Subscribe Now Subscribe Today
Research Article

Cryptosporidium Antigen Detection in Aids Patients with Diarrhea

Chaudhri S. Hayat, Cynthia L. Chappell , A. Clinton White , Kay Kimball , Richard W. Goodgame and Robert M. Genta
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail

The microscopic detection of cryptosporidiosis is difficult, time-consuming and relatively insensitive, especially in patients with Iight infections. New techniques based on specific antibodies to Cryptosporidium antigens have become available. One method, an indirect immunofluorescence assay, has increased sensitivity over conventional microscopy. A more recent technique, an antigen detection ELISA, has not been adequately evaluated in a clinical setting. The purpose of this study was to compare the antigen detection ELISA to the quantitative IFA (QIFA) using stool samples from HIV-infected patients with diarrhea. The antigen detection test had a high degree of repeatability with a coefficient of variation of 0.08. This method compared very favorably with the QIFA and yielded a sensitivity of 96.4% and a specificity of 100%. The two tests gave comparable results when high numbers of oocysts were present. In contrast, the antigen detection assay was often positive for QIFA-negative samples from patients intermittently shedding oocysts. Hence, in the lighter infections, the antigen detection ELISA appeared to be more sensitive. The ELISA absorbance value was proportional to the number of oocysts present in the test as defined by QIFA. Antigen detection ELISA appears to be an effective alternative to immunofluorescence for detecting fecal oocyst excretion and may be particularly useful when large numbers of stools must be tested.

Related Articles in ASCI
Similar Articles in this Journal
Search in Google Scholar
View Citation
Report Citation

  How to cite this article:

Chaudhri S. Hayat, Cynthia L. Chappell , A. Clinton White , Kay Kimball , Richard W. Goodgame and Robert M. Genta, 2000. Cryptosporidium Antigen Detection in Aids Patients with Diarrhea. Pakistan Journal of Biological Sciences, 3: 1100-1103.

DOI: 10.3923/pjbs.2000.1100.1103



Cryptosporidium parvum is an intracellular protozoan parasite, which infects the intestinal epithelium of animals. and man (Tzipori, 1983; Kennedy et al., 1977). The parasite life cycle is complex and involves both sexual and sexual stages (Current, 1989). Transmission of the parasite occurs when C. parvum oocysts from an infected animal (Current et al., 1983) or person (Janoff et al., 1990; Fafard and Lalonde, 1990) are ingested by a naive host. In same cases, community outbreaks of the infection have been linked to municipal drinking water (D’Antonio et al., 1985; Hayes et al., 1989) as in the U.S. outbreak in Milwaukee. Oocysts, which are resistant to disinfection and chlorination (Angus et al., 1882; Campbell et al., 1982), are incompletely removed by water treatment and filtration (Madore et al., 1987). Also, treated water can become contaminated with untreated water during periodic of heavy, rainfall and flooding. Another mode of transmission, person-to-person contact, is suspected of causing cryptosporidiosis in day care centers, where it is especially prevalent, as well as in other institutions housing children (Janoff et al., 1990; Fafard and Lalonde, 1990). Cryptosporidium is also a recognized cause of traveler’s diarrhea (Soave and Ma, 1985).

In immunocompetent individuals, the parasite can cause a self-limited, asymptomatic infection or symptoms which range from a mild to severe. A 7-day incubation period is typically followed by an illness of 2-14 days. The most commonly reported symptoms include diarrhea, nausea and vomiting, abdominal pain and low-grade fever (Jokipii and Jokipii, 1986; Elsser et al., 1986; Fayer and Ungar, 1986). However, Infectious oocysts may be excreted for an additional 1-2 weeks after symptoms resolve (Jokipii and Jokipii, 1986). In contrast, imrnunosuppressed patients with CD4 counts <200/mm3 may have a more severe and chronic disease, which in some cases may be life-threatening (Wolfson et al., 1985; Current et al., 1983).

In many laboratories the diagnosis of Cryptosporidium infection currently relies on the microscopic examination of stool samples. Various stool concentration methods and staining techniques, such as the DMSO-modified acid fast stain, are typically employed (Bronsdon, 1984; Anderson, 1981). However, with acid fast stains Cryptosporidium can be confused with other acid fast organisms or with comparably sized yeasts. Evidence also suggests that staining methods are relatively Insensitive and that detection requires a large number of organism to be present, especially in formed stools (Weber et al., 1991).

An indirect immunofluorescence assay (IFA, Meridian Diagnostics. Cincinnati, OH) based on a oocyst, wall-specific monoclonal antibody (Sterling and Arrowood, 1986) has provided enhanced sensitivity and specificity over other techniques (Garcia et al., 1987; Rusnak et al., 1989). Moreover, a direct immunoassay, which requires less incubation time, has refaced the earlier indirect method. We have adapted this method to a quantitative immunofluorescence assay (QIFA) technique for detecting Cryptosporidium oocysts. Our data indicate that in HIV-infected patients oocyst excretion is relatively constant from stool to stool and day to day in the same person. However, intensity of infection varies widely from patient to patient ranging from a mean daily oocyst concentration of 2×103/ml to 2.3×106/ml (Goodgame et al., 1993). Patients with very light Infections, i.e. excreting fewer than 104 oocysts/ml, may pose a diagnostic problem unless the more sensitive techniques are employed.

Improvement in sensitivity may come with the development of new antigen capture techniques, which could be designed to detect soluble and particulate fractions, as well as intact oocysts detected by previous methods. An ELISA-based assay would also provide a less time-consuming and more objective method of Cryptosporidium detection. One recent study employed a monoclonal antibody to a 40 kDa cyst wag antigen to detect Cryptosporidium in bovine feces (Anusz et al., 1990). The sensitivity of the antigen capture assay exceeded that of conventional stains but was less sensitive than IFA.

The purpose of this study was to compare a commercially-available antigen detection EUSA assay to the quantitative IFA using specimens from HIV-infected individuals. Secondly, in samples positive by both tests, EUSA absorbance values were compared to the number of oocysts present in the micro titer well. And lastly, stools tram Individuals intermittently shedding oocysts were employed to evaluate the sensitivity of the EUSA assay for low level infections.

Materials and Methods

Four hundred eighty two stool samples were collected from 34 HIV-infected individuals who presented at a local clinic with diarrhea. Steals were weighed and an aliquot was preserved in 10% buffered formalin at a 1:4 dilution. Preserved specimens were stored et 4°C before QIFA testing. Samples were then stored frozen at -90°C until further use.

Samples were divided lieu 3 groups: 1) 307 stool sample Cryptosporidium-positive by QIFA, 2) 15 stool samples from a group of patients with microsporidia infection confirmed by electron microscopy and negative for Cryptosporidium by modified acid fast staining, QIFA and colon and small bowel biopsy and 3) 160 stool samples negative by QIFA but from symptomatic patients, some of whom were previously positive (by QIFA) for Cryptosporidium.

A commercial kit was used for quantitative immunofluore-cence assay as previously described (MeriFluor Cryptosporidium Kit, Meridian Diagnostic, Inc., Cincinnati, OH) (Goodgame et aI., 1993). In this assay a monoclonal antibody to a Cryptosporidium oocyst wall antigen was used to visualize the oocyst via an fluorescein isothiocyanate-labeled conjugated second antibody. Apple green fluorescence and characteristic morphology indicated the presence of an intact oocyst. A 5 μL aliquot of preserved stool was examined; each stool was assayed in triplicate. Oocyst concentration was calculated by counting oocysts in a defined area and correcting for the proportion of the slide examined. Frozen samples were thawed and thoroughly mixed before testing. The Color Vue-Cryptosporidium Kit (Seradyn, Inc., Indianapolis, IN) was used to detect Cryptosporidium antigen in stool samples. All reagents used in the following procedure were provided in the kit. This assay employed a double-antibody "sandwich" ELISA performed In a microtiter format. One hundred μL of negative or positive control samples or 50 μL. of preserved fecal samples were added to microtiter wells in a plate coated with rabbit anti-Cryptosporidium antiserum. The plate was incubated at 37°C for 45 minutes and subsequently washed three times with a phosphate buffered wash solution (supplied with the kit) using a Titertek Microplate Washer S8/12 (Flow Laboratories Inc., McLean. VA). A murine monoclonal antibody to oocyst wall antigen (100 pL) was then added to each well and incubated at 37°C for 20 minutes. Plates were washed free of unbound antibody and 100 μL of biotin-conjugated anti-mouse IgG was added to each well and incubated 10 minutes at 37°C. Following a thorough wash, 100 μL of streptavidin-corijugated horseradish peroxidase was added to each well and incubated 10 minutes at 37°C. Wells were again washed prior to the addition of 50 μL of H202-activated tetramethyibenzidine solution. The plate was incubated at room temperature for 10 minutes. The reaction was terminated by the addition of 50 μL of phosphoric acid solution and the plates were read at 450 mm by a Microplate Autoreader (Model, Elio Tek Instrument, Inc., Winooski, VT). Positive results were defined as any absorbance value that exceeded 2 SD above the mean negative control.


The antigen detection ELISA assay was standardized using positive and negative control samples. To assess assay variability, 6 replicates of positive and negative control samples were tested in a single plate (Table 1). For plate-to-plate variation, duplicate positive and negative controls (contained in test kit) were included in assays done on 15 different days over a period of 6 months. Well-to-well coefficient of variation was remarkably low, 2.9% and 8.0%, for the positive and negative samples, respectively. A greater degree of variation, 0.354 and 0.264, was seen in the positive and negative control samples tested on different days.

Table 1:
Variablity in the Cryptosporidium antigen detection assay. The mean absorbance values (A450), the standard deviation ISO), coefficient of variation (CV). and the number of trials are given for positive and negative control reagents
*Plates were run on different days over a period of 3 months

Fig. 1:
Comparison of the antigen detection (ELISA) assay to the quantitative immunofluorescence assay (IFA). Three hundred seven stools confirmed positive and 15 stools confirmed negative for Cryptospondium were assessed by both methods. Sensitivity and specificity of the antigen detection assay are indicated

Absorbance values of the positive or negative controls differed as much as 2-fold between some days. This may have been the result of aging of reagents: however, no consistent pattern was seen over the 3 month period of study. It should be noted that even in 2 plates where absorbance values of the positive controls were lower than usual, there was still a 5-fold difference between positive and negative controls, These data suggest that controls should be included at the time of each assay. From among the 482 specimens, 108 of the stool samples were tested in the antigen detection assay 2-3 times. Repeat assays on 60 (65%) of the samples yielded the same result; whereas, discrepancies, i.e. positive versus negative results, were seen between repeat tests in 48 (35%) of the samples. All of the discrepancies, however, were found in samples with low oocyst counts (<104/50 μL). These data suggest that in those specimens with low oocyst concentrations, sampling of the fecal specimen did not always retrieve a detectable number of oocysts. This non-uniformity of results in low level infections was also noted with the QIFA. The sensitivity and specificity of the Cryptosporklium antigen assay was assessed by testing samples previously confirmed to be positive (14 = 307) or negative (N=15) (Fig. 1). Negative samples were from HIV-infected patients with confirmed microsporidiosis. The antigen assay performed well with a sensitivity of 96.4% and a specificity of 100%. There were 11 QIFA-positive samples that the antigen assay did not detect. Seven of these 11 had oocyst counts with a mean and median level of 1.2 X 104 and 5.3 X 103 per ml. respectively. The remaining 4 had higher oocyst counts (mean = 5.8 X 105 per ml). Since the 322 samples in this analysis were collected from 30 patients, the data were also evaluated on a per patient basis.

Fig. 2:
Companion of antigen assay absorbance values to the number of oocysts present as defined by quantitative immunofluorescence. Samples were grouped according to infection intensity: <1000 oocyste/50 μL (light infection), 1,000-10,000 oocysts/50 μL (moderate infection) and 10,000-100 oocysts/50 μL (heavy infection). Each box represents the 25-75% quartiles; the median value for each group is shown as the line bisecting the box. The 99% confidence interval is indicated by the vertical lines, rid outliers are indicated by circles

Results were similar to the "per specimen" analysis. The relationship of the EUSA abs barb value to the oocyst concentration was examined (Williamson et al., 1989). The number of oocysts in the Miatgitif well was estimated from the QIFA results. For the 294 samples positive by both tests, the number of oocysts per 50 μL was compared to the absorbance value generated in the antigen detection assay. ELISA absorbance values were plotted according to the intensity of infection, which was divided into three groups: <2 X 104 oocystsiml. (light infection), 2 X 104 -1 X 105 oocysts/mL (moderate infection) and 2 X 105-1 X 106 oocysts/mL (heavy infection). The number of samples in each group was 43, 158 and 97, respectively. A boxplot (Fig. 2) displays the 25th and 75th quartile (rectangle), the median (line bisecting rectangle and the approximate 99% lie (vertical line) for each group. Outliers are indicated by circles. A positive relationship can be seen in the intensity of infection (number of oocysts present) and the absorbance value worsted in the EUSA assay. A separate group (N = 180) of QIFA-negative samples was tested for the presence of Cryptosporidium antigeen. One hundred thirty four of these samples were collected from patients that were intermittently positive by QIFA. Twenty six samples were from symptomatic Individuals that were QIFA-negative on multiple stool specimens. One hundred five (65.6%) of the temples were positive by the andiron detection assay. These data suggest that the antigen detection assay may be a more sensitive Indicator of infection than the QIFA test.


Cryptosporidiosis is now recognised as an important cause of daintiest illness, especially in the iminunocompromised, in day care centers and in traveler's diarrhea. Outbreaks due to drinking water contaminstion have also been documented. in immunocompromleed individuals Cryptosporidium infection can be especially serious and even life-threatening. In the peat. diagnosis of the disease has been difficult and has required special staining procedures. Documentation of infection in persons with low level infections has been especially problematic. The QIFA detection system is more sensitive than conventional staining procedures, including modified acid fast stains. This approach, however, is time consuming and requires fluorescence microscopy. An EUSA-based assay, such as the antigen detection test reported hers, has the advantage of detecting parasite products in addition to whole oocysts and the enzyme-conjugated reporter antibody amplifies the signs! several fold. Also, many specimens can be tested on the same plats.

An earlier study (Anusz et al., 1990) found that the ELISA assay was less sensitive than the IFA. In that study bovine feces were seeded with known numbers of oocysts and directly tested in the ELISA. In our experience we have found that if the sample is frozen and thawed before the ELISA assay, the absorbance value is increased 2.3-fold (data not shown). This result is presumably due to the disruption of oocysts end disbursement of antigenic material. Thus, a freeze/thaw step should increase EUSA sensitivity.

To compare the antigen detection assay to the OIFA, well-defined populations were chosen. Patients were first screened for Cryptosporidium infection by standard staining techniques at the hospital laboratory (Ben Taub General Hospital. Houston. TX). Patients who tested positive were considered as having a confirmed diagnosis. A small group of patients who went negative for Cryptosporidium but symptomatic. were studied extensively and confirmed to have microsporidia infection only. These patients became the negative control group. The antigen detection assay in the above groups compered favorably to the Q1FA with a sensitivity of 95.4% and a specificity of 100% (Fig. 1).

The antigen detection assay was evaluated for test repeatability by using the negative and positive control reagents that come with the kit. A minor degree of well-towell variation (CV = 0.08 or less) was seen within each of the 6 microtiter plates tested (Table 1). A higher variation in absorbance values was seen from day-to-day. That is, negative and positive controls run an one plate may very as much as 35.4% from values generated in a separate plats on a different day. These data indicate the importance of including positive and negative control wells with each test. Repeat assays on 108 stool samples revealed discrepancies in 35% of these specimens. However, all of the discrepant samples wire from lightly infected patients as defined by QlFA and may represent sampling variations. it is likely that most, if not all, of these samples would have been negative by acid fast staining. The QlFA results were based on a mean of 3 separate measurements. In low level infections, the majority (approximately 75%) of stools assessed by QIFA also showed discrepencies wrong the aliquots tested. In some cases only 1 of the 3 aliquots contained a single oocyst. If any oocysts wets seen in any of the WAN measurements, the sample was considered eve. Antigen essays were also considered positive if any of the implicates showed positive absorbance values. Thus, both tests have problems detecting oocysts or antigens when working near the lower limit of detection sensitivity. This likely reflects the non-uniformity of oocyst numbers in store aliquots. Hence, duplicate or triplicate ammo should be performed on symptomatic patients that are initially negative in either the QIFA or antigen detection assays. It should be noted, however, that the cattiest significance of light infections is not clear.

The association between absorbance value (antigen assay) and the estimated number of oocysts (QIFA) suggests that the absorbance value of a sample may be used to estimate the intensity of the infection, i.e. high, moderate or low level. It was not, however, indicative of the exact number of oocysts present within each range.

It should be noted that 1 patient with Cryptosporidium infection had tools uniformly negative in the antigen detection assay despite having a high number of oocysts in the QIFA. The earns monoclonal antibody employed in the antigen detection test was positive when tested by IFA. There may be some unknown factor in the stool that intarferred with the antigen assay. Coating antibody could have been adversely affected by pH but this could not be evaluated since the stools were diluted in buffered preservative solution. To our knowledge this patient is not receiving any unusual treatment or dietary supplement that would account for the results. Subsequent to the collection of these stool., the patient developed cytomegalovirus colitis with associated gastrointestinal bleeding. It is possible that excessive blood in the stool may have interfered with the assay.

In summery, the antigen detection assay appeared to be more sensitive and was easier to use than the QIFA procedure. Although both methods have similar incubation times, the antigen detection assay in the microliter format allows many samples to be screened at one time and provides a numerical (objective) assessment of the results.


This work was supported, in part, by Seradyn, Inc. who provided the antigen detection assay kits and support necessary for the successful completion of the study. Support was also obtained from the King Ranch Family Foundation (CLC), Public Health Service FIRST Award Al24490 (CLC) from the National Institute of Allergy and Infectious Diseases and the Agency for International Development, Post-doctoral Training Program (CSH). Stool samples from patients with microsporidia infections were a gift from Dr. Linda Rabenek, Baylor College of Medicine. We also wish to thank Irene Bleyzer for her excellent technical assistance.

1:  Tzipori, S., 1983. Cryptosporidiosis in animals and humans. Microbiol. Rev., 47: 84-96.
Direct Link  |  

2:  Kennedy, G.A., G.L. Kreitner and A.C. Strafuss, 1977. Cryptosporidiosis in three pigs. J. Am. Vet. Med. Assoc., 170: 348-350.
PubMed  |  Direct Link  |  

3:  Current, W.L., N.C. Reese, J.V. Ernst, W.S. Bailey, M.B. Heyman and W.M. Weinstein, 1983. Human cryptosporidiosis in immunocompetent and immunodeficient persons-studies of an outbreak and experimental transmission. N. Engl. J. Med., 308: 1252-1257.
CrossRef  |  PubMed  |  Direct Link  |  

4:  D'Antonio, R.G., R.E. Winn, J.P. Taylor, T.L. Gustafson and W.L. Current et al., 1985. A waterborne outbreak of cryptosporidiosis in normal hosts. Ann. Int. Med., 103: 886-888.
PubMed  |  Direct Link  |  

5:  Hayes, E.B., T.D. Matte, T.R. O'Brien, T.W. McKinley and G.S. Logsdon et al., 1989. Large community outbreak of cryptosporidiosis due to contamination of a filtered public water supply. New Engl. J. Med., 320: 1372-1376.
CrossRef  |  Direct Link  |  

6:  Angus, K.W., D. Sherwood, G. Hutchison and I. Campbell, 1982. Evaluation of the effect of two aldehyde-based disinfectants on the infectivity of faecal cryptosporidia for mice. Res. Vet. Sci., 33: 379-391.

7:  Campbell, I., A.S. Tzipori, G. Hutchison and K.W. Angus, 1982. Effect of disinfectants on survival of Cryptosporidium oocysts. Vet. Rec., 111: 414-415.
PubMed  |  Direct Link  |  

8:  Madore, M.S., J.B. Rose, C.P. Gerba, M.J. Arrowood and C.R. Sterling, 1987. Occurrence of Cryptosporidium oocysts in sewage effluents and selected surface waters. J. Parasitol., 73: 702-705.
PubMed  |  Direct Link  |  

9:  Soave, R. and P. Ma, 1985. Cryptosporidiosis: Traveler's diarrhea in two families. Arch. Intern. Med., 145: 70-72.
CrossRef  |  Direct Link  |  

10:  Jokipii, L. and A.M. Jokipii, 1986. Timing of symptoms and oocyst excretion in human cryptosporidiosis. New Engl. J. Med., 315: 1643-1647.
CrossRef  |  Direct Link  |  

11:  Elsser, K.A., M. Moricz and E.M. Proctor, 1986. Cryptosporidium infections: A laboratory survey. Can. Med. Assoc. J., 135: 211-213.
Direct Link  |  

12:  Fayer, R. and B.L. Ungar, 1986. Cryptosporidium spp. and cryptosporidiosis. Microbiol. Rev., 50: 458-483.
Direct Link  |  

13:  Wolfson, J.S., J.M. Richter, M.A. Waldron, D.J. Weber, D.M. McCarthy and C.C. Hopkins, 1985. Cryptosporidiosis in immunocompetent patients. New Engl. J. Med., 312: 1278-1282.
CrossRef  |  PubMed  |  Direct Link  |  

14:  Goodgame, R.W., R.M. Genta, A.C. White and C.L. Chappell, 1993. Intensity of infection in AIDS-associated cryptosporidiosis. J. Infect. Dis., 167: 704-709.
CrossRef  |  Direct Link  |  

15:  Bronsdon, M.A., 1984. Rapid dimethyl sulfoxide-modified acid-fast stain of Cryptosporidium oocysts in stool specimens. J. Clin. Microbiol., 19: 952-953.
Direct Link  |  

16:  Anderson, B.C., 1981. Patterns of shedding of cryptosporidial oocysts in Idaho calves. J. Am. Vet. Med. Assoc., 178: 982-984.
PubMed  |  Direct Link  |  

17:  Weber, R., R.T. Bryan, H.S. Bishop, S.P. Wahlquist, J.J. Sullivan and D.D. Juranek, 1991. Threshold of detection of Cryptosporidium oocysts in human stool specimens: Evidence for low sensitivity of current diagnostic methods. J. Clin. Microbiol., 29: 1323-1327.
Direct Link  |  

18:  Sterling, C.R. and M.J. Arrowood, 1986. Detection of Cryptosporidium sp. infctions using a direct imunofluorescent assay. Pediatr. Infect. Dis. J., 5: 139-142.
Direct Link  |  

19:  Garcia, L.S., T.C. Brewer and D.A. Bruckner, 1987. Fluorescence detection of Cryptosporidium oocysts in human fecal specimens by using monoclonal antibodies. J. Clin. Microbiol., 25: 119-121.
Direct Link  |  

20:  Rusnak, J., T.L. Hadfield, M.M. Rhodes and J.K. Gaines, 1989. Detection of Cryptosporidium oocysts in human fecal specimens by an indirect immunofluorescence assay with monoclonal antibodies. J. Clin. Microbiol., 27: 1135-1136.
Direct Link  |  

21:  Anusz, K.Z., P.H. Manson, M.W. Riggs and L.E. Perryman, 1990. Detection of Cryptosporidium parvum oocysts in bovine feces by monoclonal antibody capture enzyme-linked immunosorbent assay. J. Clin. Microbiol., 28: 2770-2774.
Direct Link  |  

22:  Williamson, D.F., R.A. Parkerm and J.S. Kendrick, 1989. The box plot: A simple visual method to interpret data. Ann. Intern. Med., 110: 916-921.
CrossRef  |  PubMed  |  

23:  Current, W.L., 1989. Cryptosporidium spp. In: Parasitic Infections in the Compromised Host, Walzer, P.D. and R.M. Genta (Eds.)., Marcel Dekker Inc., New York, pp: 281-341.

24:  Janoff, E.N., P.S. Mead, J.R. Mead, P. Echeverria and L. Bodhidatta et al., 1990. Endemic Cryptosporidium and Giardia lamblia infections in a Thai orphanage. Am. J. Trop. Med. Hyg., 43 : 248-256.

25:  Fafard, J. and R. Lalonde, 1990. Long-standing symptomatic cryptosporidiosis in a normal man: Clinical response to spiramycin. J. Clin. Gastroenterol., 12: 190-191.
PubMed  |  Direct Link  |  

©  2021 Science Alert. All Rights Reserved