Glucagon-Like Peptide-1 Receptor Agonists Lead to Gastrointestinal Benefits in Patients with Type 2 Diabetes: A Real-World Study

Introduction

Type 2 diabetes mellitus (T2DM) is a medical and financial burden globally [1]. T2DM is a well-known risk factor for cardiovascular disease [2] and chronic kidney diseases [3]. Patients with T2DM frequently require multifaceted therapeutic approaches to effectively manage their condition [4]. Among the pharmacological interventions, glucagon-like peptide-1 receptor agonists (GLP1-RA) are promising in glycemic control and have favorable extra-glycemic effects [5].

GLP1, an incretin hormone from the lower gastrointestinal (GI) tract, stimulates insulin release and reduces glucagon concentration in response to high blood sugar levels after eating [6]. GLP1 receptors are found in various organs. GLP1 can reduce appetite via the central nervous system and it slows gastric emptying in the GI tract [7]. Therefore, GLP1-RA has a broad impact beyond glycemic control, providing significant weight reduction by influencing insulin, glucagon, appetite, and, possibly, energy expenditure in T2DM [8,9]. A recent meta-analysis by Wang et al revealed that people taking GLP1-RAs achieved a mean weight reduction of 4.57 kg (95% CI: 3.78–5.35 kg) in all population, with patients with diabetes showing a consistent weight loss of 2.69 kg (95% CI: 2.17–3.21 kg) [10]. Furthermore, they help treat non-alcoholic fatty liver disease and non-alcoholic steatohepatitis by improving insulin sensitivity and reducing liver fat [11].

GLP1-RA have shown promising effects in treating metabolic dysfunction-associated steatohepatitis (MASH), a severe form of liver disease often associated with fibrosis. In a phase 2 randomized trial of 320 patients with biopsy-confirmed MASH, once-daily Semaglutide demonstrated a higher resolution rate of steatohepatitis compared to placebo (59% vs 17%, P<0.001) [12]. However, another phase 2 trial in patients with MASH-related cirrhosis found Semaglutide 2.4 mg weekly did not significantly improve fibrosis (11% vs 29%, P=0.087) or resolve NASH compared to placebo [13]. The ongoing phase 3 ESSENCE trial evaluates Semaglutide’s efficacy at higher doses (2.4 mg weekly) over 240 weeks, with interim data supporting improvements in steatohepatitis, liver stiffness, fibrosis markers, and liver enzyme levels [14]. These results underscore Semaglutide’s potential in targeting both MASH and the cardiometabolic syndrome often associated with this condition.

Studies on GLP1-RA report common GI adverse effects, primarily nausea and diarrhea. Short-acting formulations frequently cause more nausea, whereas long-acting ones may cause less persistent problems, including constipation [15,16]. A phase 3 trial of subcutaneous once-weekly Semaglutide revealed that nausea occurred more frequently in patients receiving Semaglutide (20% and 24% in the 0.5 mg and 1.0 mg groups, respectively) compared to placebo (8%). Similarly, diarrhea was more prevalent in Semaglutide-treated patients (13% with 0.5 mg and 11% with 1.0 mg) than in the placebo group (2%) [17]. For oral Semaglutide (3 mg to 14 mg), the most common adverse events were nausea (5.1% to 16%) and diarrhea (5.1% to 8.6%), with symptoms generally being mild to moderate and transient [18]. In a multidisciplinary expert consensus statement, patients were advised to eat slowly, reduce meal portions, avoid fatty foods, and drink a generous amount of water to minimize occurrence/severity of GI adverse effects when starting GLP1-RA therapy [19].

Generally, GLP1-RA commonly causes GI adverse effects, including nausea, vomiting, diarrhea, constipation, and delayed gastric emptying. These effects are typically dose-dependent and transient, influencing the autonomic nervous system, particularly the vagus nerve of the parasympathetic system [20]. While gastrointestinal adverse effects are extensively recorded, there have been few comprehensive investigations of the gastroprotective effects on the GI system in individuals with T2DM. This research scrutinizes GI outcomes in patients diagnosed with T2DM who were administered GLP1-RA, utilizing a substantial real-world dataset.

Material and Methods

ETHICS STATEMENT:

This study was approved by the Institutional Review Board of Shin-Kong Wu Ho-Su Memorial Hospital (IRB number: 20240722R) and the requirement for patient informed consent was waived because TriNetX provides access to a de-identified database for patient selection.

DATA SOURCES:

TriNetX is a global health research network that connects healthcare organizations, pharmaceutical companies, and contract research organizations to promote clinical research, to improve the speed and quality of clinical research. The TriNetX platform provides real-time access to comprehensive, anonymized clinical data from millions of patients globally, thereby enabling pharmaceutical companies to design, conduct, and optimize clinical trials more efficiently [21]. Furthermore, researchers can use the analysis tools provided in the TriNetX platform to test hypotheses on selected cohorts. Some high-quality articles using TriNetX have been published [22–26]. This network provides a diverse dataset, including demographic information, diagnostic codes from the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM), and procedure codes from the International Classification of Diseases, Tenth Revision, Procedure Coding System (ICD-10-PCS). It also includes medication codes from the Veterans’ Affairs National Formulary (VA) and the Anatomical Therapeutic Chemical (ATC) classification system, laboratory tests from the Logical Observation Identifiers Names and Codes (LOINC) and TriNetX Curated values (TNX Curated), genetic data from the Human Genome Variation Society (HGVS), and details of healthcare utilization. The network includes data from hospitals, primary care clinics, and specialty practices, covering both insured and uninsured patients.

STUDY DESIGN AND STUDY POPULATION:

The retrospective investigation employing a new-user and active-comparator design utilized data derived from the TriNetX Global Collaborative Network, which involved 146 healthcare organizations (HCOs) and included 156 427 426 records. From this dataset, 2 304 761 adult patients (≥18 years of age) with T2DM (ICD-10: E11), an estimated glomerular filtration rate (eGFR) of ≥60 mL/min/1.73 m2, and visits ≥3 times were selected from January 1, 2019, to December 31, 2022. This study excluded patients with a history of neoplasm (ICD-10: C00-D49), transplant status (ICD-10: Z94), liver fibrosis (ICD-10: K74), or biliary tract and pancreas disorders (ICD-10: K80–K87) before the index date. The study focused on comparing GI (GI) outcomes between 2 groups of patients, including 228 703 GLP1-RA (ATC: A10BJ) and 202 303 DPP4i (ATC: A10BH) users (Figure 1). The index date was defined as the initiation date of GLP1-RA or DPP4i therapy. To ensure cohort consistency, no other conflicting pharmacological agents were prescribed within the 90 days after this index date.

A 1: 1 propensity score matching (PSM) was performed by incorporating all the baseline covariates to reduce confounding effects, generating 2 matched cohorts of 127 216 GLP1-RA and 127 216 DPP4i users. The GI outcomes were then evaluated within a timeframe ranging from 90 days to 4 years after the index day. Those were followed until January, 2025

COVARIATES:

The study evaluated a range of parameters encompassing demographic characteristics, lifestyles, comorbidities, history of surgery, prescription usage, and laboratory values within 2 years before the index date. Demographic parameters were age, sex, body mass index (TNX Curated: 9083; BMI), and racial identity, specifically focusing on White, African American, and Asian groups. Lifestyle habits covered tobacco use (ICD-10: Z72.0), nicotine dependence (ICD-10: F17), and alcohol-related disorders (ICD-10: F10). A comprehensive assessment of comorbidities was conducted, highlighting conditions such as dyslipidemia (ICD-10: E78), calcium metabolism disorder (ICD-10: E83.5), magnesium metabolism disorder (ICD-10: E83.4), iron metabolism disorder (ICD-10: E83.1), phosphorus metabolism disorder (ICD-10: E83.3), hypertensive diseases (ICD-10: I10–I1A), heart failure (ICD-10: I50), cardiac arrhythmias (ICD-10: I49), atrial fibrillation and flutter (ICD-10: I48), cardiomyopathy (ICD-10: I42), nonrheumatic mitral valve disorders (ICD-10: I34), ischemic heart diseases (ICD-10: I20–I25), lymphatic vessel disease (ICD-10: I80–I89), peripheral vessel disease (ICD-10: I73.9), aortic aneurysm and dissection (ICD-10: I71), other soft-tissue disorders (ICD-10: M70–M79), and chronic kidney disease, (ICD-10: N17–N19), as well as chronic lower respiratory diseases (ICD-10: J40–J4A) and upper respiratory tract disease (ICD-10: J30–J39). Conditions related to the biliary and pancreatic systems were also included, such as cholelithiasis (ICD-10: K80), cholecystitis (ICD-10: K81), other gallbladder (ICD-10: K82) and biliary tract diseases (ICD-10: K83), acute pancreatitis (ICD-10: K85), other pancreatic diseases (ICD-10: K86), and peritonitis (ICD-10: K65). Prescription usage covers an extensive list of medications, including antiarrhythmics (VA: CV300), antacids (VA: GA100), laxatives (VA: GA200), agents acting on the renin-angiotensin system (ATC: C09), lipid-modifying agents (ATC: C10A), insulins (ATC: A10A), biguanides (ATC: A10BA), sulfonylureas (ATC: A10BB), sodium-glucose co-transporter-2 inhibitor (ATC: A10BK), thiazolidinediones (ATC: A10BG), corticosteroids for systemic use (ATC: H02), thyroid therapy (ATC: H03), anti-inflammatories (ATC: M01), anti-gout (ATC: M04), beta-blockers (ATC: C07), calcium channel blockers (ATC: C08), and diuretics (ATC: C03).

Laboratory values closest to the index date were considered in the study feature, including alanine aminotransferase (TNX Curated: 9044), aspartate aminotransferase (TNX Curated: 9047), total and direct bilirubin (TNX Curated: 9050 and 9048), triglycerides (TNX Curated: 9004), glycated hemoglobin (TNX Curated: 9037; HbA1c), low-density lipoprotein (TNX Curated: 9002; LDL) and high-density lipoprotein (TNX Curated: 9001; HDL), hemoglobin (TNX Curated: 9014), albumin (TNX Curated: 9045), estimated glomerular filtration rate (TNX Curated: 8001; eGFR), microalbumin in urine (TNX Curated: LG37542–4), sodium (TNX Curated: 9029), calcium (TNX Curated: 9022), potassium (TNX Curated: 9028), chloride (TNX Curated: 9023), and alkaline phosphatase (TNX Curated: 9046). Lastly, surgical history included colon and rectum surgery (CPT: 1007591), intestinal surgery (excluding rectum) (CPT: 1007422), biliary tract surgery (CPT: 1007845), stomach surgery (CPT: 1007344), liver surgery (CPT: 1007795), and appendix surgery (CPT: 1007578). These comprehensive parameters facilitated the comparison between the 2 user groups during PSM application.

STUDY OUTCOMES:

This study evaluated clinical outcomes, including, acute pancreatitis (ICD-10: K85), biliary tract disease (cholelithiasis, cholecystitis, and other diseases of gallbladder and biliary tract; ICD-10: K80–K83), liver failure (ICD-10: K72–K73), peritonitis (ICD-10: K65), peptic ulcer (ICD-10: K25–K28), and GI bleeding (ICD-10: Z87.19, K92.0, K92.1, and K92.2). These conditions are common gastrointestinal complications that have important effects on patient morbidity and strain healthcare systems. Their pronounced clinical ramifications render them especially critical and significant for an exhaustive assessment of the safety profile of pharmaceuticals within this framework. The incidence and the risk for each of these outcomes have been analyzed over a 4-year follow-up period. Observation and data collection continued until a clinical event occurred, the last data entry was made, or the study concluded on January 28, 2025, whichever came first.

SENSITIVITY TEST:

To ensure the robustness of the findings, some sensitivity tests were conducted. A positive outcome control was used, and it included gastroesophageal reflux disease (GERD) (ICD-10: K21), constipation (ICD-10: K59.0), diarrhea (ICD-10: R19.7), and major adverse cardiovascular events (MACE) because the literature has indicated that GLP-1 RA can reduce atherosclerotic cardiovascular risk in patients with type 2 DM [27] and GLP1-RA will lead a high risk of constipation, diarrhea, and GERD [19,26]. The MACE is composed of cerebral infarction (ICD-10: I63), acute myocardial infarction (ICD-10: I21–I22), cardiac arrest (ICD-10: I46), or death. Moreover, a comparison of GI outcomes between GLP1-RA users and DPP4i users was conducted using different confounder-adjusted models of propensity score matching (PSM).

STATISTICAL ANALYSIS:

Variables are presented either categorically (as count and percentage) or numerically (as mean [SD]) based on the type of covariates. The study used a PSM method [28] to establish comparable groups of GLP1-RA and DPP4i users to create comparable groups of GLP1-RA and DPP4i users, incorporating all covariates listed in Table 1. This matching method ensures that the groups are as similar as possible, enabling a more precise comparison of treatment effects. The difference of baseline characteristics of the study participants was assessed by standardized mean difference (SMD), of which a level <0.1 was considered to indicate a small difference [29]. Intention-to-treat analysis was used to ensure all participants were analyzed in their original groups within 90 days of the index day.

Kaplan-Meier curves with log-rank test were utilized to estimate event-free probabilities. Moreover, Cox proportional hazards regression model was used to calculate the hazard ratios (HRs) with 95% confidence interval (CI) for GI outcomes between the GLP1-RA and DDP4i groups. The assumption of proportional hazards was evaluated using the generalized Schoenfeld approach integrated into the TriNetX platform. All statistical analysis was conducted in the TriNetX platform. A 2-sided P value of <0.05 was considered significant. R software version 4.3.0 (Free Software Foundation, Inc.), with the Forestploter and ggplot2 packages, was used to create the figures.

Results

BASELINE CHARACTERISTICS OF PATIENTS:

Table 1 presents a comparison of baseline characteristics and medical conditions between GLP1-RA and DPP4i users before and after statistical matching. Initially, GLP1-RA users (n=228 703) were younger (average age: 56.3±12.7 vs 62.8±12.6 years), less likely to be male (58.6% vs 39.9%), and had higher BMI (36.1 kg/m2 vs 31.5 kg/m2) compared to DPP4i users (n=202 303). Racial distribution showed substantial differences, with varying proportions of White (45.5% vs 52.6%), African American (19.9% vs 14.6%), and Asian (4.2% vs 8.7%) patients. Clinical characteristics showed imbalances in ischemic heart diseases (18.4% vs 22.9%), upper respiratory tract disease (12.4% vs 9%), and soft-tissue disorders (27.2% vs 19.1%). Significant differences were also noted in medication use, including calcium channel blockers (26.3% vs 35.8%), sulfonylureas (24.7% vs 33.1%), SGLT2i (24.4% vs 17.5%), and insulin (48.8% vs 40.1%), all with SMD >0.15. Laboratory parameters also showed notable differences in HbA1c, eGFR, serum calcium, albumin, total and direct bilirubin, hemoglobin, and triglyceride levels (all SMD > 0.1). After propensity score matching (n=127 216 per group), all these differences were successfully balanced with SMD values reduced to <0.1 (except for BMI levels), creating comparable cohorts for outcome analysis.

GASTROINTESTINAL OUTCOMES: GLP1-RA VS DPP4I:

Figure 2 compares the event-free rates for GI outcomes (acute pancreatitis, biliary tract disease, liver failure, peritonitis, peptic ulcer, and GI bleeding) between GLP1-RA and DPP4i users over 4 years and shows that the GLP1-RA group had favorable GI outcomes except for biliary tract disease. Moreover, Table 2 reveals lower incidence rates for GLP1-RA users compared to DPP4i users across all examined GI conditions. Acute pancreatitis occurred in 0.9% vs 1.0% (HR: 0.90, 95% CI: 0.83–0.97), liver failure in 0.8% vs 1.0% (HR: 0.81, 95% CI: 0.75–0.88), peritonitis in 0.4% vs 0.5% (HR: 0.85, 95% CI: 0.76–0.94), peptic ulcer in 1.8% vs 2.0% (HR: 0.89, 95% CI: 0.84–0.94), and GI bleeding in 6.2% vs 6.7% (HR: 0.95, 95% CI: 0.92–0.98) of GLP1-RA and DPP4i users, respectively. However, biliary tract disease risk did not differ significantly between groups.

SUBGROUP ANALYSIS OF SUBSET OF GLP1-RA:

The GI outcomes of GLP-1RA subsets (Semaglutide, liraglutide, and dulaglutide) were compared to DPP4i (Figure 3). Semaglutide showed significantly lower risks for liver failure (HR 0.77, 95% CI 0.69–0.86), peritonitis (HR 0.86, 95% CI 0.75–0.99), and peptic ulcers (HR 0.93, 95% CI 0.86–0.99). Liraglutide reduced risks for peritonitis (HR 0.79, 95% CI 0.66–0.95), while dulaglutide was associated with reduced risks for liver failure (HR 0.83, 95% CI 0.75–0.91), peritonitis (HR 0.84, 95% CI 0.74–0.96), and peptic ulcers (HR 0.79, 95% CI 0.74–0.85). Gastrointestinal bleeding risk varied insignificantly across all subsets.

SENSITIVITY TEST:

The results from the positive outcome control in Table 2 demonstrate that there was a slightly higher risk of developing constipation (HR: 1.02, 95% CI: 1.00–1.04), diarrhea (HR: 1.04, 95% CI: 1.01–1.06), and GERD (HR: 1.02, 95% CI: 1.00–1.03) and a notably lower risk for MACE (HR: 0.89, 95% CI: 0.88–0.91), suggesting an outcome consistent with previous reports.

Further analysis in Table 3 indicates that 3 progressively adjusted PSM models were analyzed, with Model 1 adjusting for basic demographics (age, sex, race, BMI, and lifestyles), Model 2 adding baseline comorbidities and surgery history, and Model 3 further incorporating baseline medication use. Across all models, GLP-1 RA showed consistent protective effects against various GI outcomes compared to DPP4i. The strongest protective effects were observed for liver failure (HR range: 0.57–0.59), peptic ulcer (HR range: 0.60–0.64), and peritonitis (HR range: 0.71–0.80), all with P<0.001. GI bleeding showed modest but significant protection (HR: 0.93–0.95, P<0.001), while effects on acute pancreatitis and biliary tract disease were less pronounced and varied in significance across models.

Discussion

This study used a real-world dataset from the TriNetX Global Collaborative Network to reveal that GLP1-RA users had significantly lower incidences of GI outcomes, including acute pancreatitis, liver failure, peritonitis, peptic ulcer, and GI bleeding, compared to DPP4i users. By highlighting these benefits, the study offers real-world evidence for considering GLP1-RA as a preferred treatment option for patients with T2DM, particularly those at risk for GI complications. To substantiate and confirm the results that we have obtained from our current study, it is imperative that additional randomized controlled trials are conducted in the future to rigorously verify our findings and ensure their reliability and applicability across broader populations.

GLP1-RA is frequently associated with GI adverse effects, including symptoms such as nausea, vomiting, diarrhea, constipation, and GERD [19,26]. These adverse effects are more prevalent during the early treatment stages or when the dosage is being increased. These symptoms can still be quite bothersome for patients, but are generally mild and do not last long. More serious GI issues occur in some cases, such as pancreatitis, which is pancreatic inflammation, and gastroparesis, which is delayed emptying of the stomach [30]. Dietary adjustments can be helpful for preventing such symptoms. This includes eating small, frequent meals throughout the day, staying well-hydrated, and avoiding high-fat foods [31].

GLP1-RA and DPP4i have been shown to be associated with an increased risk of pancreatitis and pancreatic cancer [32]. Chronic overstimulation of GLP1 receptors in exocrine pancreatic cells is thought to potentially cause pancreatitis and increase the risk of pancreatic cancer. Evidence from drug safety reports supports this hypothesis [33] and meta-analyses seem to weaken the idea of a connection between GLP1-RA therapy and acute pancreatitis in patients with T2DM [34,35]. Our study found that GLP1-RA provides better protection against acute pancreatitis compared to DPP4i, but the risk difference (10%) is small. GLP1-RA has a protective effect against pancreatitis through anti-inflammatory mechanisms, which involve the attenuation of pro-inflammatory cytokines (tumor necrosis factor-alpha, interleukin-1 beta, and interleukin-6) by inhibiting nuclear factor-kappa B and stimulating cyclic adenosine monophosphate/protein kinase A pathways to mitigate oxidative stress [36]. Furthermore, these agents promote the health of β cells by facilitating processes such as proliferation, regeneration, and survival [37,38], thereby alleviating the burden on exocrine cells.

Research has found a link between using GLP1-RA and a higher risk of gallbladder and biliary diseases, such as gallstones and inflammation of the gallbladder. A meta-analysis of 76 randomized clinical trials found that the use of GLP1-RAs increases the risk of developing these conditions by 37% [39]. This risk is particularly significant at higher doses, during long-term use, and when the medication is used for weight loss instead of managing diabetes. Liraglutide has received the most attention for its association with an increased risk of gallbladder-related events [40,41]. Our study using real-world evidence found that users of GLP1-RA have a lower risk of biliary tract disease compared to those using DPP4i or other diabetes medications. Therefore, using GLP1-RA might be safe in this area of clinical practice.

Regarding liver health, studies show that people taking GLP1-RA have a much lower risk of decompensated cirrhosis, hepatic encephalopathy, and liver failure compared to those not using these medications. This is especially true for patients with type 2 diabetes and liver cirrhosis [42–44]. GLP1-RA enhances fatty acid oxidation, decreases fat production, and improves glucose metabolism in the liver, making it a promising treatment for non-alcoholic fatty liver disease [11,45]. Overall, these benefits, also observed in our study, make GLP1-RA effective in slowing the progression of chronic liver disease, especially in individuals with T2DM. Moreover, our study showed that GLP1-RA users have a lower risk of peritonitis, potentially due to a reduction in liver cirrhosis. This hypothesis is supported by Simon et al [44], who reported various benefits of reducing ascites and spontaneous bacterial peritonitis in patients with cirrhosis and diabetes when they were prescribed GLP1-RA. However, there is no direct information or studies explicitly linking GLP1-RAs and peritonitis, and this topic needs further study.

Interestingly, the dual effect of GLP1-RAs on GI outcomes in our study – increased GERD risk but decreased peptic ulcer and GI bleeding risk – appears to stem from distinct physiological mechanisms. GLP1-RA delay gastric emptying, leading to gastric distension and increased intra-abdominal pressure [46,47], which can worsen GERD by compromising lower esophageal sphincter function and promoting acid reflux. However, these same agents demonstrate protective effects against peptic ulcers through reduced gastric acid secretion [48], anti-inflammatory properties [36], and improved glycemic control [5], which enhance mucosal healing. In diabetic rats with chronic gastric ulcers, exendin-4, a GLP-1 analog, accelerated ulcer healing by promoting angiogenesis, reducing inflammation, and enhancing tissue granulation [49]. While GERD primarily results from mechanical factors affecting acid reflux, peptic ulcers develop from acid-mediated damage, H. pylori infection, or NSAID use – processes that GLP-1 RAs can help mitigate. Our findings suggest that GLP-1 RAs’ GI effects are mechanism-specific, but further research is needed to fully understand these differential outcomes.

Our study revealed that GLP1-RA users had significantly lower incidences of GI events compared to DPP4i users. This result indicates the potential of GLP1-RA, not only in glycemic control, but also in providing protective benefits against serious complications that significantly affect patient morbidity and healthcare costs in real-word practice. Finally, the reduced incidence of GI events may be associated with the anti-inflammatory [36,50] and hepatoprotective properties [51] of GLP1-RA, which improve glucose and lipid homeostasis, promote weight loss [52], and modulate key pathways involved in liver lipid metabolism [53], causing these conditions. This evidence highlights the importance of considering wider biological effects when choosing antidiabetic therapies to improve patient outcomes beyond just blood sugar levels.

In the analysis of subgroups, Semaglutide distinctly mitigated the likelihood of liver failure, peritonitis, and peptic ulcers. Liraglutide is linked to a lower risk of peritonitis; however, it does not exert a significant influence on other gastrointestinal outcomes. Dulaglutide exhibits the most reliable advantages, markedly reducing the risks of liver failure, peritonitis, peptic ulcers, and gastrointestinal hemorrhage. Moreover, the probabilities of acute pancreatitis and biliary tract disease remain insignificant across all subsets, highlighting their subset-specific effects on GI outcomes.

The strengths of this study include the large, diverse dataset provided by the TriNetX platform, and the rigorous statistical methods used, including PSM and sensitivity test, which improve the reliability and generalizability of the results. Furthermore, evidence from a global real-world dataset can lead to more accurate results in clinical settings. However, the study has some limitations, predominantly stemming from its retrospective observational cohort design, which could not eliminate all the baseline difference. Unmeasured variables still could have influenced the results, while many baseline confounders were selected for matching in this study. Second, the potential inaccuracies in electronic health record data and the exclusion of certain patient groups limit the study’s comprehensiveness. Third, the absence of comprehensive information concerning both the specific dosages of medications administered and the levels of adherence to prescribed treatment regimens recorded within the TriNetX database results in a significant limitation that ultimately obstructs any meaningful efforts to conduct a thorough and nuanced dose-effect analysis, which is crucial for understanding the relationship between the quantity of medication and the resulting therapeutic outcomes. Fourth, the serum lipase concentration was not incorporated into the diagnostic criteria for pancreatitis. Given that pancreatitis is a well-characterized clinical entity diagnosed through established clinical criteria encompassing symptoms, imaging modalities, and laboratory assessments, the absence of consistent lipase measurements is likely to have had a negligible influence on the identification of cases. Finally, the study predominantly involved White people (about 50%), limiting generalization to other races.

Conclusions

This study reveals that GLP1-RA users would experience more favorable outcomes, especially in terms of reducing GI and liver-related adverse events, compared to DPP4i. This study not only enhances our understanding of the effects of GLP1-RA on protecting the GI system but also introduces new opportunities for research into the physiological pathways that benefit GI health and liver function.

Data Availability Statement

TriNetX connects various research centers, providing immediate access to anonymized data from the electronic health records of participating healthcare organizations. Researchers can easily access this data online at .