Peptic Ulcer Disease and H. pylori: Peptic Ulcer Disease and H. pylori: Ulcer Itself

0:00Welcome to Board Pearls. This is episode two of two of the Peptic Ulcer Disease and H. pylori chapter, in the Stomach and Small Bowel Disorders module. This episode is the ulcer itself. A man who bled from a nonsteroidal anti-inflammatory drug, NSAID, ulcer and still needs an anti-inflammatory shows how NSAIDs injure the stomach even when they're not swallowed, how risk factors add up to decide protection, which protection works, and what changes when aspirin is added. A woman with an ulcer that won't heal shows what is usually missed, why a gastric ulcer is biopsied again, and what's left before calling it idiopathic. A man with ulcers far down the duodenum shows the gastrinoma pattern. A man whose ulcer went through the wall shows how perforation is confirmed, why the clock matters, and what the surgeon deliberately leaves out. And a woman with a sealed perforation shows how it passes for pancreatitis.

0:56Here's the first patient. A seventy-year-old man with severe knee osteoarthritis took ibuprofen eight hundred milligrams three times a day for a year. Three months ago he had melena, and endoscopy showed a bleeding gastric ulcer, treated endoscopically. The ulcer has healed on a proton pump inhibitor, PPI. He stopped the ibuprofen, but his knees are so painful he can barely walk, and his orthopedist says he needs an anti-inflammatory until his knee replacement next year. He takes no aspirin or anticoagulant. He asks whether he could use a gel or a suppository instead, since those wouldn't touch his stomach.

1:34He had an NSAID-induced gastric ulcer that bled, and he still needs an NSAID, so the task is to protect him. How NSAIDs cause ulcers answers his question and explains which protections work.

1:46The main injury is systemic. NSAIDs block cyclooxygenase one in the stomach and duodenum, which depletes the prostaglandins that defend the lining. Those prostaglandins drive mucus secretion, bicarbonate secretion into the mucus layer, and blood flow to the mucosa, which delivers oxygen and carries away acid that has diffused back in. Without them, the mucus thins, the bicarbonate barrier falls, perfusion drops, and acid injures the lining, with erosions first, then ulcers when injury outruns repair.

2:21Because that injury travels through the blood, changing how the drug is given doesn't protect him. An NSAID given by injection, by suppository, or topically can still cause ulcers. So the answer to his question is no.

2:39There's a second, local injury where the pill touches the lining. The NSAID enters the surface cells, becomes trapped there as a charged ion, and damages them. Aspirin does both, and it also permanently disables cyclooxygenase in platelets, which is why even a low dose raises bleeding risk for the life of the platelets.

2:58How much protection he needs depends on how high his risk is, because risk factors add up and the total decides. The established risk factors are age over sixty-five; a prior ulcer, especially one that bled or perforated; a high NSAID dose; using more than one NSAID; corticosteroids; anticoagulants; antiplatelet drugs such as aspirin and clopidogrel; selective serotonin reuptake inhibitors, SSRIs; and H. pylori infection. Each roughly doubles the risk, and they stack.

3:33Timing is a risk factor too. The first thirty to ninety days are the riskiest, before the lining adapts, so a high-risk patient needs protection from the first dose.

3:44Risk is graded in three tiers. High risk is a prior complicated ulcer or more than two risk factors. Moderate risk is one or two. Low risk is none. He's over sixty-five and has a prior bleeding ulcer, a complicated ulcer, which alone makes him high risk.

4:01For a high-risk patient, the protection that works is a PPI. A high-risk patient should never take a non-selective NSAID without a PPI, and the PPI continues as long as the NSAID does. A moderate-risk patient gets a PPI if the NSAID is long-term. A low-risk patient without other reasons for acid suppression generally doesn't need one.

4:25The other protective drugs fall short of a PPI. Misoprostol, a prostaglandin E1 analog, replaces exactly what the NSAID removed. But it causes dose-related diarrhea and cramping that many patients won't tolerate, and it can cause miscarriage, so it's avoided in women who could become pregnant, and it's rarely chosen over a PPI now. Standard-dose H2 blockers don't prevent NSAID ulcers adequately. And vonoprazan isn't approved for preventing or treating NSAID ulcers.

4:55The choice of NSAID adds a second layer of protection. Celecoxib selectively blocks cyclooxygenase two and spares cyclooxygenase one, and with it the stomach's protective prostaglandins, so it causes fewer gastrointestinal complications than non-selective NSAIDs. Its cost is cardiovascular. An earlier drug in the class, rofecoxib, was withdrawn in two thousand four because it raised the risk of heart attack and stroke. Celecoxib carries a smaller version of the same signal, so it's weighed against heart risk every time.

5:28For a patient who has bled from an NSAID ulcer and still needs an anti-inflammatory, the answer is a cyclooxygenase two selective agent together with a PPI, both for as long as the anti-inflammatory is used. Celecoxib lowers the risk but doesn't remove it, and his history puts him in the top tier. He has no heart disease, so he starts celecoxib with a daily PPI.

5:51He also needs an H. pylori test, even though his ulcer has healed. The bacterium injures through inflammation and the drug through prostaglandin loss, and the effects are at least additive, so with both present, bleeding ulcer risk is roughly six times that with neither. So testing and treating is recommended for anyone on long-term NSAIDs or low-dose aspirin. Protection is greatest when the infection is eradicated before a chronic NSAID starts, and it's still worth doing in someone already taking one. The best time to test is the visit where the long-term NSAID or aspirin is first prescribed.

6:26His PPI shapes how he's tested. He's been on a PPI since the bleed, so the test must be done off PPI for one to two weeks, or by histology on biopsies. His biopsies at the time of the bleed showed no H. pylori on histology, and a stool antigen after a two-week PPI hold is negative.

6:44A positive result wouldn't have replaced his PPI. Eradication removes one risk factor, but the PPI decision follows his risk tier, and a high-risk patient cured of H. pylori is still a high-risk patient on an NSAID.

7:00Six months later the plan breaks. He has a myocardial infarction and is started on aspirin eighty-one milligrams daily for life, plus clopidogrel for a stent.

7:10His new drugs undo the logic of his plan. Aspirin is a non-selective cyclooxygenase inhibitor, so it blocks cyclooxygenase one in his stomach again and restores the risk celecoxib was chosen to avoid. Celecoxib plus aspirin largely loses its advantage over a non-selective NSAID. With clopidogrel, he's on two antiplatelet drugs with a prior bleeding ulcer. And celecoxib's cardiovascular risk now matters in a man who's just had a heart attack.

7:41So the plan changes, starting with the drugs he can't stop. His aspirin is for secondary prevention and continues at the lowest effective dose, generally eighty-one milligrams daily. His PPI is now essential and continues indefinitely.

7:56The PPI isn't avoided because of clopidogrel. The two share a liver enzyme, so a PPI could in theory reduce clopidogrel's activation, but trials haven't shown more cardiovascular events, and pantoprazole or rabeprazole interact least if there's concern. And the anti-inflammatory is reconsidered with his cardiologist and orthopedist, because the decision now balances his knees, his heart, and his stomach. The best answer may be to stop the NSAID until his knee replacement.

8:25The second patient's ulcer isn't healing, and that's a signal something was missed. A sixty-three-year-old woman had a one-and-a-half centimeter ulcer on the lesser curvature of the stomach, found during evaluation for iron deficiency. The biopsies showed benign ulceration. A rapid urease test on those biopsies was negative, though she was taking omeprazole at the time. She was told to continue omeprazole twice daily. Repeat endoscopy at twelve weeks shows the ulcer still there, barely smaller. She says she takes no other medications.

8:56A non-healing ulcer means one of the standard exclusions was probably wrong, the patient isn't taking the treatment correctly, or the ulcer isn't an ordinary peptic ulcer. So the PPI isn't simply escalated. Instead, what was missed is checked, commonest misses first.

9:11The commonest miss is a false-negative H. pylori test, and hers can't be trusted. Most patients with a non-healing ulcer were tested on a PPI, as she was, and PPIs make the urea breath test, stool antigen, and rapid urease test falsely negative. She needs a properly done test. That means either off the PPI for one to two weeks and off antibiotics and bismuth for four weeks, then a breath test or stool antigen. Or, since she has an open ulcer and needs her PPI, it means biopsies from the body of the stomach for histology with a special stain, such as Giemsa or immunohistochemistry.

9:46The next miss is an NSAID or aspirin she hasn't mentioned. Patients often don't count over-the-counter ibuprofen, naproxen, low-dose aspirin, or combination cold remedies as medications. So instead of asking about "any medications," she's asked specifically about every over-the-counter pain reliever, every cold or flu product, every herbal preparation, and every pill taken only now and then. Hidden NSAID use explains many ulcers that look idiopathic.

10:16Adherence is checked too, because a PPI taken wrongly barely works. Pharmacy refill records are checked, and she's asked exactly when she takes her omeprazole, because a PPI taken at bedtime or with food barely works. It belongs thirty to sixty minutes before breakfast, and before dinner when twice daily. After H. pylori treatment, she'd also be asked whether every dose was taken, with the bismuth quadruple components taken together with food.

10:45The check that's never skipped is for cancer. A gastric adenocarcinoma can look like a benign ulcer, fail to heal, and be missed on the first biopsies. So every non-healing gastric ulcer is biopsied again at follow-up, with multiple biopsies from both the rim and the base, because sampling one area can miss a tumor in the other. Negative first biopsies never prove a gastric ulcer is benign.

11:08Lymphoma is the other malignancy that hides this way. Lymphoma, mucosa-associated lymphoid tissue, MALT, or diffuse large B-cell, can present as a non-healing ulcer, so when it's a possibility the biopsies go for flow cytometry or immunohistochemistry too. Her iron deficiency fits slow bleeding from either a benign ulcer or a cancer, so it doesn't settle the question.

11:32These four checks are done together, at the same follow-up endoscopy and visit, not one after another over months, because the cheap ones are the most common and the dangerous one can't wait. Biopsies from her gastric body show no H. pylori on special stains. Biopsies from the rim and base show benign ulceration. And when asked specifically, she says she takes two naproxen tablets most nights for back pain. She didn't think they counted.

11:59Doubling her omeprazole would have treated the acid while the naproxen kept injuring her stomach every night. She stops the naproxen, confirms she takes the omeprazole before meals, and gets a different plan for her back pain. The ulcer heals by the next endoscopy. Had the repeat biopsies shown signet ring cells, her ulcer would have been a gastric cancer missed the first time, which is why the repeat biopsy is never skipped.

12:24In a patient whose ulcer still doesn't heal after H. pylori, NSAIDs, adherence, and cancer are excluded, the list widens to rarer causes, each with a clue. Crohn disease of the stomach or duodenum gives ulcers that fail PPI therapy, with multiple lesions, an aphthous pattern, ileal disease on cross-sectional imaging, granulomas on biopsy, or extraintestinal features. Then the workup becomes an inflammatory bowel disease evaluation rather than more acid suppression.

12:55The other rare causes are found by their setting. Vasculitis, such as polyarteritis nodosa or granulomatosis with polyangiitis, causes gastric ulcers as part of systemic disease, recognized by systemic features and vasculitis on biopsy of the ulcer rim. Cytomegalovirus causes ulcers in human immunodeficiency virus, HIV, infection with a low CD4, cluster of differentiation four, count, in transplant recipients, and in patients on chemotherapy. Deep biopsies show owl's-eye inclusions in endothelial and stromal cells.

13:31Two more causes come from the tissue and the drug list. Eosinophilic gastritis is diagnosed by more than thirty eosinophils per high-power field on multiple biopsies, in the right clinical setting. And other drugs injure the lining: potassium chloride, bisphosphonates, sirolimus, mycophenolate, and cocaine, which injures through vasoconstriction.

13:54The third patient shows that one cause has its own pattern. A thirty-eight-year-old man has several ulcers, including one in the third part of the duodenum and one in the jejunum, plus severe erosive esophagitis and chronic watery diarrhea. His ulcers keep returning despite a PPI and a negative H. pylori test. His father had a parathyroid tumor.

14:17That's Zollinger-Ellison syndrome. Multiple ulcers, ulcers beyond the duodenal bulb, recurrence despite treatment, severe esophagitis, secretory diarrhea, and a personal or family history of multiple endocrine neoplasia type one, MEN1, point to a gastrin-secreting neuroendocrine tumor in the duodenum or pancreas. The gastrin drives the parietal cells to make enormous amounts of acid, which ulcerates far down the small bowel, burns the esophagus, and causes diarrhea.

14:48His screening test has to be done off acid suppression. The screen is a fasting gastrin with a simultaneous gastric pH, interpreted off PPI, because PPIs raise gastrin on their own. A gastrin over one thousand picograms per milliliter with a pH under two confirms it. A gastrin of one hundred to one thousand needs a secretin test, where a rise of more than one hundred twenty above baseline confirms it, because secretin normally suppresses gastrin and a gastrinoma paradoxically releases more. Localization with somatostatin receptor imaging and endoscopic ultrasound, and screening for MEN1, follow.

15:25An ulcer is called idiopathic only after all of that. H. pylori must be negative on proper retesting, hidden NSAIDs excluded, malignancy excluded on repeat biopsy, and the rare causes considered. Truly idiopathic ulcers exist, and they recur and cause complications more often than H. pylori ulcers. They're treated with long-term PPI maintenance, follow-up, and a low threshold to investigate again.

15:53The fourth patient's ulcer has gone through the wall. A fifty-eight-year-old man who smokes and takes naproxen daily for back pain was watching television when a sudden, severe pain hit his upper abdomen. He can name the exact minute it started. Within two hours the pain spread across his whole abdomen. He arrives lying perfectly still, with a rigid, board-like abdomen and rebound tenderness. His pulse is one hundred twenty, his blood pressure one hundred over sixty, his temperature thirty-eight point four, and his white count eighteen thousand.

16:25He has a free perforation of a peptic ulcer. A free perforation means the ulcer has eroded through every layer of the wall, and acid, pepsin, food, bile, and mouth bacteria have spilled into the peritoneal cavity. Acid and pepsin cause an immediate chemical peritonitis, and the bacteria turn it into a bacterial peritonitis within hours.

16:47That explains his story. The pain has a precise onset because it began when the wall gave way. It spread across his abdomen quickly because the spill spreads. He lies still because any movement shifts inflamed peritoneum. His fever, tachycardia, low blood pressure, and high white count are the peritonitis becoming sepsis. Daily naproxen and smoking explain the ulcer.

17:11It's confirmed by finding free air, and the best test is CT. An upright chest x-ray shows air under the diaphragm in about seventy to eighty percent of perforations and is fast when computed tomography, CT, isn't immediately available, but a normal film doesn't exclude perforation. CT shows free air when the chest film doesn't, locates the perforation, shows whether the leak is free or contained, and excludes look-alike diagnoses. His CT shows free intraperitoneal air and fluid around a perforation in the front wall of the duodenal bulb.

17:45Time to surgery drives mortality, so resuscitation and preparation for the operating room happen together. He gets intravenous crystalloid. He gets broad-spectrum antibiotics covering gram-negative gut bacteria, anaerobes, and mouth flora, such as piperacillin-tazobactam, or ceftriaxone with metronidazole. A nasogastric tube decompresses his stomach so less leaks through the hole. An intravenous PPI infusion cuts the acid.

18:11Surgery is called right away, because delay kills. He goes to the operating room as soon as rapid resuscitation is done. Delay of more than twenty-four hours from perforation to repair sharply increases mortality, especially in older patients and those who arrive septic.

18:26The operation is usually a Graham omental patch, because the ulcer's edges won't hold stitches. The edges of a perforated ulcer are swollen and friable, and sutures through them would tear out. So the surgeon brings a piece of living omentum over the hole and sutures it in place. Omentum is mobile, has its own blood supply, and seals inflammatory leaks with living tissue. In a modified Graham patch, a small perforation whose edges will hold sutures is closed first and then covered with omentum. Either can be done open or laparoscopically, and laparoscopy shortens recovery in stable patients.

19:02The surgeon deliberately leaves out definitive ulcer surgery. Vagotomy and antrectomy were once added to prevent recurrence. They aren't anymore, because PPIs and H. pylori eradication treat the underlying disease afterward, and a bigger operation in an unstable patient with peritonitis adds risk without enough benefit. His surgeon does a laparoscopic Graham patch the same evening, about six hours after his pain began.

19:27Afterward, every cause of his ulcer is addressed. He continues a high-dose intravenous PPI, then an oral one. He's tested for H. pylori once stable, with eradication if positive and a test of cure. He's strongly encouraged to stop smoking, and a follow-up endoscopy confirms healing.

19:46His NSAID is stopped in a way that sticks. He stops naproxen for good. Over-the-counter NSAIDs are easy to restart without mentioning them, as the second patient showed, so every NSAID, whatever the brand and however it's taken, is named as off-limits, and he leaves with a different plan for his back pain.

20:07The last patient shows that an ulcer can go through the wall without spilling. A sixty-six-year-old woman with a long history of duodenal ulcer has two weeks of constant epigastric pain boring through to her back. Her lipase is three times the upper limit of normal, and she was admitted with acute pancreatitis. After five days of standard treatment, she isn't improving. CT shows no free air, and a deep ulcer in the back wall of the duodenal bulb reaching into the head of the pancreas.

20:35That's penetration, an ulcer that has eroded into a neighboring organ, which seals the hole. A posterior duodenal ulcer classically penetrates the pancreas, which becomes inflamed where the ulcer meets it. So the pain bores to the back, the lipase rises, and the patient is labeled with pancreatitis. The clues that point elsewhere are a known ulcer history, constant boring pain, no free air, and failure to improve on pancreatitis treatment.

21:04Because the leak is sealed, penetration is usually managed medically. That means a high-dose PPI, H. pylori eradication, stopping NSAIDs, and close observation. Ulcers can also penetrate the liver, biliary tree, gallbladder, or colon. A gastrocolic fistula causes vomiting of feculent material, and it and other complex penetrations need surgery.

21:32The smallest perforations are made at endoscopy, and they can often be closed the same way. They happen, for example, during mucosal resection or submucosal dissection. When one is recognized at the moment it happens, in a stable patient without diffuse peritonitis, it can often be closed endoscopically. Through-the-scope clips close perforations under one centimeter, an over-the-scope clip closes larger ones up to two or three centimeters, and endoscopic suturing closes larger full-thickness defects.

22:04Endoscopic closure has strict conditions. The patient must be stable without diffuse peritonitis, because peritonitis needs surgical washout. The perforation should be recognized early, before contamination spreads. The endoscopist must be experienced, and surgery must be immediately available. Selected stable patients with small contained ulcer perforations can be managed this way too. Whatever the plan, surgery is consulted for every perforation, because a patient managed medically or endoscopically can deteriorate quickly.

22:35Here are the traps the boards set in these patients. A suppository, injection, or gel doesn't avoid gastric injury. One prior bleed makes a patient high risk on its own. A high-risk patient doesn't get a non-selective NSAID without a PPI. An H2 blocker or misoprostol isn't a substitute for a PPI. Celecoxib isn't given without a PPI after an ulcer bleed, and it doesn't keep protecting once aspirin is added. H. pylori testing isn't skipped in a long-term NSAID or aspirin user. And a PPI isn't avoided because of clopidogrel.

23:14The refractory ulcer traps follow. A PPI isn't escalated without asking why an ulcer didn't heal. A negative H. pylori test done on a PPI isn't accepted. Over-the-counter NSAIDs aren't missed. A non-healing gastric ulcer isn't assumed benign because the first biopsies were, and biopsies come from both the base and the rim. Gastrin isn't measured on a PPI. And an ulcer isn't called idiopathic too early.

23:44The perforation traps close the list. Surgery for free perforation doesn't wait beyond rapid resuscitation. A normal chest x-ray doesn't exclude perforation. Vagotomy or antrectomy isn't added at the emergency operation. A penetrating posterior duodenal ulcer isn't treated as plain pancreatitis. And endoscopic closure isn't attempted in an unstable patient with diffuse peritonitis.

24:09So, start to finish. NSAIDs cause ulcers mainly by depleting cyclooxygenase one prostaglandins, however they're given. A prior complicated ulcer or more than two risk factors is high risk, and high-risk patients and moderate-risk patients on long-term NSAIDs need a PPI for as long as the NSAID continues.

24:33After an NSAID ulcer bleed, a patient who still needs an anti-inflammatory gets celecoxib plus a PPI, but adding aspirin cancels much of celecoxib's benefit. Test and treat H. pylori in every long-term NSAID or aspirin user, and keep the PPI with clopidogrel.

24:50A non-healing ulcer means something was missed. Retest H. pylori properly, ask specifically about over-the-counter NSAIDs, check adherence and PPI timing, and biopsy every non-healing gastric ulcer again at the rim and base. Then consider Crohn disease, vasculitis, cytomegalovirus, eosinophilic gastritis, other drugs, and Zollinger-Ellison syndrome.

25:15Free perforation is sudden, precisely timed pain becoming generalized peritonitis, with free air best shown on CT. Resuscitate, give antibiotics and an intravenous PPI, and repair promptly with a Graham patch, without definitive ulcer surgery. Penetration mimics pancreatitis and is managed medically. Small perforations recognized at endoscopy in stable patients can be closed with clips or suturing, with surgery on standby.

25:42The next chapter covers the rest of the stomach: atrophic gastritis and the H. pylori cancer cascade, Zollinger-Ellison syndrome, gastric adenocarcinoma, lymphoma, neuroendocrine tumors, and gastrointestinal stromal tumors, GIST.

25:59For 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 two of two of chapter six, and I'll see you in the next one.