Exploring small intestinal bacterial overgrowth in functional upper gastrointestinal disorder: A comprehensive case-control study

Abstract

BACKGROUND

Small intestinal bacterial overgrowth (SIBO) is suspected and excluded frequently in functional gastrointestinal (GI) disorders. Children presenting with various esophago-gastro-duodenal (upper GI) symptoms are rarely subjected to investigations for SIBO.

AIM

To estimate the frequency of SIBO in children having functional upper GI symptoms (as cases) and to compare the result of the SIBO status to that of the controls.

METHODS

Children aged 6 to 18 who presented with upper GI symptoms were selected for the study. All children were subjected to upper GI endoscopy before being advised of any proton pump inhibitors (PPIs). Children with normal endoscopy were assigned as cases, and children having any endoscopic lesion were designated as controls. Both groups were subjected to a glucose-hydrogen breath test by Bedfont Gastrolyser.

RESULTS

A total of 129 consecutive children who were naive to PPIs and had normal baseline investigations were included in the study. Among them, 67 patients had endoscopic lesions and served as the control group, with six cases being excluded due to the presence of Helicobacter pylori in gastric biopsies. Sixty-two children with normal endoscopy results formed the case group. In the case group, 35 children (59%) tested positive for hydrogen breath tests, compared to 13 children (21%) in the control group. The calculated odds ratio was 5.38 (95% confidence interval: 2.41-12.0), which was statistically significant. Further analysis of symptoms revealed that nausea, halitosis, foul-smelling eructation, and epigastric fullness were positive predictors of SIBO.

CONCLUSION

It is worthwhile to investigate and treat SIBO in all children presenting with upper GI symptoms that are not explained by endoscopy findings.

Key Words: Small intestinal; Bacterial overgrowth; Hydrogen breath test; Functional gastrointestinal; Proton pump inhibitor

Core Tip: Small intestinal bacterial overgrowth (SIBO) is defined as an increased number of non-pathogenic bacteria over 10⁵ organisms/mL of small intestinal content. The most common predisposing factors include, among others, gut motility disorders and chronic use of proton pump inhibitors. SIBO is suspected and excluded frequently in functional gastrointestinal (GI) disorders like functional abdominal pain, irritable bowel syndrome, functional dyspepsia, or functional defecation disorder. However, the extent of SIBO in functional GI disorders, which mainly manifest with upper gastrointestinal symptoms such as nausea, vomiting, epigastric pain, belching, dyspepsia, etc, has not been documented earlier. This study uniquely divides the proton pump inhibitor naive children into two groups: One with demonstrable endoscopic lesions (control) and the other without the endoscopic lesions (cases) and compares the prevalence of SIBO between these two groups instead of comparing the cases with healthy children without any symptoms. The result documented a significantly increased number of SIBO in the children with normal endoscopic evaluation. Henceforth, this study documents a need to evaluate the children with functional upper GI symptoms for SIBO.

INTRODUCTION

Small intestinal bacterial overgrowth (SIBO) is defined as an increased number of nonpathogenic bacteria over 10⁵ organisms/mL of small intestinal content. The most common predisposing factors include, among others, gut motility disorders and chronic use of proton pump inhibitors (PPIs). Glucose hydrogen (H₂) breath test is being used in various functional gastrointestinal (GI) disorders for estimation of SIBO. SIBO is suspected and excluded frequently in functional GI disorders (FGID) like functional abdominal pain, irritable bowel syndrome, functional dyspepsia or functional defecation disorder. The H₂ and methane breath tests have gained popularity due to their widespread availability in healthcare settings, affordability, practicality, and non-invasive nature, making them particularly well-suited for pediatric use. Moreover, the result of the test can be interpreted on the same day. As per the ROME IV criteria, a few Paediatric FGIDs like functional nausea and functional vomiting, rumination disorder, aerophagia and functional dyspepsia often have mainly upper gastrointestinal (upper GI) symptoms. Children presenting with various upper GI symptoms, are rather subjected to upper GI endoscopy and ultrasonography of the abdomen than to investigations for SIBO. This case-control study was conducted with the following objectives: (1) To estimate the frequency of SIBO in children having functional upper GI symptoms (as cases); and (2) To compare the result of the SIBO status to that of the controls.

MATERIALS AND METHODS

This prospective study was undertaken in a study in tertiary paediatric GI center in Kolkata from July 2022 to July 2023.

Study design

This is a prospective case-control study using children with functional upper GI symptoms as cases and children having endoscopic lesions (organic upper GI disorder) as controls.

Participants

Children between the ages of 6 and 18 years who presented with upper GI symptoms such as nausea, regurgitation, heartburn, halitosis, belching, foul-smelling eructation, epigastric pain, epigastric fullness, and vomiting, and who had not received PPIs for at least two months prior to their visit, were selected for the study. They might or might not have fulfilled the Rome IV criteria of FGIDs. All children were investigated to rule out celiac disease, diabetes, thyroid disorder, and chronic kidney disease. They also underwent ultrasonography of the abdomen to check that it was normal. Written consent and/or assent were obtained from all selected participants.

Inclusion criteria

Children presenting with upper GI symptoms mentioned above who were PPI naive.

Exclusion criteria

Children were excluded if they had abnormal abdominal ultrasounds, diabetes, celiac disease, thyroid function abnormalities, or gastrointestinal issues such as constipation, diarrhea, or overlapping conditions like irritable bowel syndrome, functional abdominal pain, and functional defecation disorders. Additionally, those who had taken PPIs within the past two months were also excluded.

Procedure

All selected children underwent upper GI endoscopy before any PPIs were prescribed. Those with normal endoscopy results were categorized as cases, while children with endoscopic findings such as ulcers, erosions, nodularity, erythema, mosaic mucosal patterns, or white lesions were designated as controls. Both groups then underwent a H₂ breath test using the Bedfont Gastrolyser with glucose as the substrate (up to 2 gm/kg, with a maximum of 100 gm), following all necessary precautions. After a basal reading children consumed the glucose solution. Thereafter readings were taken at 15-minute intervals up to 8 readings. A persistent increase in the values more than 12 ppm from basal value was regarded as positive[1]. Any symptom during the test was also noted. All SIBO-positive patients were treated with 3 cycles of sequential rifaximin and a probiotic. Each cycle consisted of Rifaximin at a dose of 10mg/kg/dose twice/day for 14 days followed by a probiotic Lactobacillus reuteri DSM 17938, one billion unit/day for 14 days for 3 cycles.

Sample size calculation

To ensure a standard power of 80% and an alpha level of 0.05, the sample size for each group (case and control) was initially calculated to be 60. However, considering potential variability in patient behavior and participation rates, the target was adjusted to 70 participants in each group.

Statistical analysis

Statistical calculations were obtained using SPSS software for Windows version 23.0 (SPSS Chicago, IL, United States). The odds ratio (OR) was calculated and Fisher’s exact test was used to find out the association between the categorical variables. A P value < 0.01 was accepted as significant.

RESULTS

A total of 129 consecutive children who were naive to PPIs and had normal baseline investigations were included in the study. Out of them, 79 were male (61%). The median age of the children was 10 years (interquartile range: 8-12.15). Different presenting symptoms of these subjects were documented in Table 1. The median duration of symptoms was 49 days (interquartile range: 43-54). Only one child fulfilled the ROME IV criteria of functional dyspepsia and the other 3 had functional nausea disorder as per the same criteria with a duration of more than 2 months. Post-endoscopic evaluation sixty-seven subjects had visible endoscopic lesions, but six were excluded due to Helicobacter pylori detected in their gastric biopsies, resulting in 61 subjects in the control group. Different endoscopic lesions included esophagitis LA grade A, B, and C; gastric erosions, gastric ulcers, duodenal ulcers, duodenal erosions, gastric nodularity, and mosaic pattern of the gastric mucosa. Sixty-two had normal endoscopies (cases). 3 did not participate in the breath test and were excluded. In the case group out of 59 children, 35 (59%) children had positive H₂ breath test. Out of these 35 children, 18 (51%) had different symptoms during the test tenure (nausea, belching, regurgitation, vomiting). In the control group amongst 61 children, 13 (21%) showed positive H₂ breath test and none had any symptom within the test duration. The calculated OR was 5.38 [95% confidence interval (CI): 2.41-12.02], indicating a statistically significant association (Table 2). Further analysis using the Fisher exact test revealed that nausea, halitosis, foul-smelling eructation, and epigastric fullness were positive predictors of SIBO. Conversely, epigastric pain and heartburn were found to be negatively associated with SIBO (Table 1).

Table 1 Different presenting symptoms and association with small intestinal bacterial overgrowth.

Symptoms	Case group (n = 59)	Control group (n = 61)	SIBO (n = 48)	Non-SIBO (n = 72)	P value
Nausea	49	9	39	19	< 0.00001
Regurgitation	20	27	25	22	< 0.0224
Heartburn	3	29	5	27	< 0.0013
Halitosis	36	25	31	20	< 0.00001
Belching	29	23	22	30	0.7087
Foul-smelling eructation	41	10	42	9	< 0.00001
Epigastric pain	23	56	14	65	< 0.00001
Epigastric fullness	45	28	38	35	0.0011
Vomiting	11	13	8	16	0.4946

SIBO: Small intestinal bacterial overgrowth.

Table 2 Hydrogen breath test in case and control groups, n (%).

Group	Positive H2 BT	Negative H2 BT	OR (95%CI)
Case group (n = 59)	35 (59)	24 (41)	5.38 (2.41-12.02)
Control group (n = 61)	13 (21)	48 (79)

H₂: Hydrogen; BT: Breath test; OR: Odds ratio; CI: Confidence interval.

DISCUSSION

This prospective case-control study found SIBO in the context of functional upper GI symptoms for the first time in Indian children. SIBO (by positive H₂ breath test) was found in 59% of the cases compared to 21% of the controls, with an OR of 5.38 (95%CI: 2.41-12.02). This indicated a significant association of SIBO in children with functional upper GI symptoms having normal endoscopic evaluation compared to the control group with similar symptoms but demonstrating some endoscopic lesions.

H₂ breath tests are a non-invasive diagnostic tool commonly used in the evaluation of functional gastrointestinal disorders. The H₂ breath test has a lower sensitivity but good specificity compared to the duodenal aspiration for detecting SIBO in endoscopy-negative adults[2]. The prevalence of SIBO varies significantly, ranging from approximately 9% in children using PPIs[3] to as high as 90% in those experiencing stunted growth[4] and chronic abdominal pain[5]. The discrepancies in the reported prevalence rates result from the fact that SIBO syndrome may be asymptomatic or manifest with symptoms (such as nausea, dyspepsia and epigastric fullness) that may be thought to arise from other causes. Moreover, the clinical presentation of SIBO is a common epiphenomenon accompanying a number of systemic diseases (such as diabetes) or diseases of other organ systems (such as chronic kidney disease) and is not usually considered in the differential diagnosis. However, the relation of SIBO with exclusive functional upper GI symptoms in children is not precisely known, and literature regarding this is extremely limited. The pathophysiological mechanism, though not very well described, may involve the associated impairment of GI motility, which might be the causative as well as the resultant phenomenon of SIBO.

Korterink et al[6] found SIBO using a H₂ breath test with glucose in 14.3% of children and pointed out belching and loss of appetite as potential predictors. Collins et al[5] utilized H₂/methane breath testing to diagnose SIBO in a cohort of children suffering from chronic abdominal pain, including conditions such as functional dyspepsia, irritable bowel syndrome, and functional abdominal pain. Their findings revealed a positive H₂ breath test in 91% of the cases compared to 35% of healthy controls. Interestingly, no significant differences were observed in gastrointestinal symptoms such as bloating, gas, incomplete evacuation, constipation, diarrhea, mucous in stool, or straining between subjects with and without SIBO. However, it is noteworthy that the comparison was skewed due to the unequal distribution of patients between the two groups, with 68 SIBO-positive patients compared to only 7 SIBO-negative patients. Scarpellini et al[7] evaluated the prevalence of SIBO in children with irritable bowel syndrome using the lactulose H₂/methane breath test. Their findings revealed a significantly higher prevalence of SIBO among children with IBS (65%) compared to healthy controls (7%). Interestingly, no notable differences were observed in gastrointestinal symptoms such as bloating, gas, incomplete evacuation, constipation, diarrhea, mucus in stool, or straining between SIBO-positive and SIBO-negative subjects. However, these studies collectively highlight that SIBO is a frequently underlying condition in FGIDs. Despite this, there remains a lack of research specifically addressing the prevalence and characteristics of SIBO in children with isolated upper GI functional disorders. Our study sought to address this knowledge gap.

In the context of this study, it can be questioned that as all children were subjected to upper GI endoscopic evaluation, why not quantitative culture of duodenal/jejunal aspirate used as a diagnostic tool for SIBO, which is considered the gold standard? The answer is that though this was considered in the initial plan, the expenditure for Bactec culture exceeded the total cost of the breath test (as it was available free of cost) and the results of the breath test were obtainable immediately compared to the time consumed to get hold of the culture-report. Moreover, the air is used for endoscopy at our center, making the culture of anaerobic bacteria difficult[8]. Therefore, children were subjected to breath tests.

Sieczkowska et al[9] conducted a unique study evaluating the prevalence of SIBO in 40 children treated with PPIs for three months who continued to experience persistent symptoms. Their findings revealed that 22.5% of these children developed SIBO. Compared to those without SIBO, children with SIBO reported a higher frequency of symptoms such as abdominal pain, bloating, eructation, and flatulence. Based on these results, the authors concluded that children with persistent gastrointestinal symptoms following PPI therapy should be evaluated for SIBO rather than having their PPI treatment empirically extended. This is similar to the result of our study depicting a 59% positive H₂ breath test in children with upper GI symptoms but without demonstrable endoscopic lesions.

PPI use is theoretically associated with more risk of developing SIBO; Cares et al[3] documented SIBO in 8.9% of children receiving PPI vs 3.7% in children not consuming PPI. However, this was not statistically significant in that study. However, the present study had taken care of this fact by excluding patients already receiving PPI within 2 months of the study.

This study also identified nausea, foul-smelling eructation, halitosis, and epigastric fullness as positive predictors of SIBO. Though Korterink et al[6] found belching as a potential associator in our study, belching as a pointer was not statistically significant (Table 1).

An antibiotic is usually started empirically in cases of suspected or proven SIBO. In the adult cohorts mostly rifaximin was used. A recent systematic review and meta-analysis on rifaximin therapy for SIBO compiled data from over 1300 patients[10]. The study demonstrated a dose-dependent response in eradication rates, indicating that the effectiveness of treatment increased with the dosage. This meta-analysis included only one pediatric study, which analyzed the effect of rifaximin and documented a 64% response rate in the normalization of the breath test at a dose of 600 mg/day for 1 week[11]. This study used a longer duration of rifaximin alternating with probiotics. Probiotics have been explored as potential therapeutic agents in managing SIBO. Although it might seem counterintuitive to add more bacteria to an overgrowth condition, probiotics have shown promise in alleviating symptoms and reducing bacterial burden in SIBO patients[12]. Although the position paper from the ESPGHAN special interest group does not recommend any probiotics for the treatment of SIBO[13], a recent meta-analysis evaluating the efficacy of this agent revealed that the application of probiotics resulted in vastly reduced H₂ and increased rates of decontamination (when compared to placebo). While probiotics have shown some promise in managing SIBO, the most effective outcomes have been observed when they are used in combination with rifaximin or minocycline[14]. Although the current study did not aim to assess the decontamination rate, this combined approach appears to offer better benefits.

One of the limitations of this study was its relatively small sample size. The sample size could not be large as PPI naive subjects were difficult to find in the present era. The non-estimation of methane along with H₂ was another limitation that might have underestimated the fraction of children with SIBO.

CONCLUSION

Though the gold standard for diagnosing SIBO is a culture of jejunal aspirate, the H₂ breath test is more popular, being noninvasive, widely available, and easy to interpret. Children presenting with functional upper GI symptoms demonstrated a significantly higher positivity rate on the glucose H₂ breath test, indicating a strong likelihood of SIBO, compared to those without such symptoms. This finding suggests that identifying and treating SIBO may be beneficial for all children with endoscopy-negative upper GI symptoms.

ACKNOWLEDGEMENTS

We acknowledge the contribution of our technical support team members, Mr. Jha A, Mr. Dhungel D and Ms. Debnath S for their constant support during the study.