The Breath Metabolome in Patients With Pouch Disorders

Abstract

Introduction: Ileal Pouch Anal Anastomosis (IPAA) in patients with ulcerative colitis (UC) is associated with complications like pouchitis or Crohn's disease of the pouch (CD). Pouch disorders are diagnosed with a combination of pouchoscopy, histology, clinical findings and response to antibiotics. Our previous work indicates an inflammatory bowel disease (IBD) specific breath metabolome in patients naive to surgery. In a novel, non-invasive, approach we therefore analyzed volatile organic compounds (VOCs) in the breath and serum headspace of IPAA patients as biomarkers to diagnose and differentiate IPAA disorders. Methods: A prospective study of diagnostic testing was conducted at a tertiary referral center, recruiting subjects under the categories refractory pouchitis (RP) (refractory to or dependency on antibiotics), CD (small bowel mucosal involvement, perianal complications, pouch fistula at least 3 months after ileostomy closure) or normal pouch (NP). Exclusion criteria were surgical complications of IPAA. Clinical and endoscopic sub scores of the pouchitis disease activity score (PDAI) were obtained. The VOC profile was analyzed using selective ion flow tube-mass spectrometry in 1) the breath of subjects following a mouth rinse with water and 2) the headspace of serum samples warmed to 37°C after being nil orally for eight hours. Appropriate statistical tests were used (Table 1). Twenty two distinct VOCs were analyzed. Results: Thirty UC subjects with IPAA (NP (n=7), RP (n=10) or CD (n=13), (mean age 46.8 years, 50% female, mean disease duration 106 months) were included. We observed no difference in any VOCs when comparing NP with RP or RP with CD. Acrylonitrile was the only compound elevated in CD compared to NP (p < 0.05) (Table 1). None of the serum headspace VOCs were different between NP, RP and CD. There was no significant correlation (rho value) between the level of VOCs and clinical or endoscopic sub scores of the PDAI. Ethanol, nonene and trimethylamine showed significant (p < 0.05) correlation between serum and breath VOC levels. Conclusion: Given our previous findings of a vastly distinct breath metabolome in IBD patients naive to surgery compared to healthy controls, this data shows no difference in VOCs of patients with IPAA, irrespective of occurrence of inflammation or not. This suggests that the IBD specific VOC patterns may evolve from the intramural or intraluminal milieu of the colon.Figure 1