Intestinal Flora in Children with IBD at the Time of Diagnosis

Barkman, C1; Saalman, R2; Adlerberth, I1; Wold, A1

Journal of Pediatric Gastroenterology & Nutrition: November 2006 - Volume 43 - Issue - p S39
doi: 10.1097/01.mpg.0000256228.93266.a7

1Department of Clinical Bacteriology, Sahlgrenska University Hospital, Göteborg University, Göteborg, Sweden

2Department of Paediatrics, Queen Silvia Children's Hospital, Göteborg University, Göteborg, Sweden

Introduction: The intestinal microbiota may play a role in the disease process of IBD. In this study we examined the small and large intestinal microbiota in children at the time of diagnosis of IBD, before medical treatment had been instituted.

Patients and methods: Faecal samples were obtained from children with suspected IBD before bowel cleaning and duodenal fluids at duodenalscopy. The samples were transported anaerobically and cultivated quantitatively for aerobic and anaerobic bacteria within 24h. After diagnosis was established, microbiota composition was compared between the diagnostic groups; ulcerative colitis (UC n=20), Crohn's disease (CD, n=10) and diseased controls i.e. children in whom the IBD diagnosis was excluded and with no histological signs of mucosal inflammation (n=24). Faecal samples from healthy control children (n=11) were also included for comparison. Children with indeterminate colitis (n=1), other type of colitis (allergic, infectious, unspecific n=8) and celiac disease (n=2) were excluded from the analyses due to small group numbers.

Results: The duodenal microbiota did not differ significantly between the diagnostic groups. In contrast children with UC had a distinctly different faecal microbiota compared with diseased and healthy controls. Thus, children with UC had a significantly lower ratio of anaerobic to facultative bacteria in faeces (respectively 30 compared with 500 in diseased and healthy controls, p=0.0001 and p=0.001). Further, children in the UC group had lower counts of anaerobic bacteria (average: 108.8 Colony Forming Unit, CFU/g faeces) compared with diseased (1010.1 CFU/g faeces) and healthy (1010.2 CFU/g faeces) controls (p = 0.0001 for both). The low proportion of anaerobic bacteria in the UC group was unrelated to frequency of diarrhoea or to precence of blood in stools. Among the anaerobes, bifidobacteria and clostridia were both significantly reduced in children with UC compared with diseased controls (p = 0.002 and p = 0.03, respectively) and healthy controls (p = 0.006 and p = 0.02, respectively), whereas the counts of Bacteroides did not differ between the diagnostic groups. Furthermore, the children presenting with UC had significantly increased ratio of Gram-negative to Gram-positive bacteria in faeces compared with diseased (p = 0.02) and healthy (p = 0.03) controls.

Conclusions: The results suggest that the microbiota in UC is dysbalanced with a relative lack of certain Gram-positive anaerobes and a relative increase in facultative and Gram-negative bacteria. This dysbalanced microbiota in the colon might provoke and perpetuate inflammation.

© 2006 Lippincott Williams & Wilkins, Inc.