Current guidelines recommend stress ulcer prophylaxis only for critically ill patients who carry specific risk factors for gastrointestinal bleeding, most consistently mechanical ventilation lasting more than 48 hours and coagulopathy. The recommendation sounds straightforward, but behind it lies a decades-long debate about whether the drugs used actually save lives, which agent to choose, and what to do when patients leave the ICU still taking medications they may no longer need. The evidence has shifted considerably in the past few years, and the practical questions surrounding prophylaxis are more interesting than the guideline bullet points alone suggest.
Why Stress Ulcers Happen in the First Place
When someone is critically ill, blood flow to the gut drops. The body redirects circulation to the heart and brain, and the stomach lining suffers. This reduced blood flow, followed by the damage that occurs when circulation returns, breaks down the stomach’s normal defenses: the mucus layer, the bicarbonate that neutralizes acid, and the rapid cell turnover that normally patches small injuries. Endoscopic studies find shallow mucosal damage in the vast majority of ICU patients within the first day of admission.1Europe PMC / Critical Care. Stress ulceration: prevalence, pathology and association with adverse outcomes That sounds alarming, but the distinction that matters is between those microscopic erosions and actual bleeding that changes a patient’s clinical course. Clinically significant bleeding occurs in only about 1 to 4 percent of general ICU patients, and that number has been falling as overall critical care has improved.
Who Actually Needs Prophylaxis
Not every ICU patient benefits. The two strongest and most consistently cited risk factors are prolonged mechanical ventilation (generally defined as more than 48 hours) and coagulopathy.2PubMed. Prevention and treatment of stress ulcers in critically ill patients Guidelines from the Surviving Sepsis Campaign, most recently updated in 2021, include a recommendation for prophylaxis in patients with these risk factors, though the strength of that recommendation was downgraded to conditional because of the potential for harm.3Polish Archives of Internal Medicine. Surviving Sepsis Campaign Guidelines 2021: highlights for the practicing clinician
Beyond those two major risk factors, other conditions commonly cited include severe burns, traumatic brain injury, spinal cord injury, hepatic failure, sepsis, and a history of gastrointestinal bleeding in the preceding year. Traumatic brain injury deserves special mention because the resulting ulcers, sometimes called Cushing’s ulcers, carry a particularly high mortality rate when they bleed. The mechanism involves a surge of gastric acid driven by vagal stimulation and hormonal disruption from brain injury, and the stakes are high enough that prophylaxis in this population is essentially universal in practice.
Patients who are eating normally, breathing on their own, and have no coagulopathy sit at the other end of the spectrum. For them, the baseline risk of clinically meaningful bleeding is so low that the potential harms of prophylactic acid suppression outweigh any benefit. Many older recommendations for prophylaxis were based on studies conducted before modern resuscitation practices, fluid management, and early enteral feeding became standard, which makes them less applicable today.4PubMed. Stress Ulcer Prophylaxis
Does Prophylaxis Actually Prevent Bleeding
The short answer is yes, modestly. The more revealing answer is that it does not appear to save lives. The SUP-ICU trial, a large randomized trial published in the New England Journal of Medicine, assigned ICU patients at risk for gastrointestinal bleeding to receive either pantoprazole (a proton pump inhibitor) or placebo. Clinically important bleeding occurred in about 2.5 percent of the pantoprazole group compared to about 4.2 percent on placebo, a real reduction. But 90-day mortality was virtually identical: roughly 31 percent in both groups.5PubMed. Pantoprazole in Patients at Risk for Gastrointestinal Bleeding in the ICU The one-year follow-up confirmed the same pattern, with mortality around 37 percent in each arm and no difference between them.6PubMed. Pantoprazole in ICU patients at risk for gastrointestinal bleeding-1-year mortality in the SUP-ICU trial
A more recent and even larger trial, the REVISE trial published in 2024, focused specifically on patients receiving invasive mechanical ventilation. Here the bleeding reduction was more dramatic: clinically important upper GI bleeding occurred in 1.0 percent of patients on pantoprazole versus 3.5 percent on placebo. Yet again, 90-day mortality showed no statistically significant difference, with roughly 29 percent dying in the pantoprazole group and 31 percent on placebo.7PubMed. Stress Ulcer Prophylaxis during Invasive Mechanical Ventilation The REVISE trial is significant because it offers the strongest evidence to date that prophylaxis genuinely prevents bleeding in ventilated patients, but even with that benefit, the needle on mortality barely moved. This suggests that while bleeding episodes are prevented, they may not be the kind that ultimately kill patients, or that the harms of the drugs partially offset the gains.
Choosing an Agent
Three classes of drugs are used for stress ulcer prophylaxis: proton pump inhibitors like pantoprazole and omeprazole, histamine-2 receptor antagonists like ranitidine and famotidine, and sucralfate, which works locally by forming a protective coating over damaged mucosa rather than suppressing acid. Each comes with tradeoffs.
A network meta-analysis of randomized trials found that PPIs were the most effective at preventing clinically important bleeding, outperforming H2 receptor antagonists, sucralfate, and placebo.8PubMed Central. Efficacy and safety of stress ulcer prophylaxis in critically ill patients: a network meta-analysis of randomized trials However, PPIs also carried a higher risk of pneumonia compared to sucralfate and H2 blockers. The tradeoff is real: you get better ulcer prevention at the cost of more lung infections.
When comparing PPIs directly to H2 receptor antagonists, an updated meta-analysis pooling data from more than 28,000 patients found a small but concerning increase in mortality with PPIs: roughly 9 additional deaths per 1,000 patients treated, with the confidence interval just touching the line of no difference.9PubMed. Proton Pump Inhibitors Versus Histamine-2 Receptor Antagonists Likely Increase Mortality in Critical Care: An Updated Meta-Analysis Whether that signal represents a true harm or statistical noise remains debated, but it has pushed some intensivists to reconsider whether H2 blockers deserve a comeback. In cardiac surgery patients, a subanalysis from the large PEPTIC trial found no differences in mortality, bleeding, or Clostridium difficile infection between PPIs and H2 blockers.10PubMed Central. Efficacy and safety of proton pump inhibitors versus histamine-2 receptor blockers in the cardiac surgical population: insights from the PEPTIC trial
Sucralfate is the odd member of the trio. It does not suppress acid at all, instead physically shielding the stomach lining. A meta-analysis comparing sucralfate to H2 receptor antagonists found no significant difference in bleeding prevention but a statistically significant reduction in ICU-acquired pneumonia with sucralfate.11PubMed. Sucralfate versus histamine 2 receptor antagonists for stress ulcer prophylaxis in adult critically ill patients: A meta-analysis and trial sequential analysis of randomized trials A separate study in intubated patients found that ventilator-associated pneumonia rates were roughly a third as high with sucralfate compared to PPI or H2 blocker groups, and the bacteria involved in those infections were more commonly dangerous hospital-acquired pathogens in the acid-suppressing drug groups.12PubMed. Pneumonia prevention in intubated patients given sucralfate versus proton-pump inhibitors and/or histamine II receptor blockers The mechanism makes intuitive sense: stomach acid normally kills swallowed bacteria, and drugs that suppress acid let those bacteria survive, colonize the stomach, and get aspirated into the lungs.
Adverse Effects Worth Knowing About
Pneumonia is the most studied side effect of acid-suppressing prophylaxis, but it is not the only concern. Clostridium difficile infection, a potentially severe and recurrent gut infection, is more common in patients receiving PPIs than in those on H2 blockers. One study in critically ill patients found C. difficile rates of 6.7 percent with PPIs versus 1.8 percent with H2 receptor antagonists, with PPI use emerging as an independent risk factor.13Gut and Liver. Risk of Clostridium difficile Infection with the Use of a Proton Pump Inhibitor for Stress Ulcer Prophylaxis in Critically Ill Patients A systematic review of eight studies confirmed a statistically significant association between PPI use and C. difficile across a range of risk levels.14PubMed Central. The Positive Association between Proton Pump Inhibitors and Clostridium Difficile Infection
In patients with acute myocardial infarction on mechanical ventilation, H2 blockers were associated with lower rates of ventilator-associated pneumonia compared to PPIs, even after adjusting for other variables.15JACC: Advances. Stress Ulcer Prophylaxis in Mechanically Ventilated Patients With Acute Myocardial Infarction This is relevant because heart attack patients already face enormous physiological stress, and adding a preventable pneumonia to that burden matters.
Drug interactions also deserve attention in the ICU. Critically ill patients typically receive many medications simultaneously, and PPIs are metabolized primarily by the liver enzyme CYP2C19. Genetic variation in this enzyme affects how quickly different people break down the drug, ranging from ultrarapid metabolizers who clear it so fast it may not work well, to poor metabolizers in whom it accumulates.16PubMed Central. CYP2C19 Gene Profiling as a Tool for Personalized Stress Ulcer Prophylaxis With Proton Pump Inhibitors in Critically Ill Patients – Recommendations Proposal On top of genetic variability, the frequent presence of kidney and liver impairment in ICU patients further complicates dosing and raises the risk of interactions with other drugs processed through the same pathways.
Enteral Nutrition as a Protective Factor
One of the more intriguing questions in this area is whether feeding patients through the gut reduces the need for pharmacological prophylaxis. The thinking is straightforward: food in the stomach buffers acid, stimulates blood flow to the gut lining, and supports the mucosal barrier. Animal studies generally support this, and indirect human data from large stress ulcer prophylaxis trials suggest that enteral nutrition is associated with lower rates of gastrointestinal bleeding and lower mortality.17PubMed. Does enteral nutrition protect against stress ulceration in the critically ill? The catch is that no randomized trial has directly tested enteral feeding against no feeding with bleeding as the primary outcome, so the evidence remains indirect. There is also a possible interaction between enteral nutrition and pharmacological prophylaxis that researchers are still working out. In practice, many clinicians consider patients who are receiving enteral feeds at a reasonable rate to be at lower risk, but how much that should influence the decision to withhold or stop acid suppression remains an open question.
The Overprescribing Problem
If the risk factors for stress ulcer bleeding are well defined and the guidelines are clear about who qualifies, you might expect prescribing to be reasonably targeted. It isn’t. Study after study shows that acid-suppressing drugs are prescribed to patients who do not meet criteria, continued long after the risk has passed, and sent home with patients at discharge even when no indication exists.
In one study of patients over 65 admitted to general medical wards, about one in five received stress ulcer prophylaxis, and the prescription was inappropriate in over 96 percent of those cases. Nearly 29 percent of those patients were discharged still taking the drug with no justified reason.18European Geriatric Medicine. Appropriateness of stress ulcer prophylaxis among older adults admitted to general medical wards in a university hospital A separate prospective study found that stress ulcer prophylaxis was used inappropriately in about 80 percent of hospital days and continued inappropriately in over 77 percent of cases at discharge.19PubMed Central. Variables Associated with Adherence to Stress Ulcer Prophylaxis in Patients Admitted to the General Hospital Wards: A Prospective Study
The ICU-to-discharge pipeline is a particularly leaky one. Among patients started on PPIs specifically for stress ulcer prophylaxis in the ICU, about half were still taking the drug at hospital discharge, and only 10 percent of those had a documented reason to continue.20Critical Care and Resuscitation. Incidence and cost of stress ulcer prophylaxis after discharge from the intensive care unit: a retrospective study A similar study found inappropriate continuation at discharge in roughly 45 percent of patients, with H2 blockers actually continued inappropriately more often than PPIs, possibly because clinicians perceive them as safer and give less thought to stopping them.21BMJ Open. Prevalence and factors associated with inappropriate continuation of stress ulcer prophylaxis at discharge
What makes this problematic beyond the obvious waste is that patients often continue these drugs in the community for months or years. An outpatient physician sees the PPI on the medication list, assumes it was started for a good reason, and keeps prescribing it. The patient develops a vague sense that they “need” it. Meanwhile, the ongoing exposure to acid suppression carries cumulative risks including C. difficile, reduced absorption of calcium and magnesium, and the possibility of rebound acid hypersecretion when the drug is eventually stopped.
Pharmacist-Led Stewardship Programs
The most effective response to overprescribing has been embedding pharmacists into the review process. One program that implemented clinical pharmacist oversight of stress ulcer prophylaxis prescriptions achieved a 58 percent reduction in inappropriate prophylaxis days in the ICU and an 84 percent reduction on general wards, cutting total prophylaxis-related inpatient drug costs from about $20,000 to $3,300 in the study period, with estimated annual savings exceeding $200,000.22PubMed. Impact of a clinical pharmacist stress ulcer prophylaxis management program on inappropriate use in hospitalized patients These results highlight that the problem is not physician ignorance of guidelines but rather the absence of a systematic checkpoint. When someone is explicitly tasked with reviewing whether the drug is still needed, it gets stopped. Without that checkpoint, inertia takes over.
Prophylaxis in Critically Ill Children
Pediatric practice has its own set of challenges. A multicenter survey found that stress ulcer prophylaxis was used in about 78 percent of pediatric ICU patients, with mechanical ventilation and “routine use” cited as the two most common reasons in almost equal proportion. Only one of the participating units had a formal protocol for when to prescribe it, and ranitidine was the drug of choice in the vast majority of cases.23Jornal de Pediatria. Stress ulcer prophylaxis in pediatric intensive care units
The pediatric evidence base is thinner than in adults, but what exists points in a similar direction. A study in children with mild to moderate organ dysfunction found that omeprazole did not appear to prevent gastrointestinal bleeding while it did increase the risk of central line-associated bloodstream infections, leading the authors to recommend restricting prophylaxis to mechanically ventilated children.24Anales de PediatrÃa (English Edition). Stress ulcer prophylaxis for critically ill children: routine use needs to be re-examined The parallel to the adult data is striking: outside the highest-risk subgroup, the drug may cause more harm than it prevents. The widespread “routine use” justification in pediatric units echoes the same prescribing inertia seen in adult hospitals.
Genetic Variation and PPI Dosing
An emerging area that has not yet changed bedside practice but could reshape future guidelines involves the genetics of drug metabolism. PPIs are broken down in the liver primarily by the CYP2C19 enzyme, and the activity of that enzyme varies widely between individuals based on genetic variants. Ultrarapid metabolizers clear the drug so quickly that standard doses may provide inadequate acid suppression, while poor metabolizers accumulate the drug and face greater exposure to its side effects.25PubMed Central. CYP2C19 Gene Profiling as a Tool for Personalized Stress Ulcer Prophylaxis With Proton Pump Inhibitors in Critically Ill Patients – Recommendations Proposal The distribution of these genetic variants differs across ethnic populations, meaning a flat one-dose-fits-all approach will predictably over-treat some patients and under-treat others. Pharmacogenomic testing is already routine for some other drugs in critical care, and researchers have proposed that incorporating CYP2C19 profiling into PPI prescribing could improve both effectiveness and safety. For now, this remains a research recommendation rather than a guideline requirement, but it represents one of the more plausible near-term changes to how prophylaxis is individualized.