Search in the Abstract Database

Abstracts Search 2019

P572 Half of children with acute severe colitis have predominant single faecal bacterial species, mostly Escherichia coli: Microbiome results from the PRASCO trial

J. Bishai1, G. Abitbol2, G. Focht2, M. Schirmer1, D. Marcus2, B. Yerushalmi3, M. Aloi4, A. M. Griffiths5, L. Albenberg1, K-L. Kolho6, S. Cohen7, A. Assa8, A. Levine9, H. Vlamakis1, R. Xavier1, D. Turner*2

1The Broad Institute of MIT and Harvard, Center for Computational and Integrative Biology, Boston, USA, 2Shaare Zedek Medical Center, The Juliet Keidan Institute of Paediatric Gastroenterology and Nutrition, Jerusalem, Israel, 3Faculty of Health Sciences, Ben-Gurion University of the Negev, Department of Gastroenterology and Hepatology, Beer Sheva, Israel, 4Sapienza University of Rome, Rome, Italy, 5The Hospital for Sick Children (SickKids), Toronto, Canada, 6Hospital for Children and Adolescents, Children′s Hospital, Helsinki University, Helsinki, Finland, 7Tel Aviv Sourasky Medical Center, Tel Aviv, Israel, 8Schneider Children's Medical Center, Petach Tiqua, Israel, 9Wolfson Medical Center, Holon, Israel


Acute severe ulcerative colitis (ASC) is one of the few medical emergencies of gastroenterology. We previously reported the clinical results of the PRASCO randomised controlled trial in which oral antibiotics (AB) improved disease activity after 5 days of treatment in children with ASC. This suggests that the microbiome is implicated in the aetiology and progression of ASC. We thus aimed to explore the microbiome of children enrolled in the PRASCO trial.


In the PRASCO trial, 26 children with ASC were randomised: 11 received IV corticosteroids (IVCS) and 15 received four oral AB's (amoxycillin, doxycillin/ciprofloxacin, metronidazole, vancomycin) in addition to IVCS. Stool samples were collected at regular intervals during admission. Metagenomic sequencing was performed using Illumina Nextera XT library preparation kit on a HiSeq platform. After filtering low quality and human reads using the KneadData pipeline, species-level taxonomic abundances were inferred for all samples using MetaPhlAn2.


At baseline before treatment, 14/26 (54%) children harboured more than 30% of a single species: 9 (35%) with Escherichia coli, 2 (7.7%) Haemophilus species, and 1 (3.8%) each with Klebsiella pneumoniae, Barnesiella intestinihominis or Parabacteroides sp. (none with Fusobacterium varium). interestingly, of the 15 children in the AB arm, 11 (73%) had a transient relative bloom of >50% Enterobacteriaceae (mainly E. coli) after treatment, compared with only 3/11 in the IVCS only arm (p = 0.02). Despite this, disease activity at Day 5 was lower in the AB arm and time to remission was shorter (p = 0.049). When analysing all samples, several species of Lachnospiraceae family and Clostridium bartlettii were significantly and positively associated with remission. C. bartletti and R. torques (previously observed to be depleted in IBD and known butyrate producers) were lower in children with more severe disease.


This is the first report of microbiome pattern in ASC, showing provocative results. Over half of children with ASC have gut microbiomes highly enriched (>30%) at baseline with a specific species, mostly Enterobacteriaceae. This association may suggest that some ASC episodes are triggered by yet unknown enteric infections. While our antibiotic cocktail transiently exacerbated this enrichment it reduced disease activity, possibly due to niche availability after treatment, which can subsequently be filled by butyrate-producing microbes. In addition, the enrichment is relative to other species; it is possible that the absolute count was much lower in the AB group. Dysregulation of butyrate production may be associated with ASC in children.

This study was supported by grants from: IOIBD, Helmsley Charitable Trust and Janssen