Efficacy of different methods for detection of low Cryptosporidium parvum oocyst numbers or antigen concentrations in stool specimens.

The detection of Cryptosporidium parvum oocysts in stool specimens by acid-fast (AF) stains or immunofluorescence assays (IFA) requires the presence of large numbers of oocysts. To determine whether new commercially available enzyme immunoassays (EIAs) are more sensitive alternatives, three EIAs, a direct IFA, and the modified cold Kinyoun AF stain were compared, particularly with respect to detection of low oocyst numbers or antigen concentrations. Thirty-one negative and 31 calf stool-enriched human stool specimens were tested. One EIA method detected only nine positive specimens, demonstrating a sensitivity significantly less (p < 0.0001) than that of the IFA, the AF stain, and the other two EIAs. No differences could be found with respect to specificity. In addition, serial dilutions of 28 patients' stool samples containing cryptosporidian oocysts were prepared and examined using two EIAs, IFA, and the AF stain. One EIA yielded significantly inferior results (p < 0.0001), whereas the other one and the two microscopic methods did not differ significantly in either part of the study. The results indicate that the new EIAs do not exhibit higher sensitivities for detection of Cryptosporidium parvum than the two routinely used microscopic methods. Thus, for most laboratories, the IFA or AF stain may still represent the preferred method for the diagnosis of cryptosporidiosis.