抄録
Seven different staining techniques, enzyme linked immunosorbent assay, immunofluorescence and two counting methods were compared for the detection of Cryptosporidium oocysts in faecal samples. Auramine phenol and modified Ziehl-Neelsen techniques were the most useful staining procedures with a detection level of 20,000 oocysts per gram. Both of these methods are suitable for the detection of C. parvum oocysts in samples from clinically affected hosts, where large numbers of oocysts are generally excreted. The former method was marginally clearer when lower numbers of oocysts were present. Immunofluorescence was more sensitive with a detection limit of 5,000 oocysts per gram, whereas enzyme linked immunosorbent assay had a detection limit of 50,000 oocysts per gram of faeces. The Fuchs Rosenthal counting method was relatively insensitive with a reproducible detection limit of 5×105 oocysts per gram, but very accurate with a percentage standard error over the range assessed of less than 5%. The “sensitive” counting method, though extremely laborious, had a detection limit of 50 oocysts per gram, but was less accurate at the lower end of the range with a percentage standard error of more than 20% when concentrations were of 500 oocysts per gram or less.
