The performance of a polymerase chain reaction (PCR) method for detection of Escherichia coli O157, previously validated on DNA extracted from pure cultures, was evaluated on spiked cattle swabs through an interlaboratory trial, including 12 participating laboratories from 11 European countries. Twelve cattle swab samples, spiked at 4 levels (0, 1-10, 10-100, and 100-1000 colony-forming units, in triplicate) with E. coli O157 were prepared centrally in the originating laboratory; the receiving laboratories performed pre-PCR treatment followed by PCR. The results were reported as positive when the correct amplicons were present after gel electrophoresis. The statistical analysis, performed on 10 sets of reported results, determined the diagnostic sensitivity to be 92.2%. The diagnostic specificity was 100%. The accordance (repeatability) was 90.0%, calculated from all positive inoculation levels. The concordance (reproducibility) was 85.0%, calculated from all positive inoculation levels. The concordance odds ratio (degree of interlaboratory variation calculated from all positive inoculation levels) was 1.58, indicating the robustness of the PCR method. Thus, the interlaboratory variation due to personnel, reagents, minor temperature or pH fluctuations and, not least, thermal cyclers, did not affect the performance of the method, which is currently being considered as part of an intenational PCR standard.
Read full abstract