Is it safe to use a glucagon‑like peptide‑1 (GLP‑1) receptor agonist in a patient with an esophageal ulcer or Barrett’s esophagus?

GLP-1 Receptor Agonists in Patients with Esophageal Ulcers or Barrett's Esophagus

GLP-1 receptor agonists should be used with significant caution—and potentially avoided—in patients with esophageal ulcers or Barrett's esophagus, particularly short-acting formulations, due to their association with delayed gastric emptying, increased gastroesophageal reflux complications, and documented aspiration risk even after prolonged fasting.

Mechanism of Concern: Delayed Gastric Emptying and Reflux Risk

GLP-1 receptor agonists delay gastric emptying by inhibiting gastric peristalsis and increasing pyloric tone through vagal pathways, leading to increased fasting gastric volumes and prolonged gastric residence time 1. This delayed emptying persists even with long-acting formulations like semaglutide, as demonstrated by scintigraphy studies showing retained gastric contents in 24.2% of semaglutide users versus 5.1% of controls despite 10-14 day discontinuation and 12-hour fasting 1, 2.

The clinical significance is underscored by a documented case of a 42-year-old patient with Barrett's esophagus who, despite fasting for 18 hours before upper endoscopy, had substantial gastric contents requiring suctioning before intubation, ultimately resulting in intraoperative pulmonary aspiration requiring bronchoscopic removal of food remains from the trachea and bronchi 3.

Evidence Linking GLP-1 Agonists to Esophageal Complications

Short-acting GLP-1 receptor agonists (half-life ≤1 day) are specifically associated with increased risk of erosive reflux disease and Barrett's esophagus complications. In a propensity-matched cohort study of 177,666 patients in each group, short-acting GLP-1 RAs demonstrated:

• Erosive reflux disease: HR 1.215 (95% CI 1.111-1.328) 4
• Esophageal stricture: HR 1.284 (95% CI 1.135-1.453) 4
• Barrett's without dysplasia: HR 1.372 (95% CI 1.217-1.546) 4
• Barrett's with dysplasia: HR 1.505 (95% CI 1.164-1.946) 4

Critically, long-acting GLP-1 RAs (half-life ≥5 days) showed no increased risk for these complications (HR 0.994; 95% CI 0.924-1.069 for erosive reflux disease) 4. This differential risk appears related to the more pronounced gastric emptying delay with short-acting agents: lixisenatide delayed gastric emptying half-time by 52 minutes versus 25 minutes with liraglutide 5.

Clinical Decision Algorithm

For Patients with Active Esophageal Ulcers:

Avoid GLP-1 receptor agonists entirely until ulcers are healed. The combination of delayed gastric emptying, increased gastric volumes, and documented aspiration risk creates an unacceptable safety profile in this population 1, 4, 3.

For Patients with Barrett's Esophagus:

If GLP-1 therapy is deemed essential (e.g., established cardiovascular disease requiring proven cardiovascular benefit):

1. Choose long-acting formulations exclusively: Semaglutide or dulaglutide, which show no increased risk of Barrett's complications 4
2. Avoid short-acting agents: Liraglutide, lixisenatide, and exenatide are contraindicated given their 1.5-fold increased risk of Barrett's with dysplasia 4
3. Implement aggressive reflux management: Proton pump inhibitors should be optimized, with consideration of twice-daily dosing 4
4. Establish enhanced surveillance: More frequent endoscopic monitoring may be warranted, though specific intervals are not established in the literature 4, 3

Perioperative Considerations:

For any patient with Barrett's esophagus on GLP-1 therapy requiring endoscopy or surgery:

• Discontinue semaglutide/tirzepatide for at least 3 weeks (three half-lives) before elective procedures 6, 2
• Discontinue liraglutide for at least 3 days before procedures 6
• Consider gastric ultrasound pre-operatively to assess residual gastric contents, as standard fasting periods are insufficient 1, 2
• Extended fasting periods (>18 hours) do not eliminate aspiration risk in GLP-1 users 3, 2

Gastrointestinal Effects Beyond Reflux

While one study found no significant difference in the number of reflux episodes (33.7 ± 4.1 vs. 40.1 ± 5.3 for lixisenatide vs. liraglutide) or DeMeester score during 24-hour pH monitoring 5, this finding is contradicted by the large population-level study showing clear increased risk of erosive complications with short-acting agents 4. The discrepancy likely reflects the difference between physiologic reflux episodes and clinically significant erosive disease.

Notably, GLP-1 RAs reduce gastric acid secretion by approximately 20.7% 5, which theoretically could be protective, but this effect is clearly insufficient to offset the mechanical consequences of delayed gastric emptying in vulnerable esophageal tissue 4.

Common Pitfalls to Avoid

• Do not assume that long-acting and short-acting GLP-1 RAs carry equivalent esophageal risk—the evidence clearly differentiates them 4
• Do not rely on standard fasting protocols for procedures in GLP-1 users with Barrett's esophagus; the documented case of aspiration after 18-hour fasting demonstrates this is inadequate 3
• Do not prescribe liraglutide for weight loss in patients with Barrett's esophagus when tirzepatide or semaglutide would provide superior weight loss without the increased Barrett's complication risk 4
• Do not overlook the aspiration risk even in outpatient endoscopy settings; the case report involved a routine surveillance endoscopy 3

Alternative Therapeutic Approaches

For patients with esophageal ulcers or Barrett's esophagus requiring glucose-lowering or weight management:

• SGLT2 inhibitors provide cardiovascular and renal benefits without affecting gastric emptying 1
• Metformin remains the cornerstone of diabetes management without esophageal risk 1
• DPP-4 inhibitors (sitagliptin, linagliptin) offer modest glucose lowering without the gastric emptying effects of GLP-1 RAs 1
• Metabolic surgery should be considered for patients with BMI >35 kg/m² who cannot safely use GLP-1 therapy 1

Monitoring Requirements if GLP-1 Therapy is Initiated

• Symptom surveillance: Patients should report any new or worsening heartburn, regurgitation, dysphagia, or chest pain immediately 4
• Endoscopic monitoring: Consider shortening surveillance intervals for Barrett's esophagus, though specific recommendations are not established 4, 3
• Gastroenterology consultation is strongly advised before initiating GLP-1 therapy in patients with known Barrett's esophagus 4, 3

In summary, the combination of documented aspiration events, population-level evidence of increased Barrett's complications with short-acting agents, and the persistent gastric retention despite prolonged fasting creates a high-risk scenario that generally argues against GLP-1 use in patients with active esophageal ulcers and warrants extreme caution with long-acting agents only in patients with Barrett's esophagus when alternative therapies are inadequate.