Background: Crohn's disease usually recurs after “curative” resection. This may relate to specific microbial populations playing a pathogenic role.
Methods: Faecal samples were obtained peri-operatively (baseline) and at 6, 12 and 18 months after surgery from 130 patients enrolled in the prospective POCER study. Endoscopic disease recurrence was assessed (Rutgeerts Score ≥i2) at six months (in 2/3 of patients in the active endoscopic care arm) and at 18 months in all patients. DNA was extracted using the MoBio Powersoil DNA extraction kit, and the V2 region of the bacterial 16S rRNA gene sequenced (MiSeq). Data were processed using the QIIME pipeline; alpha and beta diversity were assessed on samples after rarefaction to 10,000 reads per sample. Alpha diversity was compared using the Shannon's index. Weighted UniFrac distances assessing beta diversity were compared using the vegan package (adonis function) in R. Differential abundance between remission and recurrence was assessed at genus level using MetagenomeSeq with cumulative sum scaling normalisation, FDR correction (FDR P value), further adjustment for the number of comparisons (Adj. P Value), and adjustment for baseline patient characteristics (smoking, age, gender, body mass index and antibiotics) in R.
Results: Diversity increased significantly after surgery (all patients, baseline versus 18 months (p=0.048). At 6 months, diversity was significantly greater for patients who remained in remission compared to those with recurrence (p=0.04); at 18 months, a similar trend was observed but the results were not significant (p=0.185).
Overall bacterial composition differed between recurrence and remission at 18 months (p=0.008), as well as over time (all patients and all samples: baseline, six, 12 and 18 months; p=0.001).
Nine genera (four from the order Clostridiales, two from the orders Lactobacillales and Bacteroidales) were differentially abundant between subjects with disease recurrence compared to remission as shown.
Conclusions: Specific bacterial genera are associated with disease recurrence after Crohn's disease resection. Positive associations need to be investigated for a possible causative role, while negative associations (decreases within the orders Clostridiales and Lactobacillales) require investigation for a possible protective role.