Erythromycin Before Endoscopy Clears the Stomach and Cuts Repeat Procedures in Acute GI Bleeding, Major Meta-Analysis Finds

When a patient arrives at the emergency department vomiting blood, every minute counts. Acute upper gastrointestinal bleeding is one of the most common and dangerous emergencies in medicine, and the definitive diagnostic and therapeutic weapon is an endoscope threaded down the esophagus to find the source. But there is a fundamental technical problem: the stomach is often filled with blood and clots that obscure the view, making it hard for the endoscopist to see anything at all. A new systematic review and meta-analysis published in Frontiers in Gastroenterology has now pooled the best randomized evidence on a simple pharmacologic fix — giving prokinetic drugs, particularly erythromycin, before the procedure — and the results are striking for what the drugs do achieve, and equally revealing for what they do not.

The research team, led by Mohamad Yousef Alabdallat of Brookdale University Hospital Medical Center and colleagues from institutions spanning the United States, Egypt, and Washington University in St. Louis, conducted a rigorous synthesis following Cochrane Handbook methodology and PRISMA 2020 reporting standards, registered in PROSPERO. They searched five major databases — PubMed/MEDLINE, Scopus, Web of Science, Cochrane Library/CENTRAL, and Embase — from inception to June 30, 2026, identifying 633 records. After removing duplicates and screening titles, abstracts, and full texts, ten randomized controlled trials enrolling 1,077 adults made the final cut. Eight of those trials evaluated erythromycin, while only two evaluated metoclopramide, a detail that turns out to be central to interpreting the entire analysis.

The biological rationale is elegant. Erythromycin is not just an antibiotic; it is also a motilin receptor agonist, meaning it mimics the natural hormone that triggers the stomach’s housekeeping contractions. When given intravenously — typically 250 milligrams or 3 milligrams per kilogram, roughly 20 to 90 minutes before endoscopy — it stimulates powerful antral contractions that sweep retained blood, clots, and gastric contents out of the stomach, clearing the endoscopist’s field of view. Metoclopramide works differently, mainly through dopamine D2 receptor antagonism and serotonergic effects, and its prokinetic activity is generally weaker. Current guidelines from the American College of Gastroenterology and the European Society of Gastrointestinal Endoscopy already support intravenous erythromycin before endoscopy in selected patients with clinically severe or ongoing active bleeding, but the evidence for metoclopramide has remained far less certain.

The pooled numbers tell a clear procedural story. For the primary outcome of adequate gastric emptying or adequate endoscopic visualization, prokinetic therapy produced an odds ratio of 3.68 (95% confidence interval 2.06–6.60; p < 0.0001), meaning patients who received the drugs had more than threefold higher odds of a clear stomach at the time of the procedure. Moderate heterogeneity was present (I² = 59.9%), reflecting variation in drug type, comparator design, use of gastric lavage, bleeding etiology, and outcome definitions, but leave-one-out sensitivity analysis confirmed the effect remained statistically significant after excluding any single study. Continuous endoscopic visualization scores, pooled using standardized mean difference because the trials used different scoring scales ranging from 0–3 to 0–16, also favored prokinetic therapy (SMD 0.47, 95% CI 0.21–0.72; p = 0.0003), though the authors caution that this measure is less clinically interpretable than the dichotomous endpoint.

Perhaps the most practically important finding was the reduction in repeat endoscopy. Across ten comparisons involving 1,034 participants, prokinetic therapy reduced the odds of needing a second-look or repeat procedure by approximately 46% (OR 0.54, 95% CI 0.37–0.79; p = 0.0015), with low and non-significant heterogeneity (I² = 20.8%). This is a big deal in real-world emergency care: repeat endoscopy means additional sedation, additional risk, additional endoscopy suite time, and additional cost. The trials were conducted across diverse geographic settings — France, Switzerland, Saudi Arabia, Iran, Korea, Pakistan, and Thailand — and included both single-center studies and multicenter trials, including Pateron and colleagues’ six-emergency-department French trial and Manupeeraphant and colleagues’ multicenter Thai metoclopramide trial, lending the finding broad external validity.

Length of hospital stay also came out in favor of the prokinetics, with a pooled mean difference of −1.30 days (95% CI −2.11 to −0.49; p = 0.0016) and low heterogeneity (I² = 20.8%). The leave-one-out analysis showed this result remained statistically significant after excluding each study individually. However, the authors are careful to note that hospital stay is influenced by healthcare-system factors — admission practices, discharge criteria, bleeding severity, and local care pathways — that extend well beyond the quality of the endoscopic view, so this finding should be interpreted with appropriate caution.

Here is where the story takes its most important turn: the hard clinical outcomes did not move. All-cause mortality, pooled across six studies and 937 participants, showed a point estimate favoring prokinetic therapy (OR 0.59, 95% CI 0.32–1.09) but the confidence interval crossed the line of no effect (p = 0.0904), with zero heterogeneity. Rebleeding, pooled across five comparisons and 356 participants, showed no difference at all (OR 1.12, 95% CI 0.64–1.97; p = 0.6978). Blood transfusion requirement, pooled across ten comparisons and 752 participants, was not statistically significant under the random-effects model (MD −0.64 units, 95% CI −1.44 to 0.15; p = 0.1126), with substantial heterogeneity (I² = 83.5%) driven largely by a single Iranian trial. Endoscopy duration, pooled across eleven comparisons and 1,036 participants, was similarly non-significant under the random-effects model (MD −1.67 minutes, 95% CI −5.59 to 2.25; p = 0.4046), with very high heterogeneity (I² = 93.8%). Identification of the bleeding source showed a borderline improvement (OR 1.55, 95% CI 1.00–2.40; p = 0.0499) that was not consistently maintained in leave-one-out analysis.

Using the GRADE framework, the certainty of evidence was rated moderate for the two procedural outcomes — adequate visualization and reduced repeat endoscopy — and very low for mortality, rebleeding, transfusion requirement, endoscopy duration, and safety. The authors emphasize that these findings should be interpreted primarily as procedural benefits rather than evidence of improved hard clinical outcomes. Mortality and rebleeding are influenced by a constellation of factors far beyond gastric clearance: bleeding etiology, hemodynamic instability, comorbidity burden, quality of resuscitation, endoscopic hemostasis, pharmacologic therapy, and access to rescue interventions. A clearer stomach helps the endoscopist find and treat the lesion, but it does not change the underlying physiology of the bleed.

Safety deserves its own careful reading. Adverse-event reporting across the included trials was limited and inconsistent. Several trials reported no drug-related or endoscopy-related adverse events, while others did not provide detailed safety data for specific events of interest. No consistent signal of serious harm was identified across the pooled evidence, but the authors are explicit that the trials were not powered to detect uncommon adverse events such as QT prolongation, clinically significant arrhythmias, or extrapyramidal reactions — the known cardiac risks of erythromycin and the neurologic risks of metoclopramide, respectively. Clinicians considering erythromycin should weigh baseline QT risk, electrolyte abnormalities, cardiac comorbidities, and concomitant QT-prolonging medications; for metoclopramide, risk factors for extrapyramidal symptoms should be considered.

The clinical bottom line is selective, not universal. The evidence most strongly supports pre-endoscopic intravenous erythromycin in patients with suspected retained blood, significant clot burden, hematemesis, ongoing active bleeding, or anticipated poor visualization — precisely the scenarios where a clear stomach changes the diagnostic and therapeutic yield of the index procedure. Routine administration to all patients with acute upper gastrointestinal bleeding is not supported by the available evidence, and patients with low-risk presentations or delayed presentation after bleeding has stopped may derive less benefit. Metoclopramide should not be considered a fully evidence-equivalent substitute for erythromycin based on current data, though it may be reasonable where erythromycin is unavailable or contraindicated. The authors call for future head-to-head trials comparing erythromycin and metoclopramide directly, standardized visualization scoring systems, systematic safety reporting, and cost-effectiveness analyses to determine whether selective prokinetic use reduces healthcare resource use by decreasing repeat endoscopy and shortening hospital stay. For now, the message is clear: a cleaner stomach makes for a better endoscopy, and better endoscopy means fewer repeat procedures — even if the drugs cannot claim to save lives on their own.

Subject of Research: Gastroenterology

Article Title: Role of prokinetic drugs prior to endoscopy in acute upper gastrointestinal bleeding: a systematic review and meta-analysis

Article References: Alabdallat, M. Y., Bakain, T., Atasi, A., Oudit, O. A., Ghannam, M., AlRashid, Z., Moghib, K., Obi, E. S., Sarfaraz, M., Omaliko, C. E., & Cheung, D. (2026). Role of prokinetic drugs prior to endoscopy in acute upper gastrointestinal bleeding: a systematic review and meta-analysis. Frontiers in Gastroenterology, 5, Article 1930145. https://doi.org/10.3389/fgstr.2026.1930145

Image Credits: AI Generated

DOI: 10.3389/fgstr.2026.1930145

Keywords: erythromycin, metoclopramide, prokinetic drugs, upper gastrointestinal bleeding, endoscopy, meta-analysis, systematic review, gastric emptying, repeat endoscopy, GRADE, Role, prokinetic