Society: AGA

Introduction:
It is controversial whether glucagon-like peptide 1 (GLP-1) receptor agonists (RA) are associated with adverse pancreatic, gallbladder or biliary outcomes. Using a global aggregated electronic health record database, we assessed if starting patients with T2DM on GLP-1 RAs was associated with an increased risk of pancreaticobiliary disease.
Materials and Methods:
A retrospective query was constructed on the TriNetX platform querying 91 healthcare organizations from 12 countries. We identified all non-deceased adult patients with active records who had T2DM at the time of GLP-1 RA initiation between 2002 to 2022. The control group was patients with T2DM at the time of sodium-glucose co-transport 2 inhibitors (SGLT2i) initiation. Patients in both cohorts were excluded if they had a history of pancreatic or hepatobiliary disease or had undergone any procedure related to this system up until 1 day after the index event. Patients were also excluded if they ever had a neoplasm of any digestive organ or a history of malignant neuroendocrine tumors. Furthermore, both groups were exclusive and propensity-matched based on key factors that would predict GLP-1 RA vs. SGLT2i initiation, as well as risk factors for pancreatic and hepatobiliary disease (Table 1). We extracted the number of patients who were diagnosed with acute pancreatitis or cholelithiasis, as well as those who had a first-time diagnosis of cholecystitis, chronic pancreatitis, or underwent cholecystectomy. We combined bile duct stones, bile duct obstruction, biliary colic, and biliary fistula under a single outcome of biliary disease. Finally, we extracted the number of patients who underwent first-time endoscopic retrograde cholangiopancreatography (ERCP) as well as the number of hospitalizations or emergency visits. All outcomes were identified between 1 day after the index event and up to 3 years. Odds ratios were calculated within the TriNetX Analytics Platform with significance set at a 2-sided P value <0.05.
Results:
After propensity matching, there were 206, 717 patients on GLP-1 RA without SGLT2i and vice versa. Within 3 years of initiating GLP-1 RA, (1) patients were at increased risk of developing cholecystitis (OR 1.42; 95% CI 1.25-1.61) and cholelithiasis (1.32; 1.24-1.39), and undergoing cholecystectomy (1.41; 1.24-1.61); (2) patients also had an increased risk of emergency department visits (1.27; 1.24-1.30) and hospitalization (1.14; 1.13-1.16); these associations persisted with initial A1c <7%; however, (3) patients were not more likely to experience a biliary disease outcome or acute/chronic pancreatitis, or undergo an ERCP.
Conclusion:
GLP-1 RAs are associated with developing cholecystitis and cholelithiasis, as well as undergoing a cholecystectomy; however, they are not associated with acute or chronic pancreatitis or other biliary disease events.

<b>Table 1: Characteristics of GLP-1 RA and SGLT2i Cohort Before and After Propensity Matching</b>

Table 1: Characteristics of GLP-1 RA and SGLT2i Cohort Before and After Propensity Matching

<b>Figure 1: Comparison of Risk of Pancreatic, Gall bladder, and Hepatobiliary Complications in GLP-1 RA and SGLT2i</b>

Figure 1: Comparison of Risk of Pancreatic, Gall bladder, and Hepatobiliary Complications in GLP-1 RA and SGLT2i

Katherine Liang

Appearances