Comparison of intestinal microbiota profiles of faecal samples, rectal swabs and mucosal biopsies in patients with Inflammatory Bowel Disease. A multicentre, observational study

Abstract Background Since the gut microbiota is altered in patients with Inflammatory Bowel Disease (IBD), microbiome-based biomarkers may be useful for diagnosing and monitoring of IBD. Currently, endoscopy is the gold standard for diagnosis and monitoring disease activity, which can be a high burden for patients. This study aimed to compare intestinal microbiota profiles between three different sampling methods faecal samples, rectal swabs and colonic mucosal biopsies. This could provide more insight into the microbiota composition in different sample types and might contribute to the development of a less invasive biomarker for diagnosing and monitoring IBD. Methods Patients with IBD (Crohn’s Disease or Ulcerative Colitis) who were scheduled for endoscopy were asked to participate. Prior to bowel preparation, a fecal sample and rectal swab were collected. During colonoscopy, mucosal biopsies were derived 20 cm ab ano. Microbiota composition was analyzed by IS-pro, a PCR technique based on species-specific differences in the 16S-23S interspace region of the bacterial ribosomal DNA. Microbiota profiles of the three different sample types were compared within the same patient (for each patient: fecal sample vs. mucosal biopsy, fecal sample vs. rectal swab, rectal swab vs. mucosal biopsy), and between different patients (for each patient: fecal sample of one patient vs. mucosal biopsies of all patients, fecal sample of one patient vs. rectal swabs of all patients, etcetera). Results A total of 200 patients were included. For each patient, similarity of microbiota composition between two sample types was assessed by calculating the Pearson’s correlation (expressed as R2). R2 values of all patients were combined in boxplots. We found a significantly higher correlation between the microbiota profiles of different sample types within the same patient, than between microbiota profiles of different sample types of different patients (median R2 0.27–0.33 and 0.02–0.03 respectively, Figure 1). However, correlation between different sample types from the same patient was still relatively low. The highest correlation was found between microbiota profiles of faecal samples and rectal swabs (median R2 0.33, ICR 0.17–0.54). On phylum level, the highest correlation was found in the Bacteroidetes phylum (Figure 2). For a global analysis of all versus all samples, we generated a clustered heat map, which confirmed the previous finding that microbiota profiles from faecal samples and rectal swabs were most similar to each other. Conclusion Microbiota composition in different sample types from the same patient were more similar to each other than to profiles from different patients. Microbiota profiles of faecal samples and rectal swabs were most identical.