See Dubinsky MC et al on page 1105 in CGH;See editorial on page 1038. Background & Aims:Childhood-onset inflammatory bowel disease (IBD) might be etiologically different from adult-onset IBD. We analyzed disease phenotypes and progression of childhood-onset disease and compared them with characteristics of adult-onset disease in patients in Scotland. ; 43% vs 3%; P < .0001; OR, 23.36; 95% CI, 13.45-40.59) with less isolated ileal (L1; 2% vs 31%; P < .0001; OR, 0.06; 95% CI, 0.03-0.12) or colonic disease (L2; 15% vs 36%; P < .0001; OR, 0.31; 95% CI, 0.21-0.46). UC was extensive in 82% of the children at diagnosis, versus 48% of adults (P < .0001; OR, 5.08; 95% CI, 2.73-9.45); 46% of the children progressed to develop extensive colitis during followup. Forty-six percent of children with CD and 35% with UC required immunomodulatory therapy within 12 months of diagnosis. The median time to first surgery was longer in childhood-onset than adult-onset patients with CD (13.7 vs 7.8 years; P < .001); the reverse was true

The gastrointestinal microbiota is considered important in infl ammatory bowel disease (IBD) pathogenesis. Discoveries from established disease cohorts report reduced bacterial diversity, changes in bacterial composition, and a protective role for Faecalibacterium prausnitzii in Crohn ' s disease (CD). The majority of studies to date are however potentially confounded by the effect of treatment and a reliance on established rather than de-novo disease. METHODS:Microbial changes at diagnosis were examined by biopsying the colonic mucosa of 37 children: 25 with newly presenting, untreated IBD with active colitis (13 CD and 12 ulcerative colitis (UC)), and 12 pediatric controls with a macroscopically and microscopically normal colon. We utilized a dual-methodology approach with pyrosequencing (threshold >10,000 reads) and confi rmatory real-time PCR (RT-PCR). RESULTS:Threshold pyrosequencing output was obtained on 34 subjects (11 CD, 11 UC, 12 controls). No signifi cant changes were noted at phylum level among the Bacteroidetes, Firmicutes, or Proteobacteria. A signifi cant reduction in bacterial α -diversity was noted in CD vs. controls by three methods (Shannon, Simpson, and phylogenetic diversity) but not in UC vs. controls. An increase in Faecalibacterium was observed in CD compared with controls by pyrosequencing (mean 16.7 % vs. 9.1 % of reads, P = 0.02) and replicated by specifi c F. prausnitzii RT-PCR (36.0 % vs. 19.0 % of total bacteria, P = 0.02). No disease-specifi c clustering was evident on principal components analysis. CONCLUSIONS: Our results offer a comprehensive examination of the IBD mucosal microbiota at diagnosis, unaffected by therapeutic confounders or changes over time. Our results challenge the current model of a protective role for F. prausnitzii in CD, suggesting a more dynamic role for this organism than previously described.SUPPLEMENTARY MATERIAL is linked to the online version of the paper at

Microbial community structure differs between healthy controls, patients who have an enduring response to exclusive enteral nutrition, and those who relapse early on introduction of normal diet. Our novel Bayesian approach to these differences is able to predict sustained remission after exclusive enteral nutrition.

These NOD2/CARD 15 variants in the Scottish early onset CD population have a definite, albeit relatively small contribution to CD susceptibility (PAR 7.9%) but a major impact on phenotype. In particular NOD2/CARD15 variants are strongly associated with several markers of disease severity in pediatric CD, notably need for surgery.