Leukocytosis in children with Escherichia coli 0157 : H7 enteritis developing the hemolytic-uremic syndrome

Background Fewer than 10% of children with Escherichia coli O157:H7 enterit is develop hemolytic-uremic syndrome (HUS). Objective. To determine whether circulating leukocytes are independent risk markers of developing HUS during E. coli O157:H7 enteritis. Methods. We reviewed the charts of all children with culture-proved E, coli O157:H7 infections seen at Sainte-Justine Hospital between 1987 and 1997, Epidemiologic data, laboratory indices and circulating leukocytes counts we re noted. HUS diagnosis was validated with independent HUS patient lists fr om the pediatric nephrology services of tertiary care hospitals in the Mont real metropolitan area. The date of onset of enteritis was determined by tw o independent observers. Leukocyte counts were compared among the following independent groups: (1) uncomplicated O157:H7 enteritis (Group 1); (2) O15 7:H7 enteritis with the subsequent development of HUS (Group 2); (3) HUS al ready present at the time of medical consultation (Group 3). Results, There were 369 children with E, coli O157:H7 infection. A complete blood count was not performed in 114 (31%) patients. Observers disagreed o n the date of onset of gastroenteritis in 34 (9%) children only (kappa 0.92 ). The study population thus included 221 patients: Group 1, n = 161; Group 2, n = 27; and Group 3, n = 33. Patients developing HUS (Group 2) presente d greater total leukocyte (P < 0.008), polymorphonuclear (P < 0.008) and mo nocyte (P < 0.07) counts than those with an uncomplicated course (Group 1), Logistic regression analysis showed that young age [odds ratio (OR), 0.98; 95% confidence interval (CI), 0.96 to 0.99], duration of enteric prodrome less than or equal to 3 days (OR 4.8, 95% CI 1.13 to 20.7) and initial leuk ocytosis (OR 1.22, 95% CI, 1.11 to 1.35) were independent predictors of HUS . Conclusions. Based on the variables identified above, further studies are n eeded to determine whether the inflammatory response of the host represents only a marker of the severity of gastrointestinal infection or whether, al ternatively, it is a pathophysiologic factor that leads to HUS.