Sodium-glucose Cotransporter 2 (SGLT2) Inhibitors and Risk of Cardiovascular Events

The purpose of this study is to compare the risk of cardiovascular events associated with the use of sodium-glucose cotransporter 2 (SGLT2) inhibitors in comparison with the use of dipeptidyl peptidase-4 (DPP-4) inhibitors among patients with type 2 diabetes. The investigators will carry out separate population-based cohort studies using health care databases in seven Canadian provinces and the United Kingdom. The study cohort will be defined by the initiation of a SGLT2 inhibitor or a DPP-4 inhibitor after SGLT2 inhibitors entered the market. Patients will be followed up until the occurrence of a cardiovascular event. The results from the separate sites will be combined by meta-analysis to provide an overall assessment of the risk of cardiovascular events in users of SGLT2 inhibitors. The investigators hypothesize that the use of SGLT2 inhibitors will be associated with a decreased risk of cardiovascular events in comparison with the use of DPP-4 inhibitors.

The objective of this study is to compare the risk of major adverse cardiac events (MACE) associated with the use of sodium-glucose cotransporter 2 (SGLT2) inhibitors in comparison with the use of dipeptidyl peptidase-4 (DPP-4) inhibitors among patients with type 2 diabetes. A common-protocol approach will be used to conduct retrospective cohort studies using administrative health care data from seven Canadian provinces (Alberta, British Columbia, Manitoba, Nova Scotia, Ontario, Quebec, and Saskatchewan) and the United Kingdom (UK) Clinical Practice Research Datalink (CPRD). Briefly, the Canadian databases include population-level data on physician billing, diagnoses and procedures from hospital discharge abstracts, and dispensations for prescription drugs. The data in Ontario will be restricted to patients aged 65 years old and older. The CPRD is a clinical database that is representative of the UK population and contains the records for patients seen at over 680 general practitioner practices in the UK; these data will be linked to the Hospital Episode Statistics (HES) database, which contains in-hospital diagnosis and procedure data. The investigators will use a prevalent new-user cohort design (Suissa et al., 2017). In each jurisdiction, the investigators will assemble a source population that includes all patients who received an antidiabetic medication (metformin, sulfonylureas, thiazolidinediones, DPP-4 inhibitors, SGLT2 inhibitors, GLP-1 receptor agonists, alpha-glucosidase inhibitors, meglitinides, insulin, or combinations of these drugs) between January 1, 2006 and June 30, 2018 (or the latest date of data availability at each site). From this source population, a study cohort including all patients who received a prescription for a SGLT2 inhibitor or DPP-4 inhibitor on or after the date of the first dispensing (or prescription in CPRD) of a SGLT2 inhibitor in each jurisdiction and on or before June 30, 2018 (or latest date of data availability at each site) will be created. Study cohort entry date will be defined by the SGLT2 dispensation date or the corresponding dispensation date for a DPP-4 inhibitor in the matched exposure set. Patients will be followed until the occurrence of an event (defined below), death, end of healthcare coverage (plus 30 days), or end of the study period, whichever occurs first. Patients will be eligible to enter the cohort twice, a first time with a DPP-4 prescription and a second time with a SGLT2 prescription. Exposure will be defined as a prescription for a SGLT2 inhibitor or a DPP-4 inhibitor on the date of cohort entry. DPP-4 inhibitors will serve as the reference category as both classes are second- to third-line therapy, and DPP-4 inhibitors have no known association with the outcome. Analyses will be conducted using an as-treated approach. Patients will be followed until drug discontinuation or the initiation of the other study drug. The primary outcome will be MACE, defined as a composite endpoint of myocardial infarction, ischemic stroke, or cardiovascular death. Secondary outcomes will include the individual endpoints of MACE, all-cause mortality, and hospitalization for heart failure. Using a prevalent new-user design with time-based exposure sets, each user of a SGLT2 inhibitor will be matched to a DPP-4 inhibitor user on the number of prior antidiabetic medication classes and on time-conditional propensity score. Cox proportional hazards models will be used to estimate site-specific adjusted hazards ratios (HR) and corresponding 95% confidential intervals (CI) for each outcome of interest among patients exposed to a SGLT2 inhibitor in comparison to those exposed to a DPP-4 inhibitor. As secondary analyses, the MACE and heart failure outcomes will be stratified by age (≥70 and \<70 years), sex, prior insulin use, and SGLT2 molecule. The MACE composite outcome will be stratified by history of cardiovascular disease, and the heart failure outcome by history of heart failure. Sensitivity analyses will be performed to assess the robustness of study results and address some of the study limitations. Meta-analyses of the site-specific results will be performed using random effects models.