Evaluation of the Triage Micro Parasite Panel for Detection of Giardia lamblia, Entamoeba histolytica/Entamoeba dispar, and Cryptosporidium parvum in Patient Stool Specimens

Abstract

A study comparing the Triage Micro Parasite Panel (Biosite Diagnostics, Inc., San Diego, Calif.) to conventional O&P examination (O&P) was performed using patient fecal specimens. Five hundred twenty-three stool samples were compared. Nineteen specimens were found to be positive by Triage, and 29 were found to be positive by O&P. Seven specimens were positive for Giardia lamblia, four were positive for Entamoeba histolytica/E. dispar, and three were positive for Cryptosporidium parvum as determined by both methods. There was one false positive by Triage (C. parvum) and four false negatives by O&P (two G. lamblia, one E. histolytica/E. dispar, and one C. parvum). The Triage test accurately detected all 18 specimens that contained one of the three organisms that it was designed to detect. The Triage test is a rapid, easy-to-use enzyme immunoassay for the detection of G. lamblia, E. histolytica/E. dispar, and C. parvum in fresh or fresh-frozen fecal specimens. These data suggest that the Triage test can be used as a screen for the immediate testing of stool specimens for these three pathogenic parasites. If Triage test results are negative, O&P can be performed if parasitic infections other than G. lamblia, E. histolytica/E. dispar, or C. parvum are suspected.

Human parasites account for millions of infections every year throughout the world. The Centers for Disease Control reported that the five most common parasitic infections in the United States were caused by Giardia lamblia, Trichuris trichiura, Ascaris lumbricoides, Enterobius vermicularis, and Entamoeba histolytica (2). Cryptosporidium parvum, a rarely recognized parasite in humans in the pre-AIDS area, accounted for 13.8% of all pathogenic protozoans in 1984 in one New York City hospital (11). Routine parasitological diagnostic procedures are manual in nature and take a considerable amount of experienced technologist time to perform and interpret. With many laboratories decreasing staff as a cost-containment measure and cross-training individuals between laboratory disciplines, the necessary expertise in parasitology may be waning. Thus, in efforts to improve diagnostic testing, new methods for testing for the presence of the most common pathogenic parasites are being developed. Several products have been evaluated, including direct immunofluorescence assays and other immunoassays for antigen detection of G. lamblia, E. histolytica, and C. parvum directly from stool samples. These assays range in sensitivity from 66.3 to 100% and in specificity from 92.6 to 100% (1, 5, 6, 8, 9, 13).

The Triage Micro Parasite Panel (Triage) is the first immunoassay to simultaneously test stool for antigens from three common pathogenic protozoan parasites, namely, G. lamblia, E. histolytica/Entamoeba dispar, and C. parvum. Antigens from clinical specimens that are specific for these three parasites are isolated and immobilized on a membrane using specific antibodies. An antibody-enzyme conjugate then binds to specific sites on these antigens. The antigens are detected after the addition of substrate by the formation of color bars in different areas depending on the parasite present and show on the test device as blue-black lines. We report here an evaluation of the Triage Method compared to routine O&P examination (O&P) in a hospital setting.

(This study was presented in part at the 100th American Society for Microbiology General Meeting in Los Angeles, Calif., in May 2000.)

Fresh, unfixed stool specimens were received in the laboratory in their original containers and were tested that day or frozen at −20°C until tested. Only liquid or semiformed stools were assayed. Formalin-ethyl acetate concentration with the Fecal Parasite Concentrator Kit (Remel, Lenexa, Kans.) was performed on fresh unpreserved stool samples. Concentration was achieved by centrifugation at 500 × g for 10 min. Sediments were examined by preparing two 22-by-22-mm-coverslip preparations (one with saline and one with iodine). Permanent smears were prepared using polyvinyl alcohol preservative and were stained by the trichrome staining method. Approximately 250 fields were examined under oil immersion. In addition, a modified acid-fast stain was performed on direct and concentrated samples to evaluate for the presence of C. parvum, Isospora belli, and Cyclospora species. The concentration and permanent-stained smears were prepared and read by medical technologists specifically trained in the area of parasitology.

Only fresh or fresh-frozen unpreserved stool samples can be tested in the Triage assay. Frozen specimens were allowed to reach room temperature prior to testing. To begin specimen preparation for the Triage assay, diluent was dispensed in the indicated volumes into tubes provided by the manufacturer. Stool samples were then added to these tubes and vortexed for 10 s. A filtration device was then inserted into each specimen tube and centrifuged for 5 min at 1,500 to 1,800 × g. A transfer pipette provided in the kit was used to add 500 μl of filtered sample to the center of the Detection Zone of the Test Device. Then, 140 μl of enzyme conjugate was added directly to the Detection Zone with a provided pipette and incubated for 3 min. The Detection Zone was then washed twice, each with 6 drops of Wash Solution. Four drops of substrate were then added to the center of the Detection Zone, followed by incubation for 5 min. Immediately after this last incubation step the results were read. A scientific representative from Biosite Diagnostics, Inc., initially trained two medical technologists who subsequently tested all of the patient samples. For the purposes of this study, a positive control antigen and a negative control antigen provided by the company were tested on each day of use. Under normal circumstances, these controls would only be run once every 30 days or for the first use of a new lot or a new shipment. In addition, the onboard procedural controls had to be acceptable in order to interpret patient results. The permanently stained smear was reviewed for any specimen found to be Triage positive and O&P negative. The supervisor reviewed the smears of such specimens for no less than 30 min. For any unresolved discrepant results, a second antigen detection system, an enzyme-linked immunoassay specific for the parasite in question, was performed according to the manufacturer's recommendations (Alexon, Inc., Sunnyvale, Calif.). A true-positive specimen was defined as one that was positive for any two of the three assay methods (i.e., O&P, Triage, or Alexon).

Five hundred twenty-three stool samples were assayed by routine O&P and by the Triage method. A total of 33 specimens were reported to be positive by one or both methods. Twenty-nine samples were found to be positive for parasites by O&P (some for multiple parasites), and 19 were found to be positive by Triage. The parasites identified by the two methods are shown in Table 1.

TABLE 1.

Comparison of O&P and Triage test results

Parasite	No. of specimens found to be positive by:
	O&P	Triage
Giardia lambliaa	7	9
Entamoeba coli	5	NAd
Entamoeba histolytica/E. disparb	4	5
Iodamoeba buetschlii	4	NA
Cyclospora cayetanensis	3	NA
Endolimax nana	3	NA
Blastocystis hominis	3	NA
Taenia species ova	3	NA
Trichuris trichiura ova	3	NA
Cryptosporidium parvumc	3	5
Entamoeba hartmanni	1	NA

^aThere were two false-negative O&P results for G. lamblia. 

^bThere was one false-negative O&P results for E. histolytica/E. dispar. 

^cThere was one false-positive Triage result and one false-negative O&P result for C. parvum. 

^dNA, not applicable. 

Discrepant results were further analyzed using supervisory review and/or a microtiter enzyme immunoassay (EIA) specific for each of the three parasites. The Triage method accurately detected 18 specimens that contained one of the three parasites that it was designed to detect but did not react in any sample that contained other parasites (Table 1). Two false-negative G. lamblia results, one false-negative E. histolytica/E. dispar result, and one false-negative C. parvum result were recorded for O&P, and one false-positive C. parvum result was recorded for the Triage. Supervisory review was performed on the four specimens with false-negative O&P results. One of the specimens was found to contain rare numbers of G. lamblia, and another contained rare numbers of C. parvum. However, supervisory review was negative for the two remaining specimens with false-negative O&P results. Traditional O&P would have missed 4 of 18 (22.2%) of the 3 parasitic infections included on the Triage panel or 4 of 33 (12.1%) of all specimens with parasites. Based on O&P, supervisory review, Triage findings, and discrepant microtiter EIA analysis, the sensitivities and specificities for the three organisms were, respectively, as follows: G. lamblia, 100 and 100%; E. histolytica/E. dispar, 100 and 100%; and C. parvum, 100 and 99.8% for the Triage and G. lamblia, 77.8 and 100%; E. histolytica/E. dispar; 80 and 100%; and C. parvum, 75 and 100% for the O&P. Of the 523 specimens, 22 (4.2%) needed to be repeated in the Triage assay with the one-quarter sample procedure recommended in the package insert due to background color in the detection zone that obscured the visual reading. All 22 samples upon repeat were interpretable.

There are a variety of ways in which a stool assay for G. lamblia, E. histolytica/E. dispar, and C. parvum antigens could be incorporated into the clinical laboratory. In areas of high prevalence for one or more of these three parasites, it could be used as a screen prior to a complete O&P being performed. It has also been recommended that these assays could be used to replace routine O&P when infection with one of these parasites is the most likely clinical diagnosis (1). Indeed, many studies in addition to ours have shown that the sensitivity of routine O&P is <100% (1, 8, 9, 13). Conversely, if one is located in a geographical area that services a population where a variety of parasites are seen, the Triage could be utilized to look for the presence of one of these pathogens if repeat O&P examinations fail to produce a causative agent. Since many of the EIAs available today have been shown to be superior in sensitivity to routine O&P, their use can be very beneficial in establishing a diagnosis in cases where very low numbers of parasites are present (1, 8, 9, 13). Since the sensitivity and specificity for the Triage with all three parasites are between 99.8 and 100%, a single stool specimen may allow for the correct diagnosis days before a second or third stool sample is tested.

It is well known that E. histolytica causes human disease, whereas E. dispar does not (3, 12), and that laboratory differentiation of E. histolytica and E. dispar by microscopic identification is not possible because the organisms are morphologically identical (3, 7). Isoenzyme analysis of cultured amebae will enable the differentiation of these two species of Entamoeba, but this requires weeks to complete and is not useful for initial diagnosis of parasitic disease (9). Although the Triage does not differentiate between these two species, it would allow for the initial detection of E. histolytica/E. dispar which could then be followed up by a differential test as recommended by the World Health Organization (12).

It must be kept in mind that only fresh or fresh-frozen unpreserved stool samples can be tested by the Triage assay. Specimens would need to be transported to the laboratory for testing as soon as possible after collection, kept refrigerated for up to 72 h prior to delivery, or frozen and delivered to the testing site at a later time in order to maintain specimen integrity for this assay. Each laboratory interested in using this assay will need to establish the best approach for its incorporation into the working scheme for parasitologic analysis. No one approach will be correct for all laboratories, but utilization of this assay will depend on many factors, such as patient population, the physician clientele served, expertise available in the laboratory, ease of performance, turnaround times, assay limitations, and cost-effectiveness. Many of the assays available today have been found to be sensitive, cost-effective, and a rapid means for the detection of pathogenic parasites in stool samples (1, 5).

The cost of performing the Triage assay with trained laboratory personnel in our setting would be approximately $21.19/test, which includes both reagent cost ($19.44/test) and labor (5 min of “hands-on” time/test at $21.00/h = $1.75/test). The cost associated with providing a routine O&P consists of the examination of a concentration and a permanently stained smear of the specimen. At our institution this would cost approximately $13.75/specimen ($1.50 material expense/test + 35 min of hands-on time/test at $21.00/h = $12.25/test). Since up to three O&P specimens may be necessary in order to diagnose a parasitic infection (4), depending on available materials and labor costs, it is possible to save $20.06/patient using the Triage assay if suspicion was high for one of the three parasites this assay detects. Additional savings could be realized if lower-paid, yet well-trained technical personnel were to routinely perform this assay. In addition, utilization of the Triage assay could also be a more cost-effective approach than performing separate immunoassays for each of these three parasites.

An article by Pillai and Kain (10) showed less-favorable results with the Triage compared to ours; however, this study did not include an evaluation of the C. parvum component of the assay, and it compared E. histolytica/E. dispar detection to an EIA method as the “gold standard” rather than to O&P. Their finding of a sensitivity of 68.3% with the Triage assay in the detection of E. histolytica/E. dispar, as explained by the authors, was due to the sensitivity of the two assays, with the EIA detecting down to 250 trophozoites/ml of specimen, whereas the Triage required >1,000 trophozoites/ml of specimen to give a clearly positive result. These authors found the sensitivity of the Triage assay for detecting G. lamblia to be 88.3%, which was also lower than what was found in the current study. Another evaluation of the Triage product has been published very recently (7). That article demonstrated sensitivities and specificities similar to those obtained in our study, and these authors recommend the use of the Triage as an alternate method for performing simultaneous and discrete detection of these three parasites in fecal specimens.

Although the sensitivity and specificity of the Triage assay were determined to be excellent in that study, the authors acknowledge the limited number of positive specimens that were evaluated during the study period in order to evaluate the sensitivity of the assay. A sufficient number of negative specimens was evaluated in the present study to substantiate the specificity of the Triage assay, and the recent article by Garcia et. al. (2000) confirms the sensitivity reported here.

In conclusion, the Triage test is a rapid, easily performed and interpreted EIA for the detection of three prevalent parasites in the United States and shows excellent sensitivity and specificity for all three parasites. In addition, this assay could be utilized by personnel that do not have extensive training in manual parasitological methods.