LAUSR

Over 30 years ago, Rutgeerts et al published a seminal paper detailing the endoscopic recurrence of Crohn’s disease in the neoterminal ileum following ileocecal resection. The eponymous Rutgeerts scoring scale in that paper has been widely adopted and remains in wide clinical use today. The ability to directly visualise recurrent lesions prior to any symptoms or radiologic abnormalities has changed practice. Ileocolonoscopy within the first year after ileocecal or ileocolonic resection to assess recurrence and the Rutgeerts score is now standard of care (table 1). Crohn’s disease recurrence is frequent but not universal and progresses over months to years, enough time to intervene, assuming an effective therapy. Indeed monoclonal antiTNF therapy has shown some ability to halt or prevent recurrence. However, even in this era of biological therapy, recurrence of Crohn’s disease remains high, afflicting half or more patients after surgery. A better understanding of the sequence of events and pathogenesis of ileal tissue lesions is needed to better target therapies to prevent recurrence. For decades the progression of Crohn’s disease in the neoterminal ileum has been considered an excellent opportunity to study the pathogenesis of Crohn’s disease, as it seems to recapitulate the disease in a single accessible tissue, with aphthous lesions leading sequentially to small ulcers, larger ulcers and stenosis. Because manpower and available patients are constraints at any one medical centre, this ‘human model’ of Crohn’s disease had been underused until a collaborative, multicentre group of investigators in France named by the acronym REMIND, was formed to tackle this issue. The most recent findings of the REMIND group is presented in this issue. Previously this group has identified oligoclonal expansions of T cells in the ileal mucosa which were increased in smokers and in patients with ileal recurrence, compatible with and adaptive T cell response to some antigen, possibly a microbiota antigen. Subsequently, the mucosaassociated microbiota was analysed in ileal biopsy samples taken at surgery and at ileocolonoscopy 6 months later. Surgery itself was found to change the ileal mucosal microbiota. The half of the patients with recurrence had greater changes, with significant reduction in diversity and expansion of proteobacteria (alpha and gamma), Ruminococcus gnavus group and Corynebacterium genera, and concomitant reduction in Lachnospiraceae and Ruminococcus, the latter organisms known to produce beneficial short chain fatty acids important in homoeostasis. These results confirmed previous studies based on analysis of faecal samples. In the present report, Buisson et al have extended the above study to adherent invasive Escherichia coli (AIEC), a member of the gamma proteobacteria. AIEC has intrigued investigators since their discovery in Crohn’s ileum by DarfeuilleMichaud et al. As indicated by the name, these bacteria are able to adhere to intestinal epithelial cells and to invade and replicate in epithelial cells and macrophages without killing them. AIEC forms a membrane enclosed vesicle within the cell, which might explain how the AIEC does not kill the host cell. These properties result in impaired epithelial barrier function and the production of large amounts of tumour necrosis factor by infected macrophages. AIEC is associated with microbiota dysbiosis and, in one mouse model, was able to alter microbiota composition. AIEC metabolism of fucose to proprionate has been shown to induce IL1β production and drive intestinal inflammation in mice. The prevalence of AIEC in patients with Crohn’s disease has been based on faecal samples obtained at one point in time. This study is the first and largest prospective sampling of ileum mucosa AIEC. Two biopsies were taken from the ileum at surgery and at ileocolonoscopy 6 months after surgery. The authors applied a rigorous controlled process to recover AIEC, which involved growing E. coli and other enterobacteriaceae from the biopsies, screening the cultured bacteria for invasion of a human intestinal epithelial cell line, then documenting that the recovered E. coli were able to infect a human epithelial cell line and a human macrophage cell line without causing injury or death of the cells. This is a laborious and timeconsuming process, made necessary by the lack of any molecular markers that can reliably identify AIEC, and is not likely to be adopted by clinical laboratories. Using this method, biopsy samples could be assigned as positive or negative for AIEC; the numbers of AIEC in a given sample could not be assessed, nor could the location of the AIEC in the mucus or in tissue cells. There are a number of interesting findings from this study. AIEC detection in the ileum at either surgery (M0) or 6 months later (M6) was significantly associated with higher rates of recurrence compared with those patients negative for AIEC. However, only a minority of Crohn’s patients with or without recurrence were positive for AIEC, 30% and 17%, respectively. Only 5% of patients were positive for ileum AIEC at both M0 and M6. This is a surprising finding in that AIEC would seem to have a survival advantage over other E. coli unable to adhere and invade. This remains to be explained, but AIEC is in competition with other E. coli and Enterobacteriaceae, and some of the medications used to treat Crohn’s such as azathioprine and mercaptopurine, have been shown to suppress AIEC. AIEC might be a marker of dysbiosis of the microbiota, and it may be the dysbiosis that might result in recurrence. Supporting that possibility, ileum samples positive for AIEC had higher numbers of proteobacteria, such as Klebsiella, Enterococcus and Proteus and increased number of R. gnavus and Ruminococcus torque. The association of AIEC with recurrence in this study hinges on whether a Rugeerts score of i1 with <5 aphthous lesions is a recurrence. Clinically Rutgeerts i0 and i1 are not considered a recurrence. The authors acknowledge this but have named Rutgeerts i1 as ‘very early ileal lesions’, that is, recurrence. The i1 group had the highest frequency of AIEC positivity of any