BackgroundThe implementation of cost-effective intervention strategies for zoonotic protozoa relies on the development of sensitive and accurate diagnostic methods. We carried out a study to evaluate the accuracy of a PCR method for the detection ofCryptosporidiumspp. oocysts in fecal samples from cattle.MethodsFecal samples were spiked with different numbers of oocysts and the limit of detection of the method was determined. Two methods of DNA extraction were assessed: glass beads and freeze-thawing using liquid nitrogen. A nested PCR approach was developed targeting theCryptosporidiumSSUrRNAandTRAP-C2genes. Agreement between the diagnosis ofCryptosporidiumspp. at the SSU rRNA and TRAP-C2 loci was quantified using the κ-coefficient.ResultsCompared with the freeze-thawing method, the glass beads method was found to be a better way of extracting DNA fromCryptosporidiumoocysts (sensitivities were 83 and 100%, respectively). The limits of detection for glass beads and freeze-thaw were low, 1 and 10 oocyst/g fecal samples, respectively. Forty-six percent of the field samples previously classified as negative forCryptosporidium parvumby the floatation-concentration and enzyme-linked immunosorbent assay methods showed DNA with the PCR protocol.ConclusionPrimers for SSU rRNA are more successful in producing an amplification than primers for theTRAP-C2gene which makes the former PCR protocol the approach of choice for detectingCryptosporidium parvumoocysts in field samples.