Peptic Ulcer Disease and H. pylori: H. pylori: Biology, Eradication, and Salvage

0:00Welcome to Board Pearls. This is episode one of two of the Peptic Ulcer Disease and H. pylori chapter, in the Stomach and Small Bowel Disorders module. This episode is H. pylori from its biology through eradication and salvage, and a handful of mechanisms carry the whole thing, starting with how the organism survives at all.

0:20H. pylori is a spiral bacterium that lives in one place nothing else reliably occupies: the mucus layer coating the stomach lining, where it likes a near-neutral pH. It survives the acid stomach by making enormous amounts of urease, an enzyme that splits urea into ammonia and carbon dioxide, and the ammonia neutralizes acid in a thin shell around each organism, giving it a livable pocket. That one trick is load-bearing for diagnosis, because the two urease-based tests exploit it directly: the breath test has the patient swallow labeled urea, which the bacterium splits into labeled carbon dioxide you detect in the breath, and the biopsy urease test drops tissue into a urea reagent that changes color when bacterial urease makes ammonia. So the same enzyme that protects the organism is how we find it, and, crucially, it's also why acid suppression causes false negatives, because knocking the bacteria down drops their urease output below what the test can catch.

1:21Two virulence factors matter for the strains that actually cause disease. One is a pathogenicity island that builds a molecular syringe injecting an oncoprotein into the stomach lining cells, disrupting their junctions, amplifying inflammation, and tracking with higher cancer risk. The other is a vacuolating toxin that some strains secrete to damage cells and dig in for chronic infection. About half of humanity carries the organism, prevalence is falling where childhood hygiene has improved, and only a minority of carriers ever develop ulcers, with a smaller fraction developing the cancers covered elsewhere.

1:54Where the infection settles in the stomach decides which disease it causes, and this is the idea the whole chapter turns on. One pattern inflames the whole stomach including the acid-producing body, which damages the acid-making cells and lowers acid output, and over decades that patient develops gastric ulcers, atrophy, intestinal metaplasia, and the intestinal-type gastric cancer sequence. The other pattern inflames mainly the antrum while sparing the body, and the pivot there is the loss of the antral D cells that make somatostatin. Somatostatin normally restrains gastrin, so lose the D cells and you lose the brake: gastrin rises, the untouched body cranks out more acid, and the duodenal bulb gets hit with an acid load it can't handle. The bulb responds by turning patches of its lining into stomach-type tissue, and since H. pylori can only stick to stomach-type lining, those patches give it a foothold in the duodenum, and a duodenal ulcer follows. Same organism, same host, but a different location gives opposite acid physiology and opposite disease, which is why one infected patient makes a low-acid gastric ulcer and another makes a high-acid duodenal ulcer, and why eradication improves reflux in some and worsens it in others.

3:10Diagnosis is dictated entirely by what suppresses the bacteria. Acid blockers raise the pH and knock down bacterial density and urease, and antibiotics and bismuth do the same, so any of them on board at the time of testing produces a false negative on every test that detects the live organism or its products. That's the rule the strategy is built around. The four active tests are the breath test and the stool antigen, which are non-invasive, and the biopsy urease test and histology. All four are accurate in the right conditions and all four false-negative when the bacteria are suppressed, so the patient has to be off an acid blocker for one to two weeks and off antibiotics or bismuth for four weeks, and those windows reflect how long it takes bacterial density to rebound to detectable levels. Histology is the most resilient of the four, because it visualizes the organism directly on the tissue rather than depending on its enzyme output, so the patient who genuinely can't stop their acid blocker because of severe esophagitis or active bleeding is best served by biopsy with histology on a body sample. And the board point that follows: a negative breath or urease test in someone still on an acid blocker does not exclude infection, so you hold the drug and repeat.

4:24Serology is the exception to the suppression rule, because it measures antibodies and isn't affected by bacterial density, so it needs no drug hold, but it has a fatal flaw: antibodies persist for years after the organism is gone, so it can't tell active from past infection. Its only acceptable use is first-time screening in a low-prevalence setting without access to active testing, and it is never the test of cure. A question that checks serology after treatment is testing whether you know the antibody lingers.

4:53And the test of cure is non-negotiable: everyone treated gets retested, because first-line eradication fails in a meaningful fraction and you can't tell who from symptoms, done at least four weeks after the last antibiotic and after the acid-blocker hold, using the breath test or stool antigen, not serology. The same suppression rules that govern initial testing govern the recheck.

5:16The reasons to test have broadened, and knowing when to test outside the obvious ulcer is rewarded. Test in active or prior peptic ulcer, in gastric MALT lymphoma, in uninvestigated dyspepsia under sixty, in functional dyspepsia, in long-term NSAID or aspirin users, in unexplained iron-deficiency anemia, in immune thrombocytopenia, in autoimmune gastritis and premalignant gastric conditions, and in first-degree relatives of someone with gastric cancer. The organizing idea is that any state where treating the infection changes the outcome is on the list, and the family-history-of-gastric-cancer indication is the newer one that catches people out.

5:57Two scenarios bring the suppression rule to life. A patient on an acid blocker for years gets scoped for epigastric pain, has a benign-looking gastric ulcer, and the biopsy tests come back negative. Calling that an idiopathic ulcer and continuing the drug is wrong, because the drug has been suppressing the bacteria for years, so you hold it for one to two weeks and repeat with a breath or stool test. And a patient who just finished amoxicillin for a dental abscess and has been on an acid blocker for six weeks gets a negative breath test, which is simply uninterpretable, because both the recent antibiotic and the acid blocker invalidate it; you wait out the windows and retest.

6:36Now first-line treatment, where the ground has shifted, and the reason is mechanistic. Clarithromycin resistance is a single binary mutation at the ribosome: the drug either binds or it doesn't, and raising the dose can't fix it. Levofloxacin resistance works the same all-or-nothing way. Across much of the US, clarithromycin resistance now runs above fifteen percent, and fifteen percent is the line above which you can no longer give clarithromycin-based triple therapy empirically with confidence, which is why it's fallen out of first-line in most US practice. Meanwhile resistance to amoxicillin, tetracycline, and rifabutin has stayed very low, and that low-resistance trio is the foundation the modern regimens are built on.

7:18The one regimen with a strong recommendation for a treatment-naive patient is optimized bismuth quadruple therapy for fourteen days: an acid blocker, bismuth, tetracycline, and metronidazole, with a potent acid blocker like esomeprazole or rabeprazole and the full fourteen days rather than ten. Each piece does specific work. The acid blocker raises the pH, which pushes the bacteria to replicate, and replicating bacteria are more vulnerable to the antibiotics, so the acid suppression is actually making the antibiotics work. Bismuth has its own antibacterial action and importantly still hits metronidazole-resistant strains, which is why these regimens keep working where metronidazole resistance is common. And tetracycline and metronidazole attack two different pathways at once, covering strains that have lost susceptibility to either alone. The advantage that puts this first is that it contains no clarithromycin and needs no susceptibility testing, since its components have low, stable resistance. The cost is complexity: four drugs on multiple daily schedules, with nausea, dark stool and tongue from the bismuth, and a metallic taste and an alcohol reaction from the metronidazole, plus a caution that patients with aspirin allergy shouldn't get bismuth subsalicylate. Adherence is the single biggest reason it fails, so it has to be prescribed with real counseling.

8:42Vonoprazan has entered first-line, and it's worth understanding as a mechanism. It blocks the acid pump directly and reversibly, so unlike the older acid blockers it doesn't need an acidic environment to activate, and its suppression is fast, deep, long-lasting, and not dependent on the liver enzyme that makes some patients poor responders. The result is a stomach that sits at a higher pH for more of the day than the older drugs can manage, and that sustained suppression is what makes the eradication work. There are two vonoprazan-based first-line options: a triple with amoxicillin and clarithromycin, and a dual with just amoxicillin, both for fourteen days. The dual is the cleaner choice where clarithromycin resistance is high because it drops the macrolide entirely, and the reason amoxicillin can work almost on its own here is that vonoprazan holds the pH high enough, long enough, that amoxicillin stays active against the replicating bacteria for most of the day, which it couldn't do under weaker suppression.

9:38The other regimens fill out the second tier, and several inherit the clarithromycin-resistance problem, so none is the empiric first-line answer when no susceptibility data are given. And plain clarithromycin triple survives only when local resistance is documented under fifteen percent and the patient has never had a macrolide, and remember that an azithromycin for a sinus or chest infection counts as macrolide exposure, so in most current US settings both conditions are rarely met and clarithromycin triple is the wrong answer.

10:06The selection logic you can run in your head is mechanical: ask about penicillin allergy, ask about any prior macrolide, and ask about any prior fluoroquinolone, then pick a regimen that avoids what the patient has already seen and lean on the low-resistance trio. No allergies and no prior macrolide gives the most options. A penicillin allergy takes away the amoxicillin regimens and points to bismuth quadruple, with the important caveat that most penicillin-allergy labels aren't true allergies, so a childhood rash without anaphylaxis is a reason to send for allergy testing and de-labeling rather than lock in a weaker regimen. Prior macrolide exposure takes away the clarithromycin regimens and again points to bismuth quadruple. In nearly every hard vignette, bismuth quadruple is the answer, because it has no clarithromycin and its components have low, predictable resistance.

11:00When first-line fails, you're now treating an experienced patient, and one rule drives every salvage choice: resistance rules selection. Clarithromycin and levofloxacin should not be reused after a failure unless testing shows the strain is still susceptible, while amoxicillin, tetracycline, and rifabutin can be reused even after exposure because resistance to them stays very low. That's the entire principle, and it makes the choices predictable. Bismuth quadruple is the recommended salvage for anyone who failed a regimen that didn't already contain bismuth, tetracycline, and metronidazole, which covers almost everyone who failed a clarithromycin or vonoprazan or sequential regimen, and it stays reliable because it dodges clarithromycin and reaches resistant strains through bismuth and metronidazole.

11:48The harder case is the patient who already failed bismuth quadruple, and that's where rifabutin comes in, as a triple with an acid blocker and amoxicillin for ten to fourteen days, also available as a bundled all-in-one capsule. Rifabutin is held in reserve rather than used early for two reasons: cost, and a public-health concern about driving rifabutin resistance in mycobacteria, since it's an important TB drug. Its advantage is that resistance stays exceptionally rare so it still works in heavily pretreated patients; its downsides are cost, rare marrow suppression that warrants blood-count monitoring, and rare eye inflammation, and you exclude active tuberculosis before prescribing so you don't select rifamycin resistance in an unrecognized infection. Levofloxacin triple is a salvage option only when the strain is documented susceptible, because empiric use fails often now and the drug class carries tendon-rupture, QT, and aortic warnings. And vonoprazan-amoxicillin dual is also reasonable salvage, especially after a clarithromycin-containing failure in a non-penicillin-allergic patient, working by the same sustained-suppression mechanism.

12:56After two failures, the right move is susceptibility testing rather than more guessing, either culture from fresh biopsy at a reference lab or molecular testing on stool or biopsy that detects the resistance mutations without growing the organism, and then you give whatever the strain is still susceptible to, at full dose for the full fourteen days. Two traps are worth naming. One is the patient with a severe macrolide allergy and a prior fluoroquinolone adverse event, where both clarithromycin and levofloxacin are off the table for reasons having nothing to do with resistance, so you pull from the low-resistance trio as either bismuth quadruple or rifabutin triple depending on what's been used. The other is the patient labeled penicillin-allergic from a childhood non-blistering rash, where committing to a non-amoxicillin regimen on a weak label closes off the best salvage options, so the right answer is allergy evaluation and de-labeling first.

13:50So the whole H. pylori story rests on a few mechanisms. The organism survives in the mucus by making urease, and that same enzyme is how we detect it and why acid suppression blunts the tests. The location decides the disease: whole-stomach inflammation with low acid gives gastric ulcers and cancer risk, antral inflammation with high acid gives duodenal ulcers. Diagnosis runs through active testing off the acid blocker for one to two weeks and off antibiotics or bismuth for four, serology only for first-time screening and never for cure, and a mandatory test of cure by the same rules. Eradication leads with optimized bismuth quadruple for fourteen days because it sidesteps clarithromycin and leans on the low-resistance trio, with vonoprazan-based dual and triple as the modern alternatives that work through deeper, longer acid suppression, and clarithromycin triple largely retired because resistance has crossed fifteen percent. And salvage runs on one principle: clarithromycin and levofloxacin are one-way streets, amoxicillin, tetracycline, and rifabutin stay reusable, and after two failures susceptibility testing replaces guessing.

14:59The next episode picks up with the ulcers that aren't from H. pylori. NSAID and aspirin ulcers come first, with the prostaglandin mechanism, the risk tiers, and the logic of who needs an acid blocker alongside the NSAID. Then the refractory and idiopathic ulcer workup, including the gastrinoma evaluation with its gastrin and pH pitfalls, and the malignancy and Crohn pieces. And it closes on perforation and penetration, where the omental patch is the standard repair and a contained posterior duodenal ulcer can mimic pancreatitis with a high lipase and no free air.

15:34For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode one of two of chapter six, and I'll see you in the next one.