Introduction

Functional dyspepsia (FD) is associated with duodenal eosinophilia and gut microbiota dysbiosis. Proton pump inhibitors (PPIs), as a first-line therapy, can suppress duodenal eosinophilic inflammation to ameliorate symptoms, yet the symptom relief rate is low. Rifaximin (RIF), an antibiotic known to modulate the gut microbiota, may offer an alternative therapeutic avenue. It is hypothesized that combining RIF with PPIs could more effectively alleviate FD symptoms and duodenal inflammation by targeting the gut microbiota.

Aims & Methods

A single-center, open-label, randomized controlled clinical trial was conducted on subjects who met the inclusion and exclusion criteria (Trial registration: ChiCTR2300069313). Ninety-six patients were randomly assigned in a 1:1 ratio to either the PPI+RIF group or the PPI group. The follow-up period extended to 4 weeks post-treatment cessation. The primary endpoint was the symptom relief rate, while secondary endpoints included symptom response rate, duodenal eosinophil counts, and changes in the microbiota. The study hypothesis was that the clinical efficacy of PPI+RIF would be superior to that of PPI monotherapy.

Results

	PPI+RIF	PPI	Z value	P value	PPI+RIF	PPI	Z value	P value
Abbreviations: PPI+RIF: Proton Pump Inhibitor plus Rifaximin; PPI: Proton Pump Inhibitor; ITT: the intention-to-treat analysis;PP: the per-protocol analysis.
Symptom relief rate	ITT				PP
Week 2	54.2%（26/48）	54.2%（26/48）	0.000	0.500	53.7%（22/41）	53.3%（24/45）	0.030	0.488
Week 4	72.9%（35/48）	58.3%（28/48）	1.522	0.064	76.7%（33/43）	56.8%（25/44）	2.020	0.022
4 weeks post-treatment	77.1%（37/48）	47.9%（23/48）	3.095	0.001	78.6%（33/42）	48.9%（22/45）	3.036	0.001
Symptom response rate	ITT				PP
Week 2	70.8%（34/48）	72.9%（35/48）	-0.227	0.590	68.3%（28/41）	73.3%（33/45）	-0.514	0.696
Week 4	81.3%（39/48）	77.1%（37/48）	-0.503	0.307	83.7%（36/43）	75.0%（33/44）	1.012	0.156
4 weeks post-treatment	85.4%（41/48）	68.8%（33/48）	1.982	0.024	85.7%（36/42）	71.1%（32/45）	1.688	0.046

A total of 96 patients were enrolled in the randomized controlled study. There were no statistically significant differences in baseline demographic data and clinical characteristics between the two groups. According to the intention-to-treat (ITT) analysis, the PPI+RIF group showed significantly higher rates of symptom relief (P = 0.001) and symptom response (P = 0.024) at 4 weeks post-treatment. These findings were supported by the per-protocol (PP) analysis, which revealed a superior symptom relief rate for the combination group at both 4 weeks of treatment (P = 0.022) and 4 weeks post-treatment (P = 0.001), as well as a superior symptom response rate at 4 weeks post-treatment (P = 0.046). Combination therapy also led to a greater reduction in duodenal eosinophil infiltration (P = 0.036) and significantly decreased peak hydrogen sulfide (H₂S) levels (P < 0.001), indicating a reduction in small intestinal bacterial load. Oral microbiota analysis revealed that β-diversity of the oral flora increased following the PPI+RIF intervention, with an increased relative abundance of Enterococcus (P = 0.004). Conversely, the PPI intervention led to the enrichment of the potentially pathogenic bacterium Catonella (P = 0.006). Notably, treatment responders in the PPI+RIF group showed a specific enrichment of the Eubacterium_yurii_group (P = 0.037), an alteration associated with the amelioration of duodenal eosinophilic inflammation and the enhancement of pathways related to preventing bacterial invasion of the epithelium.

Conclusion

In patients with FD, combining a PPI with Rifaximin is superior to PPI monotherapy, improving both symptoms and duodenal eosinophil infiltration. This effect is linked to a reduction in H₂S-producing bacteria and a beneficial shift in the oral microbiota. The oral abundance of Eubacterium_yurii_group may serve as a biomarker for treatment response.

Disclosure

The authors declare that they have no conflicts of interest.

Introduction

Disorders of gut-brain interaction (DGBIs), such as functional dyspepsia and irritable bowel syndrome, are chronic gastrointestinal disorders with recurrent and bothersome abdominal symptoms. Although incompletely understood, immune activation, particularly involving eosinophils and mast cells, and visceral hypersensitivity have been implicated in their pathophysiology.

Aims & Methods

Using the normoglycemic Biobreeding diabetes-prone (BBDP-N) rat, we aim to 1) characterise the immune activation in jejunal and colonic mucosa and to 2) study the effect of budesonide, a local anti-inflammatory corticosteroid, on gut physiology and sensitivity.
90-day-old BBDP-N and Biobreeding diabetes-resistant (control) rats underwent a 28-day treatment with budesonide (0.5 mg/kg/day) or vehicle. Eosinophils and mast cells were identified by staining jejunal and colonic sections with Chromotrope 2R and anti-mast cell protease 2 antibody, respectively. To assess their activity, eosinophil peroxidase (EPO) and β-hexosaminidase levels were measured. Jejunal and colonic segments were mounted in modified Ussing chambers, and permeability was assessed by the transmucosal passage of a fluorescently labelled 20 and 4kDa dextran molecule (FITC-Dx4/20), and the transepithelial electrical resistance (TEER). Sensitivity to isobaric colorectal distension was assessed(1): the ∆ area under the curve (AUC) of the visceromotor response (VMR) to wall tension was calculated at baseline and after treatment. Depending on the normality of the data, an unpaired t-test or Mann-Whitney test was performed.

Results

Comparing vehicle-treated groups, the BBDP-N rats had higher numbers of jejunal eosinophils (1366.0±201.3 vs. 561.7±55.3 eosinophils/mm²; p<0.01) and mast cells (1151.0±136.7 vs. 549.1±44.8 mast cells/mm²; p<0.001), whereas only colonic eosinophil numbers were higher (222.7±40.4 vs. 128.9±14.5 eosinophils/mm²; p=0.03). Additionally, EPO levels were higher in the BBDP-N rats in both regions (jejunum: 59.4±9.8 vs. 4.4±3.0 µg/g tissue; p<0.01 & colon: 82.0±23.6 vs. 1.6±1.1 µg/g tissue; p<0.01), with no difference in mast cell activation. Following budesonide treatment, eosinophil (jejunum, BBDP-N: 167.3±53.6 vs. 1366.0±201.3 eosinophils/mm²; p<0.001 & control: 220.1±43.7 vs. 561.7±55.3 eosinophils/mm²; p<0.01, colon: BBDP-N: 58.2±11.9 vs. 222.7±40.4 eosinophils/mm²; p<0.01 & control: 78.1±14.4 vs. 128.9±14.5 eosinophils/mm²; p=0.01) and mast cell (jejunum: BBDP-N: 418.8±51.0 vs. 1151.0±136.7 mast cells/mm²; p<0.001 & control: 305.4±67.6 vs. 549.1±44.8 mast cells/mm²; p=0.01, colon: BBDP-N: 146.5±19.1 vs. 262.6±37.5 mast cells/mm²; p<0.01 & control: 137.4±17.8 vs. 244.9±29.4 mast cells/mm²; p<0.01) levels decreased in both groups in both regions. Colonic eosinophil activation was also reduced in BBDP-N rats (21.5±9.7 vs. 82.0±23.6 µg/g tissue; p=0.05), with no effect on mast cell activation. Jejunal permeability, assessed by 20kDa passage was increased in vehicle-treated BBDP-N rats (258.0±51.5 vs. 132.0±25.1 pmol/cm²; p=0.03). Budesonide treatment significantly increased colonic TEER in BBDP-N rats (73.2±4.3 vs. 38.3±4.1 Ω.cm²; p<0.001). Colonic sensitivity to distention was significantly reduced in BBDP-N following the treatment (-20.1±17.2 vs. 207.0±85.5 mV/s x mmHg.cm; p=0.02).

Conclusion

In conclusion, our work shows the importance of eosinophil and mast cell predominant immune activation in visceral hypersensitivity in a well-characterised preclinical model of DGBI. Anti-inflammatory treatments should be re-evaluated in DGBI with intestinal eosinophil and mast cell infiltration.

References

1: Meleine M et al. Neurogastroenterology and motility, 2019.