Résumé
A diarrhoeal epidemic in a village close to Vellore was investigated in January 1996. Faecal samples were obtained from 20 subjects with diarrhoea and from 11 individuals without diarrhoea (controls) and were examined for bacterial, viral and parasitic enteropathogens. Water samples from all sources in the village were analysed. The epidemic affected all age groups (overall attack rate 15%). The mean duration of diarrhoea was 11 days. Individuals who consumed water exclusively from a borewell had a lower relative risk (RR) of disease (0.14, 95% Cl 0.02-1.01) compared to users of two open wells (RR 6.93, Cl 0.99-48.66 and RR 7.81, Cl 1.02-59.79, respectively). No conventional bacterial enteropathogens were isolated from the stool samples. Enteroaggregative Escherichia coli (EAggEC) were identified in the stool of 11 of 20 subjects with diarrhoea, and in 1 of 11 control samples (P = 0.02). All the EAggEC isolates from the patients had identical antibiotic sensitivity patterns and produced a toxin in Ussing chamber studies. Serotyping indicated that all the EAggEC from individuals with diarrhoea belonged to one or other of two serotypes. All water samples had high coliform counts and E. coli were cultured from the two open wells but not from the borewell. The evidence suggests that EAggEC was responsible for this outbreak of diarrhoea. EAggEC should be considered as a possible pathogen in unexplained diarrhoeal outbreaks in developing countries.