Colon hypersensitivity to distension, rather than excessive gas production, produces carbohydrate-related symptoms in individuals with irritable bowel syndrome

Background & Aims: Poorly digested, fermentable carbohydrates may induce symptoms of irritable bowel syndrome (IBS), via unclear mechanisms. We performed a randomized trial with magnetic resonance imaging (MRI) analysis to investigate correlations between symptoms and changes in small and large bowel contents following oral challenge. Methods: We performed a 3-period crossover study of 29 adult patients with IBS (based on Rome III criteria, with symptoms of abdominal pain or discomfort for at least 2 days/week) and reported bloating. In parallel we performed the same study of 29 healthy individuals (controls). Studies were performed in the United Kingdom from January 2013 through February 2015. On 3 separate occasions (at least 7 days apart), subjects were given a 500 ml drink containing 40 g of carbohydrate (glucose in the first period, fructose in the second, and inulin in the third, in a random order). Levels of breath hydrogen were measured and intestinal content was assessed by MRI before and at various time points after consumption of each drink. Symptoms were determined based on subjects’ responses to the Hospital Anxiety and Depression Scale questionnaire and the Patient Health Questionnaire-15. The primary endpoint was whether participants had a clinically important symptom response during the 300 minutes following consumption of the drink. Results: More patients with IBS reached the pre-defined symptom threshold after intake of inulin (13/29) or fructose (11/29) than glucose (6/29). Symptoms peaked sooner after intake of fructose than inulin. Fructose increased small bowel water content in both patients and controls whereas inulin increased colonic volume and gas in both. Fructose and inulin increased breath hydrogen levels in both groups, compared to glucose; fructose produced an earlier increase than inulin. Controls had lower symptom scores during the period after drink consumption than patients with IBS, despite similar MRI parameters and breath hydrogen responses. In patients who reached the symptom threshold after inulin intake, peak symptom intensity correlated with peak colonic gas (r = 0.57; P