Prevalence of attaching and effacing Escherichia coli in stool samples from patients and controls

Enteropathogenic E. coli (EPEC) and enterohemorrhagic E. coli (EHEC) have the ability to cause 'attaching and effacing' (AE) lesions; the genes necessary to cause AE in both of these pathogroups have been identified and termed eae. Using colony hybridization, we screened 237 stool samples from patients with diarrhea, and 237 stool samples from age-matched controls for the presence of E. coli carrying eae. Individual colonies harbouring eae could be recovered from 7 (2.9%) of the patient stools, as well as from 6 (2.5%) of the control stools. All these E. coli isolates were positive in the fluorescence actin staining (FAS) test. In addition, all the samples were also probed for Shiga-like toxin (slt) genes and the EPEC adherence factor (EAF) to evaluate whether testing for eae identified all EHEC and class I EPEC. Of the 7 patient samples harbouring E. coli with eae, 4 had E. coli with eae and slt genes, and 2 had E. coli with eae and EAF sequences. In 2 of the 237 patient stools, E. coli which were eae and EAF negative but slt probe positive could be recovered. These 2 E. coli strains were non-reactive in the FAS test. Of the control samples, none of the E. coli strains, including the 6 samples containing eae positive strains, possessed EAF or slt-sequences. In concrete terms, the similar eae incidence found in both E. coli isolates from patients and controls is currently of limited clinical diagnostic value and more importantly, the eae probe could not identify all slt-harbouring E. coli. On the basis of these results, the use of the eae-probe cannot be recommended in preference to the slt probes for the detection of EHEC.