Objectives: To determine the optimal number of stool specimens needed for the diagnosis of cryptosporidiosis.
Methods: Four hundred thirty-five admissions were reviewed (291 patients) in which stool specimens were examined for Cryptosporidium parvum oocysts (mean of 1.47 specimens per admission), using a modified acid-fast stain. The diagnostic yield of each specimen was determined.
Results: Cryptosporidium parvum oocysts were found in 81 of 435 admissions (18.6%). Ninety-six percent of the positive cases were detected on the first stool specimen analysis, and 100% were detected by the second specimen.
Conclusions: Examination of one specimen is generally appropriate for the diagnosis of cryptosporidiosis in a hospitalized patient with AIDS presenting with diarrhea. Examination of a second specimen may be appropriate if the first specimen is negative and there is a high clinical index of suspicion.