Of all the long-term PPI safety concerns, this is the one with the best evidence behind it. It’s also the only one confirmed by a large randomized trial: over three years, people taking pantoprazole had significantly more enteric infections than those on placebo. Every other outcome that trial measured — fractures, kidney disease, dementia, pneumonia, death — showed no difference.
The mechanism is straightforward. Stomach acid isn’t only for digestion; it’s a sterilizing barrier. Most bacteria you swallow don’t survive a pH of 1.5 to 3. Raise that pH substantially and more of them reach your intestines alive.
What’s worth understanding is that the risk isn’t spread evenly across every infection people worry about. Gut infections: yes. Bacterial overgrowth: probably, and it matters more for reflux sufferers than most realize. Pneumonia: the evidence has largely fallen apart.
Key Takeaways
- The COMPASS randomized trial found significantly more enteric infections on pantoprazole than placebo (1.4% vs 1.0%) — the only safety signal it confirmed.
- Observational data links PPIs to C. difficile at roughly double the odds, but a 2025 pharmacovigilance analysis found that signal largely disappeared once antibiotic co-exposure was accounted for.
- The pneumonia link doesn’t hold up: a study of 4.2 million new NSAID users found no increased risk (odds ratio 1.05).
- Small intestinal bacterial overgrowth is associated with PPI use, with odds around 2.3 — and over 7 when diagnosed by the most accurate method.
- SIBO matters particularly here because it can worsen reflux, creating a loop where the drug for your reflux contributes to something that aggravates it.
- Absolute risks remain small — roughly four extra enteric infections per thousand people over three years.
- The combination worth watching is a PPI plus antibiotics, which is where the real C. difficile risk concentrates.
Why acid suppression affects infection at all
Your stomach sits at a pH of roughly 1.5 to 3 — acidic enough to kill most of what arrives in food and drink. It’s one of the more underappreciated parts of your immune defenses, and it’s the reason you can eat something slightly questionable and usually get away with it.
A PPI raises stomach pH to around 4 or higher for much of the day. That’s the point of the drug. But bacteria that would have died in an untreated stomach can survive that, pass into the small intestine, and either establish themselves there or continue to the colon.
Unlike most of the other PPI safety debates, this one has a mechanism that is clean, expected, and doesn’t require any statistical gymnastics to believe. Which is part of why the evidence for it is more consistent.
The randomized evidence
The COMPASS trial randomized 17,598 people to pantoprazole 40mg daily or placebo, double-blind, followed for a median of three years. Researchers tracked pneumonia, C. difficile, other enteric infections, fractures, gastric atrophy, chronic kidney disease, diabetes, dementia, cardiovascular disease, cancer, hospitalizations and death.
The single statistically significant finding across all of it was enteric infections: 1.4% in the pantoprazole group versus 1.0% on placebo, an odds ratio of 1.33. C. difficile specifically was roughly twice as common on the drug, but with only 13 cases in total it didn’t reach significance [Moayyedi et al., Gastroenterology, 2019].
Worth holding both halves of that in mind. It’s the one confirmed harm — and it amounts to about four additional infections per thousand people over three years. Real, but not a reason for alarm.
C. difficile: real, but mostly about antibiotics
C. difficile is the infection that generates the most worry, and with reason — it causes severe diarrhea, can be serious in older or frail people, and recurs readily.
The observational picture looks alarming. A meta-analysis of 56 studies covering 356,683 patients found PPI users had roughly double the odds of C. difficile infection compared with non-users (pooled odds ratio 1.99), consistent across age groups and study designs. The authors noted substantial heterogeneity and evidence of publication bias, and were careful to say that further work was needed to determine whether the association is causal [Trifan et al., World Journal of Gastroenterology, 2017].
That caution turned out to be well placed. A 2025 analysis of 238,470 PPI adverse event reports set out to test how much of the association survived adjustment for competing exposures. The initial signal was strong — a reporting odds ratio of 2.36. After excluding competition from antibacterial drugs, it dropped to 1.47. After further excluding antibiotic and immunosuppressant users and cases with kidney injury, only one PPI retained a signal at all. Age-stratified analysis showed complete signal loss across every age group once antibiotic exposure was accounted for. The authors concluded the PPI–C. difficile association appears mediated predominantly by antibiotic co-exposure rather than by the PPI itself [Wu et al., Journal of Clinical Medicine, 2025].
That makes practical sense. Antibiotics are the dominant cause of C. difficile — they clear the competing gut bacteria that normally keep it suppressed. People who take PPIs also take more antibiotics, because both correlate with being unwell and seeing doctors.
The practical conclusion isn’t “ignore this.” It’s that the combination is what matters. If you’re on a long-term PPI and you’re prescribed a course of antibiotics, that’s the window where the risk concentrates — and a sensible moment to ask whether the PPI is needed during that period. Antibiotics can also stir up reflux in their own right, which I’ve covered in can antibiotics cause heartburn.
Pneumonia: this one didn’t survive
The pneumonia theory was that acid suppression allows bacterial colonization of the stomach, and micro-aspiration then carries those bacteria into the lungs. Plausible, particularly for people who already reflux.
But it appears to be an artifact of who takes these drugs. Researchers addressed the confounding problem cleverly: instead of comparing PPI users against the general population, they studied 4.2 million new users of NSAIDs, comparing those who were also started on a PPI for stomach protection against those who weren’t. Both groups had the same underlying reason to be prescribed something, which strips out a lot of the “sicker people take more drugs” effect.
After adjustment, PPIs showed no increased risk of hospitalization for community-acquired pneumonia — an odds ratio of 1.05, with a confidence interval comfortably spanning 1. H2 blockers gave a similar null result. The authors concluded the data don’t support a pharmacological effect of acid suppression on pneumonia risk [Filion et al., Gut, 2014]. COMPASS found no pneumonia difference either.
Two independent approaches, both null. I’d treat the pneumonia concern as closed.
SIBO: the one that matters most for reflux
Small intestinal bacterial overgrowth gets less attention than C. difficile, but for anyone reading this because of reflux, it’s the more relevant one.
SIBO is bacteria establishing themselves in the small intestine, where numbers should be low. Symptoms are bloating, gas, distension after eating, and irregular bowels — frequently mistaken for IBS.
A meta-analysis of eleven studies covering 3,134 people found PPI use associated with SIBO at a pooled odds ratio of 2.28. The subgroup finding is the interesting part: in studies diagnosing SIBO by duodenal or jejunal aspirate culture — the accurate method — the odds ratio was 7.59, while studies using the less reliable glucose hydrogen breath test found no association at all [Lo and Chan, Clinical Gastroenterology and Hepatology, 2013].
That pattern — a strong effect visible only with the better test — suggests the association is real and that weaker studies were diluting it, rather than the reverse.
Why this creates a loop
Here’s the part that makes SIBO worth your attention specifically.
Bacterial fermentation in the small intestine produces gas. Gas raises intra-abdominal pressure. Raised pressure pushes against the lower esophageal sphincter and promotes the transient relaxations that let reflux happen in the first place.
So the sequence can run: reflux → PPI → bacterial overgrowth → more bloating and pressure → more reflux. And because the person’s reflux is getting worse, the dose often goes up.
I want to be careful not to overstate this — it’s a mechanistically coherent loop rather than something proven to be the main story in most patients. But it fits a pattern I hear about constantly: people whose reflux and bloating both worsened over years on escalating acid suppression, and who assumed the drug was all that stood between them and something worse. SIBO and acid reflux goes into the relationship in more depth, and gut health and acid reflux covers the wider picture.
It’s also relevant that this is largely invisible in silent reflux, where people are frequently on double-dose PPIs for throat symptoms the drug was never likely to fix. Maximum acid suppression, minimum benefit — which I’ve argued in detail in why PPIs don’t work for LPR.
Keeping this in proportion
Relative risks sound dramatic; absolute risks are what you actually live with. Doubling a small number leaves a small number. In COMPASS, the difference was four additional enteric infections per thousand people over three years.
Set against that: if you’re taking a PPI to prevent gastrointestinal bleeding while on NSAIDs or anticoagulants, that bleeding risk is considerably larger than the infection risk. If you have erosive esophagitis or Barrett’s, the drug is preventing damage that matters more. Don’t trade a real benefit for a small uncertain harm.
The calculation only tips when the benefit side is empty — which, for throat symptoms, it often is. In that situation the useful move isn’t finding a gentler drug, it’s removing the need for one, which is what the food, timing and night-time changes in the Wipeout Diet Plan are designed to achieve.
Practical steps
- Be alert around antibiotics. That’s where C. difficile risk actually lives. If you develop persistent watery diarrhea during or after a course, get it checked promptly rather than waiting it out.
- Consider SIBO if you’re bloated. Long-term PPI plus significant bloating and distension is a pattern worth investigating rather than accepting.
- Think about probiotics carefully. They help some people and worsen bloating in others, particularly where overgrowth is already present. Probiotics for acid reflux covers what’s reasonable.
- Use the lowest effective dose. Twice-daily dosing means more hours at raised pH and therefore less barrier function.
- Don’t stop abruptly. Rebound acid hypersecretion will make you feel you needed the drug all along. Getting off PPIs and acid rebound explains the taper, and alginate rafts are the best cover during it — they work as a physical barrier and don’t touch stomach pH, so they don’t carry this trade-off at all.
Conclusion
Infection risk is the most credible of the long-term PPI concerns, and the only one a randomized trial confirmed — but it’s confined to gut infections, it’s small in absolute terms, and the frightening C. difficile numbers appear to be driven mostly by antibiotics rather than the acid suppression itself. Pneumonia can be set aside. SIBO is the one I’d actually pay attention to, because it can quietly feed back into the reflux you started with.
The through-line across everything I’ve written on PPI safety is the same. None of these risks is large enough to justify stopping a drug that’s genuinely working. All of them are more than large enough to make an unnecessary drug a bad deal. If you’re on long-term acid suppression for throat symptoms that haven’t shifted, the answer isn’t a safer tablet — it’s addressing the reflux itself. That’s what the Wipeout Diet Plan is built around: which foods and drinks reactivate pepsin in your throat and the pH thresholds that decide it, meal timing, the night-time routine, and a structured programme to follow rather than a list of foods to avoid. It was designed first for LPR, the throat-based form where acid suppression reliably disappoints, but because it works on the same underlying mechanisms it does the job just as well for GERD and everyday heartburn.
The Wipeout Food Reference Guide is the essential lighter companion — the foods and drinks that are safe for acid reflux and LPR with their pH values, so you can check anything at a glance. The Diet Plan is the programme; the guide is the reference beside it.
If you take one practical thing from this article, make it the antibiotic point. That’s the specific window where the risk is real, and it’s the one situation where a conversation with your doctor about pausing or reviewing your PPI is clearly worth having.
Frequently Asked Questions
How much does a PPI really raise my infection risk?
In the randomized trial, enteric infections occurred in 1.4% of people on pantoprazole versus 1.0% on placebo over three years — about four extra cases per thousand. A meaningful signal statistically, a small risk personally.
Should I stop my PPI while taking antibiotics?
Ask your doctor rather than deciding alone. It’s a reasonable question, since that combination is where C. difficile risk concentrates. But if the PPI is protecting your stomach from NSAIDs or anticoagulants, stopping it temporarily may carry its own risk. It’s a genuine trade-off, which is why it needs an individual answer.
Can PPIs cause food poisoning?
They can make it somewhat more likely, yes — that’s the same mechanism as the enteric infection finding. Reduced stomach acid means fewer swallowed pathogens are killed on arrival. Worth ordinary food hygiene care and a little extra caution with street food while travelling.
Do H2 blockers carry the same risk?
Less so, since they suppress acid less profoundly and for shorter periods. The pneumonia study found no risk with either. For gut infections, H2 blockers are generally considered lower risk, though they’re also less effective where strong acid suppression is genuinely needed.
How would I know if I had SIBO?
Typical symptoms are bloating and abdominal distension that build through the day, excess gas, and irregular bowel habits — often diagnosed as IBS. Testing is by breath test or, more accurately, small intestinal aspirate. If you’re on long-term acid suppression and significantly bloated, it’s worth raising.
Will probiotics protect me?
The evidence is mixed and depends heavily on strain and situation. Some strains have reasonable evidence for preventing antibiotic-associated diarrhea. For general use alongside a PPI the picture is less clear, and in people with existing overgrowth probiotics can make bloating worse rather than better.
Is vonoprazan different?
Vonoprazan suppresses acid more profoundly and for longer than conventional PPIs, so on mechanism you’d expect the barrier effect to be at least as pronounced. It’s newer, with less long-term safety data. Worth noting rather than assuming it sidesteps the issue.
Research & References
- [Moayyedi et al., Gastroenterology, 2019] — Randomized double-blind trial of 17,598 participants on pantoprazole or placebo for a median of three years found enteric infections were the only significantly increased safety outcome (1.4% vs 1.0%, odds ratio 1.33), with no significant difference in pneumonia, fractures, kidney disease, dementia or mortality.
- [Trifan et al., World Journal of Gastroenterology, 2017] — Meta-analysis of 56 studies covering 356,683 patients found PPI use associated with Clostridium difficile infection at a pooled odds ratio of 1.99, while noting substantial heterogeneity and evidence of publication bias.
- [Wu et al., Journal of Clinical Medicine, 2025] — Pharmacovigilance analysis of 238,470 PPI reports found the initial C. difficile signal (reporting odds ratio 2.36) fell to 1.47 after excluding antibacterial drug competition and disappeared entirely across all age groups after adjustment, indicating the association is largely mediated by antibiotic co-exposure.
- [Filion et al., Gut, 2014] — Cohort study of 4,238,504 new NSAID users across eight databases found no increased risk of hospitalization for community-acquired pneumonia with PPI use (adjusted odds ratio 1.05), with similar null results for H2 blockers.
- [Lo and Chan, Clinical Gastroenterology and Hepatology, 2013] — Meta-analysis of eleven studies covering 3,134 subjects found PPI use associated with small intestinal bacterial overgrowth (pooled odds ratio 2.28), rising to 7.59 in studies using duodenal or jejunal aspirate culture for diagnosis.
David Gray
12 years living with LPR · Consultant & researcher
I've lived with LPR for twelve years — the misdiagnoses, the PPI courses that did nothing, the slow work of figuring out what actually helps. Wipeout Reflux is where I translate the research into plain terms for people stuck in the same place. Every claim here is sourced to peer-reviewed work, and I consult one-to-one with LPR sufferers.