How Is SIBO Actually Treated? Antibiotics, Herbal Antimicrobials, and the Elemental Diet
Three approaches have published trial data behind them. What each one involves, how they compare, and why clearing the overgrowth is only half the work.
Photo: Leandro De Torres / Unsplash
The three routes, and what decides between them
SIBO is treated three main ways, and each has published trial data behind it. There is rifaximin, a prescription antibiotic taken as a short course. There are herbal antimicrobials, given as a combination of at least two agents over roughly four weeks. And there is the elemental diet, a two-week liquid formula in which every nutrient arrives already broken down. Which route suits you depends on your gas type, what you can afford, and how much disruption you can tolerate. After that comes the motility work, which decides whether the overgrowth stays away.
SIBO stands for small intestinal bacterial overgrowth. Bacteria that belong in your colon set up shop in your small intestine, where they ferment your food before you have finished absorbing it. The wider picture sits on the SIBO and motility hub.
All three assume the diagnosis is confirmed first, and breath testing has genuine reproducibility problems, since the same sample can read positive or negative depending on your lab's cutoff. Gas type shifts the plan too: methane-predominant overgrowth, now called IMO, usually needs its own approach.
Rifaximin, and what it costs to actually get it
Rifaximin, sold as Xifaxan, is the antibiotic most gastroenterologists reach for, and it stays largely inside the gut rather than being absorbed into the bloodstream. The pooled numbers look reasonable at first: a 2017 review of 32 studies covering 1,331 patients found an eradication rate of 70.8 percent by intention to treat, with adverse events in 4.6 percent, though its authors called the quality of those studies generally poor.
A 2021 meta-analysis complicates that. Across 21 observational studies and 5 randomized trials covering 874 patients, intention-to-treat eradication came out at 59 percent. Then the authors looked only at the randomized trials, and in those five, comparing rifaximin against placebo or an active control, there was no statistically significant difference (n = 203, risk ratio 1.14, 95% CI 0.59 to 2.19). The controlled comparisons are far less settled than the headline percentage suggests.
The obstacle I run into most in clinic is the price. Xifaxan is extremely expensive out of pocket, and coverage is notoriously hard to get, because some plans require you to fail other antibiotics first. I have had patients tell me they drove between clinics asking whether any of their doctors had samples from the pharmaceutical representatives. That's a treatment plan defeated by a formulary rather than by biology. Whether rifaximin suits you belongs with the clinician who can prescribe it.
Herbal antimicrobials, and the 2014 comparison study
The study people cite here was published in 2014, out of a tertiary care referral gastroenterology practice. Patients who tested positive on lactulose breath testing were offered either rifaximin or a herbal protocol for four weeks, then retested. Of the 37 who took the herbal route, 17 (46 percent) had a negative follow-up test, against 23 of 67 (34 percent) on rifaximin. The difference didn't reach statistical significance (P = .24). Nobody was randomized to an arm. The finding is genuine, but it isn't a head-to-head trial.
The rescue arm is the part I find most useful in practice. Of the patients who didn't respond to rifaximin, 14 were offered herbal rescue therapy and 8 (57.1 percent) then tested negative, against 6 of 10 (60 percent) given triple antibiotic therapy. Side effects, meanwhile, were reported almost entirely in the rifaximin arm: one anaphylaxis, two hives, two diarrhea, and one Clostridium difficile infection, against a single case of diarrhea on the herbal side, and at these numbers the safety difference didn't separate (P = .22). Herbal antimicrobials are still not risk-free, and they belong under the supervision of a practitioner who knows everything else you take.
One pattern I see in clinic, which is a clinical impression rather than something a trial has measured: single-agent herbal courses seem to stop working with repeated use. If someone has taken oregano or berberine on their own for more than about four weeks, the response is often gone by the time they reach me. This is why the herbal route gets given as a combination, and why I rotate agents when one has run a long stretch.
The elemental diet, and the taste problem
The elemental diet is a two-week liquid formula in which every macronutrient arrives pre-digested: proteins as free amino acids, fats as medium-chain triglycerides or smaller, carbohydrates as simple sugars. You absorb all of it high in the digestive tract, before it reaches the stretch of small intestine where the bacteria are waiting to ferment it.
The breath-test numbers here are the highest of the three, though the three studies aren't comparable: this one had no control arm and enrolled people with IBS. In a 2004 study of 93 people with IBS and an abnormal lactulose breath test, 74 of them (80 percent) had a normal test after 15 days on an exclusive elemental formula, rising to 85 percent once those who continued to day 21 were included.
Then there is the thing anyone who has tried one will tell you, which is that it tastes awful. I have a bias against the elemental diet: I would personally rather fast for 14 days than drink one. Palatability has improved, though. A 2025 Cedars-Sinai trial gave 30 people two weeks of an exclusive palatable elemental formula, and all 30 completed it. Breath tests normalized in 73 percent, and 83 percent reported adequate global relief. Two caveats belong alongside those numbers. There was no control group, and both figures were secondary endpoints, since what the trial set out to measure was change in the stool microbiome. One of the authors is also affiliated with the company that makes the formula.
One limitation belongs in the conversation before you start, and it's reasoning rather than a tested finding. An elemental formula delivers its carbohydrate as simple sugars, so if small intestinal fungal overgrowth is also part of your picture, that's the concern clinicians raise. Nobody has tested the two against each other. That one is a conversation for whoever is managing your case.
Clearing the overgrowth is only half of the work
Eradication rates describe the end of a treatment course. They say nothing about where you are a year later. In a 2008 study, 80 patients whose SIBO had been cleared with rifaximin were retested at three, six, and nine months. Glucose breath tests were positive again in 12.6 percent at three months, 27.5 percent at six months, and 43.7 percent at nine months, with gastrointestinal symptoms returning alongside them. Older age, a history of appendectomy, and chronic proton pump inhibitor use were all associated with recurrence.
Nearly half of the people successfully treated were positive again inside a year. That isn't the antimicrobial failing, seeing as it did clear the overgrowth; it is what happens when the conditions that allowed it are still in place afterwards.
Of the conditions still in place afterwards, motility is the one I spend the most time on, because it's the one that's modifiable. The migrating motor complex is the between-meal wave that moves residue and bacteria out of the small intestine, and a suppressed one leaves the small intestine unable to clear itself. I have written already about why that sweep decides whether SIBO returns and what chronic stress does to it. A plan that ends the day the breath test turns negative has finished half of its work.
When the standard routes stall
Some cases respond quickly, others stall, and a few get worse. Long-standing SIBO can leave behind nutrient deficiencies and a system that no longer regulates inflammation predictably, so even a well-chosen protocol can arrive at a body not ready for it. When that happens, the useful move is to stop treating the overgrowth as the first target. I describe this to patients as finding the loosest thread in a complex knot, meaning whichever piece will give first if you support it. Once that piece is steadier, the rest usually tolerates treatment better.
If you want a structured walk-through of SIBO, the SIBO Treatment Algorithm covers testing, the treatment phases, and the motility work that keeps you well after.
FAQ
Can SIBO be treated naturally?
Yes, and the herbal route has been measured against the antibiotic in a published study rather than argued for from tradition alone. In a 2014 study from a tertiary referral gastroenterology practice, 46 percent of the people who took a four-week herbal antimicrobial protocol had a negative follow-up breath test, compared with 34 percent of those who took rifaximin. The trial couldn't separate those two numbers statistically, and patients were offered a choice between the arms rather than being randomized to one, so read it as two comparable options rather than as one beating the other. Herbal antimicrobials still interact with medications and belong under the supervision of a clinician who knows your full list.
Are herbal antimicrobials as effective as rifaximin for SIBO?
On the one direct comparison available, they performed at least as well, though the study wasn't randomized and the arms were unequal at 37 and 67 people. The same study looked at what happens when rifaximin doesn't work: 8 of 14 non-responders given herbal rescue therapy tested negative afterwards, against 6 of 10 given triple antibiotic therapy. So a person who doesn't respond to the antibiotic has a second option that isn't simply more antibiotics.
Why does SIBO keep coming back after treatment?
Because clearing the bacteria doesn't change the conditions that let them accumulate. In a 2008 study of 80 patients whose SIBO had been cleared with rifaximin, breath tests were positive again in 12.6 percent at three months, 27.5 percent at six months, and 43.7 percent at nine months. Older age, a history of appendectomy, and chronic proton pump inhibitor use were all associated with recurrence. That study didn't measure motility, so it can't tell you how motility ranks against those three. In my own practice it's the condition I work on hardest after the kill phase, because it's the one you can actually change.
What is the best diet during SIBO treatment?
There isn't a single diet that clears an overgrowth on its own, and the choice depends on which route you are taking. The elemental diet is itself a treatment rather than a supporting diet: a two-week exclusive liquid formula in which proteins arrive as amino acids and carbohydrates as simple sugars. Alongside antibiotics or herbal antimicrobials, most people are steered toward a temporary reduction in fermentable carbohydrates rather than a long restrictive plan. Anything beyond a few months of restriction is a conversation to have with your clinician.
How long does SIBO treatment usually take?
The kill phase is short. Rifaximin is a course of a couple of weeks, though the published trials have used anywhere from one to four, herbal antimicrobial protocols generally run about four weeks, and the elemental diet is 14 days, extended to 21 in the studies when the breath test had not normalized. The motility work afterwards is the long part, and in my practice it runs in months rather than weeks, which is my clinical impression from practice rather than something anyone has put a number to. The study that followed people after treatment found breath tests turning positive again at three, six, and nine months, and nobody has published what happens after that.
References
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
