Clinical effectiveness of second-line antihyperglycemic drugs on major adverse cardiovascular events: An emulation of a target trial
Issued Date
2023-01-30
Resource Type
eISSN
16642392
Scopus ID
2-s2.0-85147897287
Journal Title
Frontiers in Endocrinology
Volume
14
Rights Holder(s)
SCOPUS
Bibliographic Citation
Frontiers in Endocrinology Vol.14 (2023)
Suggested Citation
Siriyotha S., Lukkunaprasit T., Angkananard T., Looareesuwan P., McKay G.J., Attia J., Thakkinstian A. Clinical effectiveness of second-line antihyperglycemic drugs on major adverse cardiovascular events: An emulation of a target trial. Frontiers in Endocrinology Vol.14 (2023). doi:10.3389/fendo.2023.1094221 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/82448
Title
Clinical effectiveness of second-line antihyperglycemic drugs on major adverse cardiovascular events: An emulation of a target trial
Other Contributor(s)
Abstract
Introduction: The cardiovascular benefits of multiple antihyperglycemic drugs as add-on therapies to metformin in the real-practice are unclear. This study aimed to directly compare major adverse cardiovascular events (CVE) associated with these multiple drugs. Methods: An emulation of a target trial was conducted using a retrospective-cohort data of type 2 diabetes mellitus (T2DM) prescribed with second-line drugs on top of metformin, including sodium-glucose cotransporter 2 inhibitors (SGLT2i), dipeptidyl peptidase-4 inhibitors (DPP4i), thiazolidinediones (TZD) and sulfonylureas (SUs). We applied inverse probability weighting and regression adjustment using intention-to-treat (ITT), per-protocol analysis (PPA) and modified ITT. Average treatment effects (ATE) were estimated using SUs as the reference. Results and Discussion: Among 25,498 patients with T2DM, 17,586 (69.0%), 3,261 (12.8%), 4,399 (17.3%), and 252 (1.0%) received SUs, TZD, DPP4i, and SGLT2i. Median follow-up time was 3.56 (1.36-7.00) years. CVE was identified in 963 patients. The ITT and modified ITT approaches showed similar results; the ATE (i.e., the difference of CVE risks) for SGLT2i, TZD, and DPP4i compared to SUs were -0.020(-0.040, -0.0002), -0.010(-0.017, -0.003), and -0.004(-0.010, 0.002), respectively, indicating 2% and 1% significant absolute risk reduction in CVE in SGLT2i and TZD compared to SUs. These corresponding effects were also significant in the PPA with ATEs of -0.045(-0.060, -0.031), -0.015(-0.026, -0.004), and -0.012(-0.020, -0.004). In addition, SGLT2i had 3.3% significant absolute risk reduction in CVE relative to DPP4i. Our study demonstrated benefits of SGLT2i and TZD in reducing CVE in T2DM patients compared to SUs when added to metformin.