Glucagon-like Peptide 1 Receptor Agonists Reduce the Risk of Venous Thromboembolism in Patients with Diabetes Irrespective of Obesity: A Propensity Score-Matched Multicenter Database Analysis

Introduction:

Glucagon-like peptide 1 receptor agonists (GLP1-RA) are antihyperglycemic agents initially approved for the treatment of type 2 diabetes mellitus (T2DM). More recently, they are also used to achieve weight loss in patients with obesity. There are robust data that demonstrate consistent cardiovascular benefit with the use of GLP1-RA in T2DM, but data are limited to evaluation of atherosclerotic disease. Experimental studies have shown that GLP1-RA may attenuate thromboxane-induced platelet activation. Moreover, obesity is an established risk factor for venous thromboembolism (VTE). Given these observations, we hypothesized that GLP1-RA use may reduce the risk of VTE.

Methods:

We conducted a retrospective, propensity score-matched multicenter database analysis using the TriNetX Analytics Network, which includes de-identified data from electronic health records from over 120 healthcare institutions globally, covering more than 250 million patients. We queried patients with T2DM as the use of a GLP1-RA is indicated for these patients. We excluded patients with use of oral anticoagulation, prior VTE, or a history of atrial fibrillation. We compared patients who received GLP1-RA with those who received Dipeptidyl Peptidase-4 (DPP-4) inhibitors as both of these medications target the incretin system and increase insulin secretion, thereby achieving glycemic control. We further performed a subgroup analysis stratified by the presence of obesity, defined as having a body mass index (BMI) ≥ 30 kg/m². The primary outcome was incidence rate per 1000-patient years of all VTE at 1-year after the index date of first initiation of GLP1-RA or DPP-4 inhibitors; secondary outcomes were pulmonary embolism (PE) and deep venous thrombosis (DVT) individually. We propensity-matched patients (1:1) on GLP1-RA with those on DPP-4 inhibitors based on predetermined clinical variables, including age, sex, race, BMI, hemoglobin A1c, use of other anti-diabetic agents including metformin and insulin, and underlying comorbidities based on the components of the Charleston Comorbidity index. We further performed a subgroup analysis stratified by the presence of obesity, defined as having a body mass index (BMI) ≥ 30 kg/m² as a secondary analysis.

Results:

We identified 656,588 eligible patients who had T2DM. Among these patients, 366,369 received GLP1-RA and 290,219 received DPP-4 inhibitors, respectively. Among patients with known BMI data, 62% of the patients classified as obese with BMI ≥ 30kg/m². After propensity-score matching, the final analysis included two cohorts of each 168,428 patients who received GLP1-RA and DPP-4 inhibitors, respectively. After matching, patients in both cohorts were well balanced (standardized mean differences <0.10) for demographics, hemoglobin A1c, BMI, other anti-diabetic drugs and preexisting comorbidities. The mean HbA1c were 8.3 ± 2.0 and 8.3 ± 2.0, and the mean BMI was 33.3 ± 7.4 and 32.4 ± 7.1 for patients on GLP1-RA and DPP-4 inhibitors, respectively. The incidence of VTE was 11.0 events per 1000 patient-years in the GLP1-RA cohort vs. 12.9 in the DPP-4 inhibitor cohort. Hence, patients who received GLP1-RA had an 18% lower risk of VTE than those who received DPP-4 inhibitors (hazard ratio [HR], 0.82 [95% CI: 0.77-0.88]). Similarly, patients who received GLP1-RA had a lower risk of PE (HR, 0.78 [95% CI: 0.71-0.86]) and DVT (HR, 0.85 [95% CI: 0.79-0.92]) than those on DPP-4 inhibitors. In the subgroup analysis, similar differences in VTE rates were observed in patients with obesity (HR, 0.80 [95% CI: 0.73-0.89]) and without obesity (HR, 0.82 [95% CI: 0.71-0.96]).

Conclusions:

In this large propensity-score matched, multicenter database analysis, the use of GLP1-RA was associated with lower VTE rates after one year in patients with T2DM. This observation was consistent regardless of obesity status. These results support the hypothesis that use of GLP1-RA can lead to a reduction in VTE risk. Further studies are needed to elucidate the mechanisms and causality underlying the association between GLP1-RA use and reduction of VTE risk, and whether these findings extend to patients using GLP1-RA for weight control without T2DM.