Psyllium reduces inulin-induced colonic gas production in IBS: MRI and in vitro fermentation studies

Abstract

Objective: Health-promoting dietary fibre including inulin often triggers gastrointestinal symptoms in patients with IBS, limiting their intake. Our aim was to test if coadministering psyllium with inulin would reduce gas production.

Design: A randomised, four-period, four-treatment, placebo-controlled, crossover trial in 19 patients with IBS. Subjects ingested a 500 mL test drink containing either inulin 20 g, psyllium 20 g, inulin 20 g+ psyllium 20 g or dextrose 20 g (placebo). Breath hydrogen was measured every 30 min with MRI scans hourly for 6 hours. Faecal samples from a subset of the patients with IBS were tested using an in vitro fermentation model. Primary endpoint was colonic gas assessed by MRI.

Results: Colonic gas rose steadily from 0 to 6 hours, with inulin causing the greatest rise, median (IQR) AUC(0-360 min) 3145 (848-6502) mL·min. This was significantly reduced with inulin and psyllium coadministration to 618 (62-2345) mL·min (p=0.02), not significantly different from placebo. Colonic volumes AUC(0-360 min) were significantly larger than placebo for both inulin (p=0.002) and inulin and psyllium coadministration (p=0.005). Breath hydrogen rose significantly from 120 min after inulin but not psyllium; coadministration of psyllium with inulin delayed and reduced the maximum increase, AUC(0-360 min) from 7230 (3255-17910) ppm·hour to 1035 (360-4320) ppm·hour, p=0.007.Fermentation in vitro produced more gas with inulin than psyllium. Combining psyllium with inulin did not reduce gas production.

Conclusions: Psyllium reduced inulin-related gas production in patients with IBS but does not directly inhibit fermentation. Whether coadministration with psyllium increases the tolerability of prebiotics in IBS warrants further study.

Trial registration number: NCT03265002.

Keywords: abdominal MRI; colonic fermentation; dietary fibre; irritable bowel syndrome.

Conflict of interest statement

Competing interests: GM has received speaker’s fees from Almirall and Vertex and research funding from Vertex and Sanofi-Aventis Deutschland GmbH. KW has served as a consultant for Danone, has received speaker’s fees from Alpro and Yakult and research funding from Clasado Biosciences, Nestec Ltd, Almond Board of California and the International Nut and Dried Fruit Council. RCS has received speaker’s fees from Alfa Wasserman and research funding from Zespri International Limited and Sanofi-Aventis Deutschland GmbH. MR has served as a consultant for USANA Health Sciences and received speaker’s fees from Alpro, California Walnuts, Associated British Foods and Yakult and research funding from Almond Board of California, the International Nut and Dried Fruit Council and Danone. JMRT has served a consultant for TakeXa, Biocodex and Alfasigma. FJW has received research funding from PepsiCo.

© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY. Published by BMJ.

Figures

Figure 1

Change in MRI colonic gas from fasting values for each test drink (n=19). Data shown are mean±95% CI. (A) Time course over the duration of the study, showing significantly greater gas production for inulin compared with psyllium, dextrose and inulin and psyllium coadministration at 360 min (p=0.0097). (B) Area under the curve (AUC) for individual participants after each test drink (on the x axis). Inulin produced a significantly larger AUC than the other three test drinks. *p

Figure 2

Area under the curve (AUC) change in MRI colonic gas from fasting values for IBS-C versus IBS-D (n=19), demonstrating significantly greater AUCs for IBS-D after both inulin and inulin and psyllium coadministration test drinks. Data shown are mean±95% CI. a, p=0.01. b, p=0.03. IBS-C, constipation-predominant IBS; IBS-D, diarrhoea-predominant IBS.

Figure 3

Colonic volumes after the test drink (n=19). (A) Total colonic volumes after dextrose remained stable but rose significantly after the other three test drinks. Area under the curve for both inulin and inulin and psyllium coadministration were significantly greater than dextrose, p=0.002 and p=0.005, respectively. Data shown are mean±95% CI. (B) Ascending colonic volume 360 min after each test drink (on the x axis), represented by Tukey box and whiskers plot. Inulin and psyllium coadministration significantly increased the volume compared with either inulin or psyllium alone (p=0.003 and p=0.02, respectively).

Figure 4

Breath hydrogen concentration (ppm) at fasting and every 30 min after the test drink (n=19). Breath hydrogen rose steadily after 30 min with the inulin test drink but area under the curve analysis demonstrated a significantly reduced rise when inulin was coadministered with psyllium, p=0.0065. Both psyllium and dextrose produced significantly less breath hydrogen than inulin alone, both p

Figure 5

Forty-eight hours of in vitro gas production of substrates inoculated using stool from patients who participated in the human MRI study (n=8). (A) Median cumulative gas production showing an early rapid fermentation of dextrose followed by the inulin and psyllium combination. (B) Area under the curve (AUC) of in vitro gas production for each test drink (on the x axis). Data shown are mean±95% CI. (A) Dextrose AUC is significantly greater than inulin (p=0.0008), psyllium (p=0.0001) and calculated inulin and psyllium (p=0.001). (B) Inulin and psyllium combination AUC is significantly greater than psyllium (p=0.002).

Figure 6

Correlations between in vitro area under the curve (AUC) gas production (mL·hour) and in vivo AUC colonic gas (mL·min) as assessed by MRI for (A) inulin r2=0.58, p=0.03, (B) psyllium r2=0.14, p=0.35 and (C) inulin and psyllium in combination r2=0.003, p=0.89.