Herbal therapy is equivalent to rifaximin for the treatment of small intestinal bacterial overgrowth.

OBJECTIVE: Patients with small intestine bacterial overgrowth (SIBO) have chronic intestinal and extraintestinal symptomatology which adversely affects their quality of life. Present treatment of SIBO is limited to oral antibiotics with variable success. A growing number of patients are interested in using complementary and alternative therapies for their gastrointestinal health. The objective was to determine the remission rate of SIBO using either the antibiotic rifaximin or herbals in a tertiary care referral gastroenterology practice.
DESIGN: One hundred and four patients who tested positive for newly diagnosed SIBO by lactulose breath testing (LBT) were offered either rifaximin 1200 mg daily vs herbal therapy for 4 weeks with repeat LBT post-treatment.
RESULTS: Three hundred ninety-six patients underwent LBT for suspected SIBO, of which 251 (63.4%) were positive 165 underwent treatment and 104 had a follow-up LBT. Of the 37 patients who received herbal therapy, 17 (46%) had a negative follow-up LBT compared to 23/67 (34%) of rifaximin users (P=.24). The odds ratio of having a negative LBT after taking herbal therapy as compared to rifaximin was 1.85 (CI=0.77-4.41, P=.17) once adjusted for age, gender, SIBO risk factors and IBS status. Fourteen of the 44 (31.8%) rifaximin non-responders were offered herbal rescue therapy, with 8 of the 14 (57.1%) having a negative LBT after completing the rescue herbal therapy, while 10 non-responders were offered triple antibiotics with 6 responding (60%, P=.89). Adverse effects were reported among the rifaximin treated arm including 1 case of anaphylaxis, 2 cases of hives, 2 cases of diarrhea and 1 case of Clostridium difficile. Only one case of diarrhea was reported in the herbal therapy arm, which did not reach statistical significance (P=.22).
CONCLUSION: SIBO is widely prevalent in a tertiary referral gastroenterology practice. Herbal therapies are at least as effective as rifaximin for resolution of SIBO by LBT. Herbals also appear to be as effective as triple antibiotic therapy for SIBO rescue therapy for rifaximin non-responders. Further, prospective studies are needed to validate these findings and explore additional alternative therapies in patients with refractory SIBO.

INTRODUCTION

Small intestinal bacterial overgrowth (SIBO) is defined as an increase in the concentration of bacteria of more than 100 000 colony-forming units per mL of proximal jejunal fluid. Although the overall prevalence of SIBO in the general public is unknown, a metaanalysis has shown that the prevalence of SIBO is approximately 56% among patients with irritable bowel syndrome (IBS). At present, the prevalence of SIBO in the community or in a tertiary-care gastroenterology referral practice remains unknown. SIBO is becoming an increasingly significant problem in clinical practice as its manifestations can be protean and range from a full-blown enteropathy causing profound malabsorption and malnutrition simulating celiac disease to mild symptoms that overlap with IBS. There are several factors involved in the pathogenesis of SIBO that have in common a disruption in the natural protective antibacterial mechanisms of the gut as shown in Tables 1–3. These factors, alone or in combination, can predispose the gut toward dysbiosis and favor the growth and colonization of aerobes and anaerobes in the proximal jejunum and may provide barrier(s) from achieving durable remission of symptoms following treatment. Currently, lactulose breath testing (LBT) is the most commonly used procedure for diagnosing SIBO., LBT is useful as an indirect testing method for SIBO, given that the human gastrointestinal (GI) tract does not absorb or use lactulose, which is metabolized by bacteria to produce hydrogen and methane gas. These gases are then measured by gas chromatography. Once suspected SIBO is diagnosed, it should be treated to prevent its myriad of adverse consequences.–

Table 1

Conditions That Predispose Toward the Development of Small Intestine Bacterial Overgrowth–

Achlorhydria (surgical, iatrogenic, autoimmune)

Motor abnormalities

Scleroderma

Intestinal pseudo-obstruction

Diabetic enteropathy

Vagotomy

Abnormal communication between colon and small bowel

Fistulas between colon and small bowel

Resection of ileocecal valve

Structural abnormalities

Systemic and intestinal immune deficiency states

Surgical loops (Billroth II, entero-entero anastomosis, Rou-en-Y)

Duodenal or jejunal diverticula

Partial obstruction of small bowel (stricture, adhesions, tumors)

Large small Intestine diverticulosis

Systemic diseases (celiac disease, cirrhosis, pancreatic exocrine insufficiency, non-alcoholic fatty liver disease)

Alcoholism

Table 2

Protective Factors That Protect Against the Development of Small Intestine Bacterial Overgrowth,,

• Gastric acid

• Pancreatic enzymes

• Bile acids

• Cholecystectomy

• Motility

• Migrating motor complex

• Biofilm

• Secretory immunoglobulin A

Table 3

Extrinsic Factors That Alter the Gut Microbiome and May Influence the Development of Small Intestine Bacterial Overgrowth

FODMAPsa (fructose, lactose, galactans, fructans, sugar alcohols)

Proton pump inhibitors

Anti-motility agents

Fiber

Prebiotics

Probiotics

Antibiotics

FODMAPs is an acronym for a group of highly fermentable foods (Fermentable Oligo-, Di-, Monosaccharides and Polyols).

What is the current knowledge?

Small intestinal bacterial overgrowth (SIBO) is widely prevalent among patients with the irritable bowel syndrome (IBS) and is associated with a worsening of symptoms and quality of life.

Current treatment of SIBO is presently limited to antibiotics with a variable response rate and none have been approved by the US Federal Drug Administration (FDA).

SIBO is often treated with a course of rifaximin 1200 mg/ day for 10-14 days, which is expensive and not routinely covered by many commercial health plans in the United States.

The majority of IBS patients use complementary and alternative therapies to control symptoms and are often looking for more “natural” options to treat their disease.

What is new here?

The high prevalence rate for SIBO of 64% in a tertiary care referral gastroenterology practice.

The response rate for normalizing breath hydrogen testing in patients with SIBO was 46% for herbal therapies vs 34% for Rifaximin.

Herbal therapy may be an effective treatment for patients with SIBO.

Patients with SIBO refractory to rifaximin can be given the choice of herbal therapy as rescue therapy.

The current state-of-the-art treatment of SIBO is the provision of a short course (10-14 days) of antibiotics (Table 4). Thus far, rifaximin, a rifampin derivative, is the antibiotic that has been most widely recognized and published for its use in the treatment of SIBO. Additionally, SIBO has been implicated as a major contributing pathophysiological factor in IBS. Over the past decade, patients with IBS are increasingly turning to complementary and alternative medicine (CAM) options for symptoms relief and disease management. Herbal remedies and nutraceutical supplements continue to dominate as the most common forms of CAM used for IBS patients. A number of herbs have a long tradition of antimicrobial activity, thus we hypothesized that the use of plant extracts possessing antimicrobial activity would be as effective as antibiotic therapy for patients with an abnormal LBT and the diagnosis of SIBO.

Table 4

Antibiotic Regimens Used for Small Intestine Bacterial Overgrowth

Agent	Dose	Frequency
Amoxicillin-clavulanate	500 mg PO	3 times/day
Cephalexin	250 mg PO	4 times/day
Chloramphenicol	250 mg PO	4 times/day
Ciprofloxacin	500 mg PO	twice daily
Doxycycline	100 mg PO	twice daily
Metronidazole	250 mg PO	3 times/day
Neomycin	500 mg PO	twice daily
Norfloxacin	400 mg PO	twice daily
Rifaximin	400 mg PO	3 times/day
Tetracycline	250 mg PO	4 times/day
Trimethoprim- sulfamethoxazole	1 double-strength tablet PO	twice daily

Abbreviation: PO, per os (by mouth).

METHODS

Study Participants

The medical records at a single tertiary-care referral center from October 2006 to November 2010 were reviewed as part of an institutional IRB-approved protocol. Study participants were included in the analysis if they had: (1) symptoms suggestive of SIBO, (2) an initial positive LBT, (3) completed a prescribed regimen of either rifaximin or herbal therapy, and (4) received a post-treatment LBT. Symptoms considered to be suggestive of SIBO included otherwise unexplained abdominal discomfort, cramping, bloating, flatulence, eructation, diarrhea, worsening of symptoms after meal ingestion, and low serum B12. Criteria considered to be risk factors for SIBO included prior history of gastric bypass surgery, known gastrointestinal motility disorder, collagen vascular disease, IBS (association), pancreatic insufficiency, and chronic proton-pump inhibitor (PPI) use. Exclusion criteria included those patients aged <18 or >85 years, use of antibiotics within 3 months, and failure to comply with regimen or to have a follow-up LBT. A subset of patients who (1) completed the rifaximin regimen, (2) had persistently positive post-treatment LBT, (3) subsequently completed a prescribed herbal or triple antibiotic regimen, and (4) had a post-herbal therapy LBT were defined as “rifaximin non-responders.”

Lactulose Breath Testing Protocol

Study participants followed a standard preparation protocol prior to the procedure. No laxatives were to be taken for at least a week prior to the test. One day prior, test subjects were advised to avoid high-fiber foods, butter, margarine, and sodas and asked to fast for 12 hours before the test, consuming no food except water. Subjects were advised not to smoke, sleep, or exercise vigorously up to 30 minutes before or at any time during the test.

A QuinTron Gas Microanalyzer (QuinTron Instrument Company, Inc, Milwaukee, Wisconsin) was used to detect breath hydrogen in samples. Baseline breath hydrogen was obtained on all subjects prior to the ingestion of 10 gm of lactulose in 30 cc of water. Serial breath hydrogen analysis was obtained every 20 minutes after ingestion of lactulose up to a maximum of 3 hours. The test was considered positive if it showed one or more of the following: (1) a baseline breath concentration of >10 parts per million (ppm) for hydrogen or >7 ppm for methane only if patients were compliant with their preparation or (2) an increase within 90 minutes (small intestine) that was followed by a larger peak (colonic), indicative of a positive study (with a decrease of at least 5 ppm following the first peak). The first increase had to have one of the following to be considered positive: (1) an increase of at least 12 ppm methane over the baseline by 90 minutes or (2) if producing hydrogen only, an increase of at least 20 ppm hydrogen over the baseline by 90 minutes. All breath tests were evaluated by a single experienced reader who was blinded to the treatment regimen (GM).

Treatment

Subjects with newly diagnosed SIBO by LBT were given two open-label treatment choices based upon individualized treatment preference; either two 200 mg rifaximin tablets three times daily (TID) or 2 capsules twice daily of the following commercial herbal preparations; Dysbiocide and FC Cidal (Biotics Research Laboratories, Rosenberg, Texas) or Candibactin-AR and Candibactin-BR (Metagenics, Inc, Aliso Viejo, California) for 4 consecutive weeks immediately followed by a repeat LBT. Table 5 illustrates the details regarding the composition for each herbal preparation. The cost of herbal therapy was no more than $120 for a 30-day supply. The primary outcome was the proportion of patients in each group who had a negative post-treatment LBT. Rifaximin non-responders were then prescribed either the herbal protocol or triple antibiotics (clindamycin 300 mg TID, metronidazole 250 mg TID, neomycin 500 mg TID) for 4 additional weeks. The statistical methods applied to analyze the results were the student t-test and chi-squared test. The study was approved by the Johns Hopkins University (Baltimore, Maryland) Internal Review Board.

Table 5

Herbal Preparations for the Treatment of Small Intestine Bacterial Overgrowth

FC Cidal	Dysbiocide	Candibactin-AR	Candibactin-BR
Proprietary blend - 500 mg: 1 capsule	Proprietary Blend 950 mg per 2 capsules	One Capsule contains:	Two Capsules contain:
Tinospora cordifolia (stem)	Dill seed	Red Thyme oil (thymus vulgaris, providing 30%-50% thymol) 0.2 mL	Coptis root and rhizome extract (coptis chinensis, containing berberine) 30 mg
Equisetum arvense (stem)	Stemona Sessilifolia powder and extract	Oregano Oil (origanum vulgare, providing 55% to 75% carvacrol) 0.1 mL	Indian Barberry root extract (berberis aristata, containing berberine) 70 mg
Pau D'Arco (inner bark)	Artemisia Absinthium shoots and leaves extract,	Sage leaf 5.5:1 extract (salvia officinalis) 75 mg	Berberine Sulfate 400 mg • Proprietary 4:1 Extract 300 mg: Coptis root and rhizome (coptis chinensis)
Thymus vulgaris (aerial part)	Pulsatilla Chinensis rhizome powder and extract	Lemon Balm leaf 5:1 extract (melissa officinalis) 50 mg	Chinese Skullcap root (scutellaria baicalensis)
Artemisia dracunculus (leaf)	Brucea Javanica powder and extract		Philodendron bark (phellodendron chinense)
Sida cordifolia (aerial part)	Picrasma Excelsa bark extract		Ginger rhizome (zingiber officinale)
Olea europaea (leaf)	Acacia Catechu stem extract		Chinese Licorice root (glycyrrhiza uralensis)
	Hedyotis Diffusa powder and extract		Chinese Rhubarb root and rhizome (rheum officinale)
	Yarrow leaf and flower extract (achillea millefolium).		Chinese Rhubarb root and rhizome (rheum officinale).

RESULTS

A flow chart of the study design and results are shown in the Figure. Three hundred ninety-seven patients were offered LBT for suspected SIBO, of which 252 (63.5%) were positive; 87 of these patients did not pursue treatment, and 61 of those who underwent therapy did not undergo repeat LBT. The remaining 104 patients were provided the option of two treatment arms as per the patients' choice and underwent post-treatment LBT. Rifaximin was completed by 67 patients, and 37 completed herbal therapy. Results were tabulated only for those who completed therapy. There was no difference in the mean age of patients who chose rifaximin (47.55 ± 15.73) and those on herbal therapy (41.6 ffl 16.3) (P=.30). There was a gender skewing of participants as 48/67 (71%) of the rifaximin arm and 29/37 (78%) of the herbal arm were females (Table 6). Of the 67 patients who completed the rifaximin arm, 9 (13%) had risk factors for SIBO (diabetes mellitus, gastroparesis, narcotics, connective tissue disease, etc) while 7 (19%) in the herbal arm had risk factors (P=.46). Of the 252 individuals who tested positive for SIBO by LBT, 160 (63.5%) had IBS as diagnosed by the Rome III criteria. Of the 37 patients who received herbs, 17 (46%) had a negative follow-up LBT compared to 23/67 (34%) of rifaximin users (P=.24) (Table 7). The odds ratio of having a negative LBT if on herbal therapy as compared to rifaximin was 1.85 (CI=0.77-4.41, P=.17) once adjusted for age, gender, SIBO risk factors, and IBS status (Table 8). Fourteen of the 41 (34.1%) rifaximin non-responders were offered herbal rescue therapy; 8 of the 14 (57.1%) had a negative LBT after completing the rescue herbal therapy. Ten of the 41 (24.4%) rifaximin non-responders were offered rescue therapy with triple antibiotics, and 6 of the 10 (60%) had a negative LBT after completing the rescue herbal therapy. Rescue therapy with herbals in rifaximin non-responders was not different (P=1.0). The average age of those SIBO patients who cleared the LBT post-rescue herbal therapy was 37.4 ± 17 years, which was not different than non-responders (41.5 ± 11.6, P=.62) (Table 9). The prevalence of IBS subtype, known risk factor(s) for SIBO, or gender status did not predict response to herbal rescue therapy (Table 9). Adverse effects (AE) were reported among the rifaximin treated arm (6/67; 9.0%) including 1 case of anaphylaxis, 2 cases of hives, 2 cases of diarrhea, and 1 case of Clostridium difficile (post-treatment). One case of diarrhea (non–Clostridium difficile) was reported in the herbal therapy arm (1/37; 2.7%) (P=.22)

Figure

Flow chart of the study design: 397 patients were recruited to participate in the study and underwent lactulose breath hydrogen testing (LBT), 252 were found to be LBT positive while 145 were negative. Due to lack of protocol adherence, only 104 of those who were had positive LBT met our inclusion criteria for the study, with 67 opting for Rifaximin therapy and 37 preferring herbals. Forty-four of the 67 (65.7%) subjects treated with Rifaximin failed to respond and had positive LBT after therapy with only 34.3% responding. Herbal therapies were associated with a (17/37) 45.9% response rate with (20/37) 54.1% having post-treatment positive LBT. For those who failed the first round of Rifaximin therapy, a crossover trial showed a 8/14 (57.1%) response rate for herbals in those who failed Rifaximin.

Table 6

Demographic Information of Rifaximin vs Herbal Users

	Rifaximin (n=67)	Herbal (n=37)	P value
Age (yrs)	47.55 ± 15.73	41.6 ± 16.3	.07
Percent female	48 (71%)	29 (78%)	.45
Risk Factors for SIBO	9 (13%)	7 (19%)	.46
Diabetes mellitus	4 (6%)	0 (0%)	.12
Gastroparesis	2 (3%)	1 (3%)	.93
Narcotic	1 (1%)	3 (8%)	.09
CTD	2 (3%)	4 (11%)	.1
Any risk factor	9 (13%)	7 (19%)	.46
IBS	53 (79%)	25 (68%)	.19
Distribution of IBS by Subtypes			.08
None	14 (21%)	12 (32%)
Diarrhea predominant	22 (32%)	6 (16%)
Constipation predominant	16 (24%)	12 (32%)
Mixed	14 (21%)	4 (11%)
Unclassified IBS	1 (1%)	3 (8%)
Negative LBT after Rx	23 (34%)	17 (46%)	.24

Table 7

Results of IBS Subjects Undergoing Intervention for Positive Lactulose Breath

Characteristics	Rifaximin	Herbs	P value
Number	67	37	N/A
Age (y), SD, range	44.4 ± 14.8 (19-81)	41.3 ± 14.8 (19-76)	.33
Gender
Female, n (%)	48 (71)	29 (78)	.97
Male, n (%)	19 (29)	8 (22)
Responses (n)	26	15	N/A
Response Rate (%)	34	46	.24
Adverse Events (n, %)	2, 2.9	1, 2.7	.83

Table 8

Odds Ratios (OR) for Negative LBT Comparing Herbal Therapy to Rifaximin

	OR	Confidence Interval	P value
Unadjusted	1.63	0.72 - 3.70	.24
Adjusted for Age and Gender	1.62	0.70 - 3.77	.16
Fully Adjusted*	1.85	0.77 - 4.41	.17

Adjusted for age, gender, small intestine bacterial overgrowth risk factors, and irritable bowel syndrome status.

Table 9

Demographic Information of Rifaximin Nonresponders After Herbal Rescue Therapy

	Responder (n=8)	Nonresponder (n=6)	P value
Age +/− SD	37.4 +/− 17	41.5 +/− 11.57	0.62
Female (%)	5 (62.5%)	4 (67%)	0.87
IBS subtype			0.74
None	1 (12.5%)	1 (16.7%)
Diarrhea predominant	1 (14%)	2 (40%)
Constipation predominant	3 (43%)	1 (20%)
Mixed	3 (60%)	2 (40%)
Any Risk Factor for SIBOa	2 (25%)	1 (17%)	0.71

Defined as narcotic use, gastroparesis, diabetes mellitus, and connective tissue disease.

DISCUSSION

There is an ongoing resurgence of interest in the role of gut microbiota-host interactions in both health and disease. The gut microbiota is a finely balanced ecosystem that helps regulate key vital functions for the host, including immunity, barrier defense, biotransformation of toxins and carcinogens, and much more. As discussed in more detail by Okeke et al in this issue of Global Advances in Health and Medicine, imbalances in the gut microbiome (also known as dysbiosis) have been linked to the development of disorders of mood and behavior, Alzheimer's disease, and numerous gastrointestinal and systemic disorders including inflammatory bowel disease, diabetes, obesity, and cardiovascular disease., Bacterial dysbiosis in SIBO can disrupt epithelial tight junctions increasing small intestine paracellular permeability, translocation of endotoxin, and induction of proinflammatory cytokines.– SIBO can have a number of extraintestinal manifestations such as rosacea, restless legs syndrome,, arthralgias, anemia, interstitial cystitis, chronic prostatitis, and polyneuropathy. A large body of work has linked bacterial dysbiosis of the small bowel and endotoxemia to non-alcoholic steatohepatitis and progression of alcoholic liver disease, non-alcoholic fatty liver disease, obesity, and others.– SIBO is a widely prevalent condition in the practice of gastroenterology that is chronic and recurrent for patients and lacks clear treatment guidelines for practitioners. Given the role of the gut microbiome in health and dysbiosis in disease, SIBO is an important entity in clinical practice to recognize and treat.

Rifaximin is the most commonly studied antibiotic treatment for SIBO, with an overall breath test resolution rate of 49.5% (95% confidence interval, CI 44.0-55.1) in 8 clinical trials. A recent meta-analysis reported that the therapeutic efficacy of rifaximin to treat SIBO in the setting of IBS showed benefit; however, the therapeutic gain was only 9.8%, and the number needed to treat is 10.2 with mild heterogeneity (P=.25, I(2)=26%). Additionally, rifaximin is currently not approved by the US Food and Drug Administration (FDA) for the treatment of SIBO and is only labeled for hepatic encephalopathy. Furthermore, a month's supply of rifaximin retails for $1247.39, and according to Medicare part D, a patient's copay in 2014 will be $638.09 for preferred ($703.70, non-preferred) pharmacy or mail-in. Antibiotics may also produce a wide range of toxicity (Clostridium difficle colitis, antibiotic induced diarrhea, anaphylaxis, Steven's Johnson reactions, hemolytic-uremic syndrome, etc). Antibiotics have also been postulated to have pervasive adverse effects on the gut microbiome and protective biofilm layer. In fact, a 7-day course of clindamycin has been shown to reduce commensal flora and gut microbial diversity and select antibiotic-resistant genes for over 2 years. Thus, research into more effective and safer therapeutic options for SIBO are ongoing.

The present study demonstrates that herbal therapy may be as effective as antibiotic therapy in the treatment of SIBO, as indirectly measured by normalization of the lactulose breath test abnormalities. Complementary and alternative medicine has gained wide popularity among the United States population, with up to 50% of patients using at least one form of alternative therapy. In addition, it has been noted that the majority of alternative medicine users are patients with chronic conditions. Examples include relaxation techniques and acupuncture for migraines, chiropractic for back problems, and herbal therapy and relaxation for digestive problems. Drossman et al reported based on an international survey that approximately 37% of patients with functional digestive disorders seek complementary and alternative therapies to control their symptoms. More recent data indicates that up to 51% of patients with IBS report using CAM., Since CAM has gained popularity among patients with gastrointestinal diseases, especially functional disorders, physicians should familiarize themselves with available CAM therapies. Alternative therapies commonly used by IBS patients that have some level of scientific basis in the literature include acupuncture, fiber, peppermint oil, herbal, traditional, yoga, massage, meditation, mind, relaxation, probiotics, hypnotherapy, psychotherapy, cognitive therapy, and behavioral therapies. The use of antimicrobial herbs to treat SIBO has not been previously reported in the literature. A number of herbs exist that have known antimicrobial activity.– In the current study, we chose a combination of herbal preparations to provide broad-spectrum coverage against enteric coliforms., Oil of oregano (Origanum vulgare) is a well-documented botanical that directly kills or strongly inhibits the growth of intestinal microbes., Oil of oregano has other beneficial properties such as inducing apoptosis in human colon cancer caco2 cells. Berberine extracts and thymus vulgaris are also well known for their broad antibacterial activities.– Wormwood (Artemisia absinthium) has substantial antimicrobial and anti-inflammatory properties that may be important to the pathogenesis of SIBO and has been used to successfully induce remission of Crohn's Disease., There are other herbals used in our study that have noteworthy properties. Lemon balm offers anti-anxiety and antidepressant effects that may benefit patients with IBS, while coptis root has growth-inhibitory effects on human bacteria.– Red thyme essential oil inhibits the growth of Escherichia coli O157: H7 and Staphylococcus aureus. Indian Barbarry root extract (Berberis aristata) contains berberine and has antimicrobial, anti-inflammatory, and antidiarrheal proper-ties. Equisetum arvense L. was shown to possesses a broad spectrum of a very strong antimicrobial activity against a variety of enteric microorganisms including Staphylococcus aureus (S aureus), Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Salmonella enteritidis and the fungi Aspergillus nigerand Candida albicans. Thymus vulgaris has potent antimicrobial and anti-inflammatory actions.– Olea europaea inhibits the growth of a number of staphylococcal species including S aureus.

Rifaximin is a non-absorbable oral antibiotic, first introduced in Italy in 1987 and in the United States in 2004, that has been approved for a variety of indications and is presently an antibiotic of choice for SIBO. Rifaximin is presently the best studied antibiotic with an overall breath test resolution rate of 49.5% (95% confidence interval, CI 44.0-55.1). Lauritano et al in 2009 showed that rifaximin is superior to metronidazole in the treatment of SIBO; however, recurrence rates after treatment with antibiotics is high as evidenced by high breath test positivity. In our study, we demonstrated that 46% of patients normalized their LBT with herbal therapy. There was no statistical difference between antimicrobial herbs (46%) and rifaximin (34%) (P=.24). Patients were offered the choice of either therapy. SIBO tends to be a recurrent disease, and frequent antibiotic use may have longterm adverse effects on the gut microbiome and be costly; thus, herbal therapy may be a reasonable treatment option for patients with SIBO. In our study, the side effect profile of herbs when compared to rifaximin was not statistically different with a P=.22. However, the prevalence of side effects in the Rifaximin group was 9% (6/67) including C difficle and 2 non-C difficle-associated diarrhea. In the herbal group, only 1 case of non-C difficle–associated diarrhea (1%) was observed. These observations are contrary to popular beliefs. Perhaps the herbal therapies are less disruptive to the gut microbiome and while producing efficacy in resolving SIBO there appears to be less risk of C difficle; extended trials will need to confirm this hypothesis. The fear of bacterial resistance including opportunistic infections such as Clostridium difficile raises concerns among recurrent antibiotic users. The rifaximin non-responders who received herbal therapy were equally successful in resolving SIBO via the LBT as were triple antibiotics, which had much higher cost and risk of toxicity. Thus, herbal therapy appears to be effective in the treatment of SIBO patients that is initially refractory to rifaximin. Patients who were treated with herbs were not different than those who received rifaximin with respect to the number of predisposing factors for SIBO, and thus the equivalence in response was not due to disparities in underlying conditions. Induction of achlorhydria by acid-suppressive medications (eg, PPIs) has been studied as a potential risk factor for SIBO, with a recent meta-analysis of 11 studies revealing that a pooled OR of SIBO in PPI users vs nonusers was 2.282 (95% confidence interval [CI], 1.238-4.205) when duodenal/jejunal aspirate culture was used to diagnose SIBO but not for glucose hydrogen breath testing. In contrast, in our study, concurrent PPI use did not appear to influence response to either herbal therapies (P=.27) or rifaximin (P=.10).

Limitations of this study include the fact that it was a retrospective chart review and not a prospective, randomized controlled trial. Furthermore, there was heterogeneity and multiplicity of herbs used and, additionally, diet was not controlled. Our patients were also not formally assessed for symptom resolution using a standardized questionnaire. The cost of the herbal regimen is traditionally not covered by commercial insurance; however, the cost of the herbals is relatively low and many individuals did find coverage through their flexible spending accounts.

In summary, we conclude that in the setting of SIBO, patients can be given the choice of antibiotic or herbal therapy depending on their individual preference with similar response rates and safety profiles. In addition, patients who are refractory to rifaximin can receive herbal therapy as a potential rescue therapy with equivalent results to triple antibiotics. This decision is left at the discretion of the treating physician and the patient depending on the clinical setting and the patient's inclination. Future prospective studies that consolidate the antimicrobial herbal regimen to a single agent would potentially enable randomization and blinding, improve patient compliance, and diminish associated costs in this challenging gastrointestinal disorder.