Microsporidiosis

Microsporidia and Cryptosporidium parvum are important agents of enteritis, capable of causing severe chronic disease in immunocompromised individuals.l,2 Currently, detection of these organisms in stool specimens requires at least 2 special stains or procedures; a chemofluorescent agent or a modified trichrome stain is commonly used to detect microsporidia,3-5 and an acid-fast stain, direct immunofluorescent assay (DFA), or enzyme immunoassay is used to detect C parvum.6-8 Recently, Ignatius et a19 described a combination acid-fast-trichrome (AFT) stain, which allows detection of these organisms, as well as Cyclospora cayetanensis and Isospora belli, with a single procedure. We modified this AFT procedure, incorporating the use of commercially available reagents, and compared the results of the modified AFT stain with those of our standard procedures for stool specimens received in the laboratory for a 15-month period.

METHODS

A total of 470 fecal specimens for which an examination for microsporidia, C parvum or both was requested were included in the study. Specimens were received in 10% formalin and concentrated using the formalin-ethyl acetate sedimentation procedure prior to testing.10 Approximately 10 iL of concentrated specimen was used to prepare smears for all staining procedures. In accordance with the ordering physicians' request, 373 specimens were examined for microsporidia using a modified trichrome stain (Meridian Diagnostics, Inc, Cincinnati, Ohio), and 422 specimens were evaluated for C parvum using the Merifluor Cryptosporidia/Giardia DFA (Meridian Diagnostics, Inc), according to the manufacturer's directions. The modified trichrome stain and DFA were performed daily, Monday through Friday. The AFT stain was performed in batch, once per week. Control material containing microsporidia, C parvum, and C cayetanensis from previously positive patient specimens was stained with the AFT stain along with each batch of patient specimens.

For the AFT stain, approximately 10 uL of concentrated specimen was thinly spread on glass slide and allowed to air dry. The specimen was fixed in methanol for 5 minutes, flooded with carbol fuchsin (TB Kinyoun Stain Reagent A, BBL Microbiology Systems, Cockeysville, Md), and incubated for 10 minutes at room temperature. The slide was then washed with tap water, decolorized with 0.5% HCl-alcohol (TB Decolorizer F, BBL Microbiology Systems), and washed again with tap water. Next, the slide was stained with Trichrome Blue for Microsporidia (Medical Chemical, Santa Monica, Calif) for 30 minutes at 37degC. The slide was then dipped in 90% acid alcohol and 95% ethanol, followed by a 5-minute soak in 95% ethanol. The slide was then placed in absolute alcohol for 10 minutes followed by 10 minutes in HistoSolv X (Biochemical Sciences, Inc, Swedesboro, NJ). The slide was allowed to air dry and examined using oil immersion, total magnification of X 1000.

When discrepancies occurred between the AFT results and the DFA findings, the technologist who initially performed the testing reviewed the original slides and also repeated the staining procedures on newly prepared smears. To further evaluate these discrepancies, the dimethyl sulfoxide (DMSO)modified acid-fast stain (Trend Scientific Inc, St Paul, Minn) and an enzyme immunoassay (EIA) for C parvum (Alexon ProSpecT Cryptosporidium Microplate Assay Alexon Inc, Sunnyvale, Calif) were also performed, and patient medical records were reviewed.

RESULTS

All 6 specimens containing microsporidia were positive by the AFT and the modified trichrome stains. Fourteen of 19 specimens containing C parvum were detected by AFT and DFA, 1 was detected by DFA only (the slide contained very few fluorescent organisms), and 4 were detected by AFT only (2 patients with 2 specimens each, received on different days). While the number of organisms present on the latter 4 slides was not precisely quantitated, several organisms were easily seen in most fields examined. For the 5 specimens with discrepancies, the DFA and the AFT stained smears were reexamined, and new smears were prepared, stained, and examined with the same results. The 4 specimens that were positive for C parvum by AFT alone were further evaluated with the DMSO-modified acid-fast stain and the EIA for C. parvum. Organisms of the characteristic size, shape, and appearance were clearly observed in the DMSO-modified acid-fast stain; however, the specimens were negative by EIA.

Review of the patients' medical records showed that 1 patient had been diagnosed as having cryptosporidiosis 8 months earlier based on the presence of organisms in the stool DFA. At that time, the patient was placed on paromomycin and experienced relief of symptoms. The patient had recently become intolerant to therapy and had discontinued taking the medication. A second patient had no previous diagnosis of cryptosporidiosis, but experienced relief of current symptoms following paromomycin therapy which was instituted as a result of the laboratory's report of C parvum in the AFT and modified acid-faststained fecal smears.

None of the specimens included in this evaluation contained C cayetanensis; however, 1 belli was observed in 1 AFT stained smear. Organisms with the characteristic morphology of 1 belli had been previously observed in the routine ova and parasite examination of this specimen, which consisted of a wet mount of concentrated specimen and trichrome stain of an unconcentrated specimen.

The Figure, A through C, shows the appearance of microsporidia, C parvum, and C cayetanensis in the positive control fecal material stained with the AFT stain and examined under oil at a power of x 1000. The Figure, D, shows the appearance of I belli in a patient specimen stained with the AFT stain and examined at the same magnification. The spores of the microsporidia are 1.0 to 2.5 (mu)m in length, have an oval to round shape, and appear light pink with an unstained polar "vacuole" or "belt." Oocysts of C parvum measure 4-6 (mu)m long, have an irregular round or oval shape, and appear bright pink. The oocysts of C cayetanensis are 8 to 10 wm long, have a regular round shape, and are pale pink or unstained. The oocysts of 1 belli are 20 to 35 (mu)m in length and have an ovoid shape with tapered ends, and the internal sporoblasts appear bright pink.

COMMENT

The AFT stain appears to be a promising method for detecting microsporidia; however, the number of microsporidia detected in our study was too few to conclusively determine the utility of the assay for this organism. Further studies need to be performed to verify our results.

The AFT stain performed well for C parvum, detecting 18 of 19 positive specimens. The specimen missed with the AFT stain contained very few fluorescent organisms, suggesting that the DFA may be better able to detect rare numbers of organism. The DFA stain was found to have a higher sensitivity than the modified acid fast stain in a study by Alles et a1,6 in which the sensitivity of the DFA was 93% compared with 54.8% for the modified acid-fast stain; however, this was not true in an evaluation by Kehl et al,ll in which the sensitivity of the DFA and the acidfast stain both were 96%. An important point made in the latter study that must always be considered in the selection of a laboratory assay was that the immunoassay reagents were more expensive than those for the acid-fast stain. This is true when compared with the AFT stain as well.

Interestingly, we found that the AFT stain detected organisms that were not recognized by either the DFA or the EIA. This type of discrepancy has been reported before. In the study by Alles et al,6 the DFA was negative but the acid-fast stain was positive for 3 of 42 C parvumcontaining specimens. The false-negative immunoassays may simply have been attributable to sampling error, but other possibilities include the presence of substances in the concentrated specimen that interfere with antibody binding or the absence of the appropriate antibody binding sites. In one of our cases, previous antimicrobial therapy may have altered these binding sites.

When performing the AFT stain, we found that careful adherence to the protocol was critical. Inclusion of a control slide was important to determine if optimal staining had been achieved. This was especially important for the microsporidia, which stain less intensely than C parvum or C cayetanensis. Occasionally during the evaluation, the control material was found to have stained poorly. In these cases, the patient and control smears were prepared again and restained with better results. We also found that, when examining the AFT stained smear, careful evaluation of the size and shape of suspicious organisms was essential, as other objects in the fecal material also stain pink. However, with these considerations in mind, use of the AFT stain can provide laboratories with a single procedure for the detection of microsporidia and C parvum.

We thank Jane Diblasi, Susan Kelso, and Bhavani Vishwanath for their assistance in completing this evaluation.

References

1. Current WL, Garcia L.S. Cryptosporidiosis. Clin Microbiol Rev. 1991; 4:325358.

2. Didier ES. Microsporidiosis. Clin Infect Dis. 1998;27:1-8.

3. DeGirolami PC, Ezratty CR, Desai G, et al. Diagnosis of intestinal microsporidiosis by examination of stool and duodenal aspirate with Weber's modified trichrome and Uvitex 2B stains. J Clin Microbiol. 1995;33:805-810.

4. Didier ES, Orenstien JM, Aldras A, Bertucci D, Rogers LB, Janney FA. Comparison of three staining methods for detecting microsporidia in fluids. J Clin Microbiol. 1995;33:3138-3145.

5. Ignatius R, Henschel S, Liesenfeld O, et al. Comparative evaluation of modified trichrome and Uvitex 2B stains for detection of low numbers of microsporidial spores in stool specimens. / Clin Microbiol. 1997;35:2266-2269.

6. Allen A, Waldron MA, Sierra LS, Mattia AR. Prospective comparison of direct immunofluorescence and conventional staining methods for detection of Giardia and Cryptosporidium spp in human fecal specimens. J Clin Microbiol. 1995;33: 1632-1634.

7. Garcia LS, Shimizu RY. Evaluation of nine immunoassay kits (enzyme immunoassay and direct fluorescence) for detection of Giardia lamblia and Cryptosporidium parvum in human fecal specimens. / Clin Microbiol. 1997;35:15261529.

8. Rosenblatt JE, Sloan LM. 1993. Evaluation of an enzyme-linked immunosorbent assay for detection of Cryptosporidium spp in stool specimens. J Clin Microbiol. 1993;31:1468-1471.

9. Ignatius R, Lehmann M, Regnath T, et al. A new acid-fast trichrome stain for simultaneous detection of Cryptosporidium parvum and microsporidial species in stool specimens, J Clin Microbial. 1997;35:446-449.

10. Garcia LS, Bruckner DA, eds. Diagnostic Medical Parasitology. 31' ed. Washington, DC: ASM Press; 1997:614-618.

11. Kehl KSC, Cicerello H, Havens PL. Comparison of four different methods for detection of Cryptosporidium species. J Clin Microbiol. 1995;33:416-418.

Barbara S. Reisner, PhD; Judy Spring, MT(ASCP)

Accepted for publication October 7, 1999.

From the Department of Pathology, University of Texas Medical Branch, Galveston.

Presented in part at the 98th General Meeting of the American Society for Microbiology, Atlanta, Ga, May 20, 1998.

Reprints: Barbara S. Reisner, PhD, University of Texas Medical Branch, 301 University Blvd, Galveston, TX 77555-0740 (e-mail: breisner@utmb.edu).

Copyright College of American Pathologists May 2000
Provided by ProQuest Information and Learning Company. All rights Reserved