Why SIBO comes back, and what motility has to do with it

Before you read on: recurring abdominal symptoms can mean many things. Further down there is a list of signs where a medical assessment comes before any explanation. If one of them applies to you, please start there.

For six weeks things were quiet. The belly stayed flat, morning and evening. You ate things that had been impossible for months, and nothing happened.

Then came an evening when your trousers felt tight again. You pushed it aside. Two weeks later the old pattern was back, with everything that goes with it.

Many people know exactly this pattern. And almost all of them read it the same way: I did something wrong.

That reading is understandable. According to the data, however, it is not the most obvious one. A recurrence after successful treatment is statistically not an exception. After nine months, almost one in two was positive again in the only cleanly traceable follow up investigation. You are therefore not the exception but part of a very large group. And it usually points in the same direction: the reason why anything was able to settle in the small intestine has not yet been found.

This text is about that reason. More precisely about a movement that most people have never had explained to them, although they feel it every day.

The relapse is not the exception: almost one in two after nine months

In this entire field there is exactly one figure that can be traced cleanly. It comes from Rome, from the year 2008, and it is uncomfortable.

The same work found something else. In the multivariate analysis three things were independently associated with the recurrence: higher age, an appendectomy in the history and the long term use of acid blockers.

With these three figures a second look is worthwhile, in the direction of caution. For age the odds ratio was 1.09 per year of life, for the appendectomy 5.9 and for the acid blockers 3.52. Sounds clear cut. It is not.

The confidence interval for the appendectomy runs from 1.45 to 24.19. That is so wide that the number 5.9 carries no precise statement, only a direction. For the acid blocker the interval starts at 1.07 and therefore sits close to the edge of irrelevance. I write this down because within that width lies the difference between a hint and a proof.

One question comes up almost every time in this context, and it deserves a clear answer. Does the second or third round make the gut flora resistant?

There is an investigation on this from a large irritable bowel study. Of 103 openly treated people 73 entered a blinded phase, 1429 bacterial and yeast isolates were identified and tested for susceptibility against eleven antibiotics. No lasting cross resistance was found. Worth noting in passing is the study design itself: it required that symptoms return after the initial response. Coming back was the entry ticket into the randomised phase.

What is used in the first treatment at all and how the breath test works is described in SIBO: bacterial overgrowth in the small intestine. This text here picks up afterwards, at the question of why it could come to this in the first place.

Before I go further, something has to go where it belongs, namely right up front.

The cleaning wave: what the migrating motor complex does

There is a sound that almost everyone knows and almost everyone misreads.

That deep rumbling in an empty belly, a few hours after eating. Most people read it as hunger in the sense of lack. Something is missing there, something has to go in soon.

Physiologically it is something else. It is operations. More precisely: it is the audible part of a movement that clears out the small intestine as long as nothing follows.

This movement is called the migrating motor complex, MMC for short. The word sounds unwieldy, the picture behind it is simple: the small intestine has a rinse cycle. It runs only in the fasted phase. And it runs only as long as nothing follows.

The key account of this physiology comes from a Belgian working group in Leuven.

The steering messenger at stake here is called motilin. It appears in almost no popular guide on this topic, although it makes the connection to something you can feel yourself.

The rumbling is the wave, not the lack

The same working group investigated something four years later that changed the order of things in my head.

Why the night is the longest opportunity

Many people describe that the matter feels different in the evening and at night than during the day. Some wake up with a flat belly although in the evening they looked five months along. Others report the opposite.

What is documented at this point is only the basic pattern: the migrating motor complex is a pattern of the fasted state. From that follows a simple calculation. The longest continuous fasted phase in the life of most people is the night. It is therefore the time in which the wave is interrupted least.

What follows from that is a derivation and not a study result: whoever eats very late and very generously shortens exactly this window. I write this with that label because on the question of whether late meals change the relapse rate I found no human study. So no figure in hours appears here either.

And one distinction absolutely belongs beside this, because it often gets lost online. Symptoms that wake you at night sit at the very top of the list of red flags. Having a flatter belly in the morning is something different from being woken by your belly. The second belongs in a medical assessment, and that before anyone thinks about motility.

And what about the vagus nerve?

In the functional scene the cleaning wave is often told as a vagus affair. Calm the nerve, the wave runs. This account does not hold for the small intestine in this form, and I think it is fair to say so openly.

In the same review it says: cutting the vagus nerve abolishes motor activity in the stomach but leaves the periodic activity in the small intestine intact. The small intestine has its own pacemaker system.

That does not mean that the nervous system and the gut have nothing to do with each other. It only means that the connection runs differently than often portrayed. How it runs is described in Gut brain axis and vagus. I am not explaining it a second time here.

What the evidence really shows, and where it stops

Now comes the section in which I correct myself before you do.

The motility hypothesis is strong. It is plausible, it has measured data, and it explains a great deal. But it is not what it mostly appears as online, namely the explanation. This can be shown with the very work that established the hypothesis.

I have rarely found this limitation in the freely available guide literature on this topic. It sits in the same abstract as the finding that is mostly reproduced. That is not a reproach to anyone, it happens when citing at second hand. But it shifts the statement from an if then to a sometimes.

Two further works have quantified the observation.

This last point matters clinically. If the overgrowth helps to slow the movement, then after a successful antimicrobial phase more runs again at first. That is exactly what many people describe: clearly better for a few weeks, then a slow fading. The cause underneath was never gone.

What these human data cannot do is prove the direction. They show associations. For causality the path leads into the animal model, and there the chain is directly visible.

The same group triggered the effect from a completely different side as well. In rats with interrupted bile flow the MMC cycle length rose from 17.3 minutes to as much as 34.1 minutes, and the bacterial count in the jejunum increased.

That is a second, independent path to the same end stretch. Anyone with stubborn relapses may include the bile side in their thinking. How that hangs together is described in Bile, bile acids and TUDCA.

What slows the cleaning wave

If symptoms return after a successful treatment, the most useful question is not which remedy comes next. It is: what in your history has slowed the movement up there?

There is not one answer to that. There is a list, and the points on it are documented to different degrees. That is exactly what often does not get mentioned. Eight factors are frequently listed as if they were equally strong. According to the works I have read on this, they are not.

After a gastrointestinal infection

For many people the story begins with a single event. A travellers diarrhoea. A food poisoning. An infection that was over after a few days, after which nothing was ever quite the same again.

What post infectious irritable bowel syndrome involves overall is described in Irritable bowel: finding the causes. Here only the bridge from that infection to the sluggish wave is of interest. And it has a name.

This chain can be mapped in human blood in part. There is a test for antibodies against CdtB and against vinculin. And here it gets interesting, because the numbers carry something different from what the test is often taken to promise in everyday use.

Transparency calls for half a sentence: among the authors of this validation work were employees of the manufacturer of rifaximin. That does not devalue the data. But it belongs in a text that claims transparency.

Underactive thyroid

The thyroid steers pace. That also applies to the smooth muscle of the gut.

If you have recurring relapses, the thyroid belongs in the picture. Which values actually say something is described in Thyroid values: which ones really count, and what nutrition can contribute in Nutrition with Hashimoto.

One point on this matters to me, and it applies without exception: existing thyroid medication is not altered, not reduced and not stopped because of this association. Whether a prescription fits is discussed medically, and every adjustment belongs in medical supervision.

Diabetes and blood sugar

On almost every page a confirmed risk factor is listed here. The best available summary says something more cautious.

Beside this belongs a work that did not depend on the breath test. A Greek group around Pyleris aspirated secretions from the duodenum in 320 people during a gastroscopy and cultured them. In 19.4 percent overgrowth was found. In the regression, irritable bowel syndrome, a history of type 2 diabetes and the use of proton pump inhibitors were independently associated with it.

Two works, two methods, two different answers. I let that stand. The most honest sentence is: blood sugar belongs on the list, but not at its top.

Connective tissue disease and scleroderma

This is the strongest association in the whole field, and it supports the motility hypothesis most clearly. Where the bowel muscle is structurally altered, something can settle up top more readily.

On other connective tissue diseases and on forms of disturbed circulatory regulation I found no robust figures in the literature, although they are regularly named on German pages. Clinically I observe that these people often have a sluggish passage. I cannot back that with a figure, and therefore it stands here as an observation and not as a finding.

Opioids and other medications

Opioids are one of the clearest brakes on motility there is, and they have a peculiarity that many people do not know.

And now the sentence that necessarily belongs beside it: a prescribed painkiller is not stopped on your own and not reduced on your own. That is a question for the consultation, not for a blog article. The same applies to laxatives, to antibiotics and to everything else that sits on a prescription.

Acid blockers, and an open contradiction

Here things get untidy, and I deliberately do not tidy that up artificially.

On one side stands a meta-analysis by Lo and Chan with 11 studies and 3134 people. The pooled odds ratio was 2.28. In the studies that directly cultured secretions from the small intestine, so with the most accurate method, it was 7.59. That figure deserves the same caveat as the appendectomy above: the confidence interval runs from 1.8 to 31.9, so it carries a direction and not a magnitude. The authors additionally point to four outlying studies and a possible publication bias. In the studies using a glucose breath test, by contrast, no association was found. This fits the Lauritano finding above, where the long term use of acid blockers was one of three independent relapse factors.

On the other side stands the critical meta-analysis by Shah and colleagues. In people with irritable bowel syndrome, the use of proton pump inhibitors was not linked with overgrowth, with an odds ratio of 0.8 and a p value of 0.55.

My impression: the contradiction hangs on the measurement method. The more accurately measurement is done, the clearer the association becomes. I cannot prove that, and therefore both stand side by side.

Anatomy: previous operations, adhesions, blind loop

Sometimes it is not about the movement but about the shape.

A bowel segment that is excluded from transit can hardly be reached by a cleaning wave. There is a small but instructive investigation on this. Di Stefano and colleagues treated people with overgrowth and an anatomical peculiarity after gastrectomy or gastrojejunostomy with a blind loop. An absorbable antibiotic lowered hydrogen excretion more clearly than a non absorbable one. The authors explain this by the fact that a non absorbable agent reaches a segment excluded from transit less well.

The study type belongs in the classification. The treatments there ran one after the other and were not randomised against each other, the work is listed as a comparative study and not as a randomised controlled trial. Both agents named are prescription only antibiotics, and their use in overgrowth lies outside the licence in Germany.

For you that means: if the same treatment keeps failing quickly in your case, that can be down to the anatomy and not to your consistency. The case number of this investigation is very small, and the question belongs in medical hands.

Adhesions after abdominal operations belong on the same list. Mechanistically that is undisputed, a robust figure on it I did not find, and therefore I name none here. And then there is the finding from the Lauritano work: an appendectomy in the history, with an odds ratio of 5.9 and a confidence interval from 1.45 to 24.19. That is a trace, not a magnitude.

The ileocecal valve, and why stomach acid comes first

Between the small and the large intestine sits a valve. It is meant to prevent large bowel contents from flowing back, into a place where far fewer bacteria belong.

Chronic stress

Finally the point where I can deliver least.

That persistent stress changes digestion is something I observe constantly in the consultation. The targeted search for controlled measurements in humans that directly link psychological stress and fasted state motility, however, gave me no usable hits. And the vagus role, as described above, is narrower for the small intestine than often portrayed.

So this is a clinical observation. I write it down because it belongs to my everyday work, and I label it as what it is. What is documented about the connection between the nervous system and the gut is described in Gut brain axis and vagus.

Meal gaps: why constant snacking does not let the wave run

Imagine a street cleaning service that can only drive when no cars are parked. As long as someone parks every ten minutes, it never gets through.

That is roughly what it looks like in the small intestine when something keeps arriving all day long.

The physiology on this is regarded as well established. The migrating motor complex is a pattern of the fasted state, and eating switches the gut into digestive mode. That is not an interpretation, that is how it stands in the review from Leuven. Anyone who keeps eating something across the day keeps the system largely in one operating mode and rarely lets it switch into the other.

What exactly counts as an interruption has not been cleanly measured in humans. Mechanistically much suggests that a coffee with milk, a biscuit or the three nuts in between are likely to fall into that category. The word likely stands here on purpose.

And now comes the reversal that carries this whole section. The hunger that appears after a few hours is, according to the 2016 data, not the sign that something is missing. It is the perceptible sign that the wave is running right now. That turns the feeling in the belly from an alarm into feedback.

What counts as an interruption at all

This is the question I am asked most often about it, and I answer it with the vagueness it deserves.

Physiologically the gut switches into digestive mode when nutrients arrive. From that it follows mechanistically that water and unsweetened tea probably do not represent a relevant interruption, while a coffee with milk, a juice, a bar or a handful of nuts are likely to fall into that category.

The word probably stands there on purpose. On the question of which amount and which composition actually interrupts the cleaning wave in humans, I found no robust measured data. On chewing gum and on sweeteners likewise none. Naming precise thresholds for this would go beyond the data.

More useful in practice than a rule is an observation. Count once on a normal day how often anything arrives at all. For many people it is between eight and twelve events, and the fewest of them were planned meals. That number often says more than any timetable.

What honestly is not documented here

I have rarely found this point in the freely available guide literature on this topic. I write it out at length, not as a reproach to anyone, but because it shifts the statement.

There is not a single human study showing that longer meal gaps lower the relapse rate in overgrowth. The targeted search for meal frequency, snacking, fasting interval and recurrence returned zero hits. Not few. None.

What does exist is a derivation from two well documented links. Eating interrupts the motor complex, that is documented. A disturbed motor complex goes along with overgrowth, that too is documented. The conclusion in between is plausible. It is still a conclusion and not a result.

And that is why you get no number of hours here.

What I can say instead is a direction. Fewer interruptions than before. Drinks without calories in between instead of snacks. And, this I consider the most useful part, reading your own feeling of hunger as feedback instead of as an emergency.

How many hours that comes to for you depends on your everyday life, on your blood sugar behaviour, on your work and on your history. This question belongs in a conversation and not in a table.

A second factor comes in, and it concerns not the timing but the composition. A review by Knez and colleagues from the year 2024 argues that the nature of the food co determines transit speed, and names among other things fibre and polyphenols. That is a narrative review and not a relapse study. It grounds a direction, not a rule. What is behind the fibre discussion and what is not is described in Fibre myths: what holds up.

And for all questions about FODMAP, dietary phases and food lists: they do not belong in this text. They are described in Using FODMAP properly.

Prokinetics: what they can do and what the studies really show

When it comes to relapses, sooner or later one word appears: prokinetic. An agent meant to prompt the movement, usually taken in the evening or at night.

What the overgrowth itself is treated with, whether antibiotic or herbal, is not the subject of this text. For that there is Oregano oil and berberine in SIBO and the foundational article on overgrowth. And if constipation and methane are in the foreground for you, partly different considerations apply, which are described in IMO: methanogens and constipation.

Before anything else comes, a framing sentence that applies without exception.

The one study that exists on this

One sentence circulates in the German language internet: erythromycin is said to have delayed SIBO relapse by five months in a study. This sentence goes back to a real piece of work. But it does not report it correctly.

Two things about this matter, and both are missing from the circulating short version.

First: for erythromycin of all things, the agent usually meant today, statistical significance was not reached. The five months quoted everywhere come from the other arm.

Second: that other arm was tegaserod. This medication was first withdrawn in 2007 because of cardiovascular concerns, reintroduced in the United States in 2019 for a narrowly defined group and withdrawn from the market again on 30 June 2022, according to the manufacturer for business reasons. In Germany it is not available. So the only statistically significant option in that study no longer exists.

What belongs with erythromycin, and is missing from the short version

Erythromycin is prescription only and licensed as an antibiotic, not as an agent for bowel movement. Anyone who receives it for that receives it off label, so outside the licence. That is permitted, but it shifts responsibility onto the prescribing physician, and you should be told about it.

And it has risks of its own that have nothing to do with the gut. Erythromycin can prolong cardiac repolarisation, which in rare cases can lead to dangerous rhythm disturbances. It can inhibit the breakdown of many other medicines, among them statins, certain blood thinners and some psychotropic drugs, and thereby raise their levels. For anyone with heart disease, a known QT prolongation or a liver disease, or anyone taking several long term medications, this is a genuine question and not a formality. On top of that: an antibiotic every night over months is also a question of resistance. I deliberately name no dosage here, that belongs in the consultation.

Prucalopride is regularly named as a prokinetic for overgrowth in German guide literature. I found no study investigating prucalopride with overgrowth or its recurrence. Prucalopride is prescription only and licensed for chronic constipation. In overgrowth its use would be outside the licence, and for relapse prevention no study on it exists. Its restrictions include known hypersensitivity, bowel perforation or obstruction and severe inflammatory bowel disease. I name it here only because it appears regularly in guide literature. Naltrexone is prescription only as well, and the low dose use for bowel symptoms lies outside the licence. On that there is a separate text, namely Low dose naltrexone in the gut. I am not presenting the evidence a second time here.

Herbal: what exists and what it applies to

For two herbal approaches there are clean randomised data. Both come from different research questions, and that belongs said.

For a well known standardised herbal mixture there is a placebo controlled study with 60 people from the year 2001, in which the symptom score improved more clearly than under placebo.

Three limitations belong beside this, and they matter more than the figure. What was studied was functional dyspepsia, so upper abdominal symptoms, not overgrowth and not its recurrence. The formulation tested at the time contained greater celandine. Because of reports of liver damage there was a regulatory procedure on this, and the formulation was changed. So the preparation of today is not the preparation of the study. And herbal finished medicinal products are medicines too: they can burden the liver, they have contraindications, and with an existing liver disease, in pregnancy, while breastfeeding and in children the question belongs in a medical conversation. Anyone who uses something like this does so on the basis of a transfer and not on the basis of evidence.

And what about bitters?

At this point the common account often goes further than the data support. Hence a counter finding from the same working group in Leuven that carries the rest of this article.

Bitters have their place, and I write about that in Bile, bile acids and TUDCA. As a universal answer to a sluggish cleaning wave they are not suited according to these data.

Where the functional view and the guideline diverge

There is a sentence that many people with this diagnosis have already heard. It goes roughly: I do not think much of SIBO.

After months of symptoms this sentence hits hard. But it has a very good reason, and I think you should know it. Not so that you accept it, but so that you can hold the conversation differently.

This work confirms at the same time that the association between irritable bowel syndrome and overgrowth exists, with an odds ratio of 3.7 and, against exclusively healthy controls, of 4.9. Both stand in the same work. It does not say that the phenomenon does not exist. It says that our measuring instruments delineate it poorly.

How the breath test works, which thresholds apply and what to watch for is described in SIBO: overgrowth in the small intestine. This text deliberately does not explain the diagnostics a second time.

And the guidelines?

Now comes the part that interests me most about this topic. Because the same professional world that is sceptical about the test says something very clear about the search for causes.

Hold these two things side by side for a moment. The same professional society that calls the definition imprecise recommends as the first direction of treatment that the underlying causes be identified and addressed.

That is exactly the thesis of this article. It does not sit at the margins, it sits in a guideline document of one of the two large American professional societies.

In practice that means something specific for your next conversation. If you come with a positive breath test and the result is presented as the sole proof, the scepticism is understandable. If you come with a history in which an infection, a thyroid diagnosis, an operation or a long term medication appears, then you are talking about something the guideline takes seriously too.

What this means for you when the symptoms come back

I deliberately give no protocol here. No step plan, no sequence with week numbers, no list to tick off.

What I give are seven questions. They are what I ask in the consultation when someone comes with a second or third relapse. You can take them with you into your next conversation.

What I would still like to say at this point concerns the gut less than the feeling that comes with the third relapse.

Many people experience it as proof that something is fundamentally wrong with them. That they cannot manage it. That they were not disciplined enough.

According to the data that is not the most obvious explanation. After nine months almost one in two is positive again, and that also holds for people who did everything exactly as it had been discussed.

And if you want to know how this question belongs in a larger context, in which stomach acid, barrier, microbiome and movement are thought about together: that is described in the overview article Gut reset: the whole picture approach.

Frequently asked questions about relapse and motility

Where this topic connects to the rest of the body

A sluggish cleaning wave rarely stands alone. It hangs on the thyroid, on sleep, on the nervous system and on what happens up in the stomach before anything moves on at all.