Search in the Abstract Database

Abstracts Search 2019

P857 The microbiota profile reflects disease severity in paediatric onset IBD

M. Malham*1, B. Lilje2, K. Winther3, G. Houen2, P. S. Andersen2, C. Jakobsen1,4

1Hvidovre University Hospital, The Paediatric Department, Hvidovre, Denmark, 2Statens Serum Institut, The Department for Bacteria, Parasites and Fungi, Copenhagen, Denmark, 3Nordsjaellands Hospital, The Paediatric Department, Hilleroed, Denmark, 4Hvidovre University Hospital, The GastroUnit, Hvidovre, Denmark


Over the last decade, reports have emerged that describe a distinct microbiotic profile (decreased diversity and density) in both Crohn’s disease (CD) and ulcerative colitis (UC) which was again distinct from healthy controls. However, in recent years we have been able to sequence the microbiotic profile to the species level, which have changed the interpretation of some of the older studies. In this study, we aimed to describe the microbiotic profile in a cohort of paediatric IBD patients.


We collected faecal samples from a cross-sectional cohort. Faeces were stored at -80 degrees Celsius before analysis. The microbiome analysis was done using 16S and 18S rRNA sequencing with the miSeq instrument. The software ‘BION’ was used for OTU picking. The statistical program ‘R’ was used for further statistical analysis. Faecal calprotectin (FC) was analysed by ELISA on the same faecal samples. As faecal samples were not instantly frozen after collection, we only report on species presence/absence. Patient charts were retrieved and data recorded regarding medical treatment and surgery in the year after faeces samples were collected.


143 patients (77 CD / 58 UC / 8 IBDU) and 34 healthy controls were included. We found a significant difference in richness (number of observed species) between disease groups (controls vs. UC (p < 0.001), controls vs. CD (p = 0.04) and CD vs. UC (p = 0.009) with controls having the highest number of different species and UC the lowest. Moreover, a high degree of intestinal inflammation (assessed by f-calprotectin) and extensive disease localisation was associated with reduced diversity in UC (p = 0.02 and p = 0.04, respectively) but not in CD (p = 0.94 and 0.11, respectively). We identified 9 species that were significantly associated with a healthy microbiome and 2 species that were associated with IBD. The 3 species that were most significantly associated with a healthy microbiome were Akkermansia muciniphila, Gemmiger formicilis and Bacteroides massiliensis. No association was found between the microbiome composition and the need of medical treatment or surgery. Finally, using 18S rRNA analysis, we found no associations between the presence of parasites and IBD.


Patients with IBD had a decreased diversity in their faecal microbiome and the IBD type influenced the degree of the reduced diversity. Moreover, we identified 9 species that were more often present in healthy controls. Finally, we found that the composition of the microbiome was affected by the grade of the intestinal inflammation. These findings should be kept in mind when planning future studies with probiotics as a possible treatment of IBD.