In total, 14 healthy controls and 14 IBS patients were recruited. Two participants, one in each group, dropped out due to intolerance to the long period of brain MR scanning, resulting in 13 evaluable subjects in each group (Consort flow diagram see Figures S5 and S6). All patients fulfilled the Rome IV criteria for IBS according to the Rome IV questionnaire. Six IBS patients met criteria for diarrhoea-predominant subtype, three for constipation-predominant subtype and four for mixed subtype. Descriptive data are shown in Table 1.
Higher scores for bloating, cramps and flatulence (group main effect: F 1,685 = 9.38, P = 0.002; F 1,684 = 23.55, P < 0.001; F 1,682 = 8.30, P = 0.004 respectively) were reported by IBS patients compared to HC, with increasing differences over time (group-by-time interaction effect: F = 10,685 = 2.96, P = 0.001; F 10,684 = 7.77, P < 0.001; F 10,682 = 2.30, P = 0.012 respectively) (Figure 3). The group by condition interaction effect was observed for cramps and nausea (F 2,684 = 5.48, P = 0.004; F 2,686 = 3.78, P = 0.023 respectively). Planned contrast analysis revealed the differences between conditions (fructans vs glucose or fructans vs saline) were seen only in IBS, but not in HC for cramps, pain, flatulence and nausea (Table 2). No significant group main effects or group by condition interaction effects were observed for the other GI symptoms (all P > 0.12) (Supplementary Material S4).
GI symptom responses to infusion of fructans, glucose and saline in IBS patients and healthy controls. (A) Bloating; (B) fullness; (C) nausea; (D) cramps; (E) pain; (F) flatulence. GI, gastrointestinal; HC, healthy control; IBS, irritable bowel syndrome
Controlling these analyses for psychological variables (PHQ12, PHQ9, GAD7) did not change the effects of interest, nor were any of the effects of the psychological variables significant (data not shown).
Psychological State: POMS and PANAS
No significant differences were observed for extra-intestinal symptoms (all P > 0.06) (Supplementary Material S5).
Small Bowel Motility. Fructans induced a higher small bowel motility index compared to both glucose and saline (condition main effect: F 2,44 = 15.39, P < 0.001; planned contrasts: fructans vs glucose: t 44 = 5.49, pHolm < 0.001; fructans vs saline: t 44 = 3.46, pHolm = 0.001). IBS exhibited similar motility indices compared to HC (group main effect: F 1,24 = 0.12, P = 0.73; group-by-condition interaction effect: F 2,24 = 0.57, P = 0.57). Planned contrasts confirmed similar differences between conditions in IBS and HC (Table 2, Figure 4A,B, Table S4).
Abdominal physiology responses to infusion of fructans, glucose and saline in IBS patients and healthy controls. (A) small bowel motility 1-hour post-infusion in HC; (B) small bowel motility 1-hour post-infusion in IBS; (C) ascending colon gas (D) ascending colon volume, after infusion of the three solutions, over time. HC, healthy control; IBS, irritable bowel syndrome
Ascending Colon gas and Volume. There were no differences at baseline between IBS and HC (both P > 0.21). Fructans infusion induced a greater increase in ascending colon gas (condition main effect: F 2,36 = 21.40, P < 0.001; planned contrasts: fructans vs glucose: t 36 = 4.96, pHolm < 0.001; fructans vs saline: t 36 = 5.43, pHolm < 0.001) and volume (condition main effect: F 2,36 = 28.47, P < 0.001; planned contrasts: fructans vs glucose: t 36 = 6.82, pHolm < 0.001; fructans vs saline: t 36 = 5.65, pHolm < 0.001). IBS and HC showed similar increases in gas (group main effect: F 1,24 = 0.78, P = 0.39; group-by-condition interaction effect: F 2,36 = 0.03, P = 0.97) and volume (group main effect: F 1,24 = 1.39, P = 0.25; group-by-condition interaction effect: F 2,36 = 0.67, P = 0.52). Planned contrasts confirmed significant differences between fructans and both glucose and saline in IBS and HC (Table 2).
The increase in ascending colon gas in IBS remained stable between 60 and 120 minutes post-infusion, while in HC, gas continued to increase (group-by-time interaction effect, F 1,23 = 5.60, P = 0.027; planned contrasts: IBS: 120 mins vs 60 mins, t 23 = −0.43, pHolm = 0.90; HC: 120 mins vs 60 mins, t 23 = 2.97, pHolm = 0.028). Such effect was not observed for volume (F 1,22 = 0.84, P = 0.37) (Figure 4C,D).
Significant group-by-condition-by-time three-way interaction effects were observed on the blood oxygenation level-dependent (BOLD) signal, relative to pre-infusion baseline.
Comparing fructans and glucose between IBS and HC over time, differential activation patterns were identified in bilateral cerebellum, left and right supramarginal gyrus, bilateral postcentral gyrus, right anterior and midcingulate cortex, left middle and right anterior insula, left rolandic operculum, right putamen and bilateral thalamus (Table 3, Figure 5A).
Comparison of brain responses (change from baseline) between IBS patients and healthy controls over time. (A) Following fructans vs glucose; (B) following fructans vs saline. Voxel-level threshold pFWE-corrected < 0.05, extent threshold k ≥ 20. Colour scale reflects F-values. FWE, family-wise error; HC, healthy control; IBS, irritable bowel syndrome
Furthermore, significant group differences were found following fructans infusion compared to saline over time in bilateral cerebellum, bilateral supramarginal gyrus, right superior temporal gyrus, left superior and right middle frontal gyrus, bilateral anterior cingulate cortex, left midcingulate cortex, bilateral anterior and middle insula and right rolandic operculum. (Table 3, Figure 5B).
Associations Between gut Physiology and GI Symptoms
Due to the limited variability for the differences in GI symptoms between conditions in HC (fructans vs glucose: >48.36% within ±2; fructans vs saline: >39.69% within ±2), associations were analysed in IBS only.
Small Bowel Motility. There was no association between the difference in small bowel motility and the difference in GI symptoms, for fructans compared to both glucose and saline, in IBS patients (P > 0.24 for all main effects).
Ascending Colon gas and Volume. The difference in ascending colon gas between fructans and saline was associated with the respective difference in bloating (main effect of gas, F 1,11 = 10.49, P = 0.008) and cramps (main effect of gas, F 1,11 = 5.77, P = 0.035) (Figure S7). No other associations were seen between ascending colon gas and other GI symptoms (fructans vs saline P > 0.40; fructans vs glucose P > 0.058 for all main effects of gas).
An association between difference in ascending colon volume and cramps was observed (main effect of volume, F 1,11 = 6.19, P = 0.03) when comparing fructans and saline. No other significant associations were seen between differences in ascending colon volume and in GI symptoms (P > 0.10 for all main effects of volume).
Associations Between Brain Activity and GI Symptoms
Due to the same reason as above, the associations between brain activity and GI symptoms were analysed in IBS patients only.
The difference in brain response after fructans vs glucose covaried significantly with GI symptoms in bilateral cerebellum, right anterior and left midcingulate cortex, left inferior and superior and right middle frontal gyrus, bilateral supramarginal gyrus, bilateral postcentral gyrus, right supplementary motor area, bilateral insula, right putamen, right pallidum and bilateral thalamus (Table S5).
The difference in brain response after fructans vs saline covaried significantly with GI symptoms in bilateral cerebellum, right anterior and left midcingulate cortex, left superior and right middle frontal gyrus, right supramarginal gyrus, bilateral postcentral gyrus, bilateral insula, left rolandic operculum, right putamen and bilateral thalamus (Table S6).
Aliment Pharmacol Ther. 2022;55(6):670-682. © 2022 Blackwell Publishing