichuris trichiura) were less common, showing a prevalence of ~0.1%. The PAF fixation was found to interfere with the acid-fast staining of Cryptosporidiutn. However, washing the stool specimens once with 0.85% saline removed most of the fixative and allowed successful staining of the oocysts. The PAF fixative also interferes with the brightness and the identification of Giardia intestinalis cysts examined at low magnification (X 100). Such an effect was not observed at a higher ma~ification (X 400). PAF was not found
6
:I 1.5 0 3.0 100
1 6
81.6
12.6 1.2 0.2
1.2 0.2
3.0 100
effective for long term preservation of helminthic ova, and most of them lost their integrity within several months. However, fixed Cryptosporid~um-oocysts and G~ardiaintestinalis-cysts remained intact over a period of more than 2 years and their antigenicity was preserved as demonstrated by both IFA assay (for Cryptosporidium) and ELISA (for Giardia intestinalis). A 100% detection rate was obtained at a concentration of 2 3 x lo4 Giardia intestinalis cysts and 2 1.4 x lo4 Cryptosporidium oocysts per g of stool, respectively. Diagnosis of Giardia ~ntestjna~isand Cry~tospor~dium mainly depends on the demonstration of the parasites either in wet smears (for Giardia intest~naiis) or in acid-fast stained preparations (for Cryptosporidium). Neither technique is highly sensitive and either may fail to detect the parasites in samples containing I 3 x lo4 (Giardia intestinalis) and I 1.4 x lo4 Cryptosporidium per gram of stool specimen, as demonstrated in this study. These results indicate that the rate of infestations with G~ardia intestinaiis and Cryptosporidium may have been higher than those observed. Indeed, the IFA assay available for Cryptosporidium (Sterling & Arrowood, 1986) and the ELISA developed for Giardia intestinalis (Stibbs, 1989) have been shown to be more sensitive and specific than microscopy. Recently, the modified formalin ethyl acetate coprodiagnostic concentration technique was found incapable of detecting Cryptosporidium in many symptomatic and asymptomatic individuals mainly due to oocyst loss during the gauze filtration procedure (Perry, Matthews & Miller, 1990; Weber, Bryan, Bishop, Wahlquist, Sullivan & Juranek, 1991). A better recovery of Cryptosporidium from stools was obtained using this technique followed by flotation over a hypertonic sodium chloride solution (Weber et al., 1991). In the present study one layer of gauze of
Research Note 1.35 mm* mesh size was used for stool filtration. The detection rate for almost all the parasites examined was not affected by the gauze filtration and most of them have been demonstrated almost similarly before and after the gauze filtration procedure. The loss of Cryptosporidium oocysts by the gauze filtration however, might be caused by oocysts being trapped in the mucus in the stool that may attach to the gauze during the filtration procedure (Perry et al., 1990). Nevertheless, none of the tested Cryptosporidium positive patients were misdiagnosed following filtration. With regard to Giardia intestinalis, while 5% of the patients were not diagnosed as having giardiasis following the filtration, 8.8% would have been missed. In the present study the rate of Giardiu intestinalis and Cryptosporidium detection in the PAF fixed specimens before FEC was not found to be significantly different from that obtained after FEC. These results were consistent during the 2 periods of the study, in both low prevalence and high prevalence of parasites. This may possibly be due to alterations in the specific gravity of the Giardiu intestinalis-cysts and the Cryptosporidium-oocysts. We therefore conclude that in a field study where large numbers of specimens are collected’ in preservatives the time consuming effort of FEC to improve Giardia intestinalis and Cryptosporidium detection rate is probably ineffective and unjustified.
Acknowledgements-This work was supported in part by ICIDR grant No. 1 POl-AI 26497 and the USA-Israel CDR grant No. DHR-5544G SS906000. We thank Mrs Ruth
411
Sneier and Mrs Frederique Abensoor for their technical assistance. REFERENCES BECKJ. W. & DAVIESJ. E. 1981. Medical Parasitology. 3rd edition, p. 318. The Mosley, CC.V., St Louis, MO. MA P. & SOAVER. 1983. Three-step stool examination for cryptosporidiosis in ten homosexual men with protracted watery diarrhea. Journal of Infectious Diseases 147: 824 828.
MATA L., BOLANOSH, PIAZARROD. & VIVES M. 1984. Cryptosporidiosis in hospital patients with gastroenteritis. American Journal of Tropical Medicine and Hygiene 33: 24-29. NAVIN T. R. & JURANEK D. D. 1984. Cryptosporidium
clinical and epidemiologic, and parasitological
review.
Reviews of Infectious Diseases 6: 313-327.
PERRYJ. L., MATTHEWSJ. S. T., & MILLER G. R. 1990. Parasite detection by efficiences of five stool concentration systems. Journal of Clinical Microbiology 28: 1094-1097. PETERSONH. 1972. Giardiasis (lambliasis). Scandinavian Journal of Gastroenterology 7 (suppl.) 14: l-44. RITCHIEL. S. 1948. An ether sedimentation technique for routine stool examination. Bulletin of U. S. Army Medical Department 8: 326.
STERLINGC. R. & ARROWOCJD M. J. 1986. Detection of Cryptosporidium spp. infections using direct immunofluorescent assay. Pediatric Infections Disease Journal 5: 139-142.
STIBBSH. H. 1989. Monoclonal antibody based enzyme immunoassay for Giardia lamblia antigen in human stool. Journal of Clinical Microbiology 27: 2882-2888.
WEBERR., BRYANR. T., BISHOPH. S., WAHLQUIST.S. P. SULLIVANJ. J. & JURANEKD. D. 1991. Threshold of detection of Cryptosporidium oocysts in human stool specimens: evidence for low sensitivity of current diagnostic methods. Journal of Clinical Microbiology 29: 1323-1327.