A PROSPECTIVE-STUDY OF SHIGA-LIKE TOXIN-ASSOCIATED DIARRHEA IN A PEDIATRIC POPULATION

Although population-based studies have shown that children have the hi ghest age-specific incidence of infection with the Shiga-like toxin-pr oducing E. coli (SLTEC), these sporadic case series were not focused s pecifically on the pediatric age group. We undertook a prospective stu dy to determine the frequency of detection of SLT in an exclusively pe diatric population. The study design minimized ascertainment and refer ral bias by systematically defining the population by the presence of diarrheal symptoms rather than by specific diagnosis, previous submiss ion of stool for culture, or referral to a diarrhea study. All childre n <10 years of age hospitalized at a tertiary care pediatric hospital, irrespective of admission diagnosis, were surveyed prospectively at a dmission and for 2 days thereafter for the presence of defined diarrhe al symptoms. From May 1, 1991, to April 30, 1999, 227 patients and 92 age- and season-matched controls were enrolled. Fecal SLT was detected in six (2.6%) patients, three of whom had E. coli O157:H7 organisms w ere isolated; SLT was not found in any of the controls. SLT was more c ommonly detected in children 2-10 years of age and in bloody stools. S almonella was isolated in six (2.6%) cases, Shigella in five (2.2%), a nd Yersinia in three (1.3%); rotavirus was detected in 46 (20.3%). Two patients with SLT-associated diarrhea had hemolytic uremic syndrome ( HUS), and four had hemorrhagic colitis. SLT-associated diarrhea occurr ed in the summer and fall months in contradistinction to that with rot avirus, which occurred in the winter and spring. Because enteric infec tion with SLTEC may have serious sequelae, such as HUS, and because it occurs with a frequency comparable to that of other bacterial enterop athogens, the evaluation of diarrhea in pediatric patients should incl ude a search for SLTEC, particularly E. coli O157:H7.