Evaluation of Cardioprotective Effects of SGLT2 Inhibitors and GLP-1 Receptor Agonists in Korean Patients With Type 2 Diabetes: A Multicenter CDM-Based Retrospective Cohort Study With Prospective AI-ECG Validation
Evaluation of Cardioprotective Effects of SGLT2 Inhibitors and GLP-1 Receptor Agonists in Korean Patients With Type 2 Diabetes: A Multicenter CDM-Based Retrospective Cohort Study With Prospective AI-ECG Validation
This observational study will evaluate the cardiovascular effects of sodium-glucose cotransporter-2 inhibitors (SGLT2 inhibitors) and glucagon-like peptide-1 receptor agonists (GLP-1 receptor agonists) in Korean adults with type 2 diabetes.
The study consists of two phases. Phase 1 will use de-identified electronic health record data from multiple Korean hospitals organized in a common data model. Cardiovascular outcomes, including myocardial infarction, ischemic stroke, and hospitalization for heart failure, will be compared among patients who started an SGLT2 inhibitor or a GLP-1 receptor agonist. Fractures, metabolic changes, and real-world medication use will also be evaluated.
Phase 2 will prospectively enroll approximately 100 adults with type 2 diabetes who are starting an SGLT2 inhibitor as part of routine clinical care at Dongtan Sacred Heart Hospital. Participants will undergo electrocardiography, AI-based ECG analysis, echocardiography, blood and urine tests, and body composition assessment at baseline. Approximately 50 participants with stage B heart failure, defined as structural or functional heart abnormalities without heart failure symptoms, will undergo the same assessments after 6 months. The primary outcome of Phase 2 is the change in echocardiographic global longitudinal strain from baseline to 6 months. The study will also examine whether changes detected by AI-based ECG are associated with changes in echocardiographic measures of cardiac function.
The study does not assign medications or alter routine treatment. All treatment decisions are made by the participants' treating physicians.
Phase 1: Retrospective Cohorts
Inclusion Criteria:
Exclusion Criteria:
Phase 2: Prospective SGLT2 Inhibitor Validation Cohort
Inclusion Criteria:
Exclusion Criteria:
hunjeechoe@gmail.com+821094935703
This is an investigator-initiated, two-phase observational study integrating a multicenter retrospective cohort analysis with a single-center prospective validation cohort. The two phases involve different study populations and data sources and will be analyzed independently. Their findings will be interpreted together to provide complementary evidence regarding long-term clinical cardiovascular outcomes and short-term changes in cardiac function among Korean patients with type 2 diabetes.
In Phase 1, de-identified electronic health record data mapped to the Observational Medical Outcomes Partnership Common Data Model will be obtained from participating Korean hospitals through a distributed data network. A new-user, active-comparator cohort design will be used to compare patients initiating an SGLT2 inhibitor with those initiating a GLP-1 receptor agonist. Medication exposure, cardiovascular events, safety outcomes, metabolic measures, and treatment patterns will be identified using prespecified common data model definitions. Confounding will be addressed using propensity score methods, stratified analyses, and sensitivity analyses. Participating institutions will execute a common analytic protocol locally and provide only aggregated, non-identifiable results for multicenter analysis.
In Phase 2, adults with type 2 diabetes who are scheduled to start an SGLT2 inhibitor during routine clinical care at Dongtan Sacred Heart Hospital will be consecutively screened. Baseline assessments will include standard 12-lead electrocardiography, portable 6-lead electrocardiography, AI-based ECG analysis, advanced echocardiography, laboratory testing, body composition assessment, and research blood sampling. Participants will be classified according to prespecified echocardiographic criteria for stage B heart failure. Approximately 50 participants who meet the protocol-defined criteria for the longitudinal cohort will undergo repeat assessments 6 months after baseline. Baseline data from other consented screening participants may be included in cross-sectional analyses.
Phase 2 will evaluate whether AI-based ECG measurements reflect echocardiographic cardiac dysfunction at baseline and whether changes in AI-based ECG measurements correspond to changes in cardiac function after SGLT2 inhibitor treatment. Analyses will focus on changes within participants, correlations and agreement between AI-ECG and echocardiographic measurements, and the feasibility of AI-ECG for monitoring subclinical cardiac dysfunction.
Medication selection, dose, continuation, and modification are determined solely by the treating physicians according to routine clinical practice. The protocol does not assign treatment, require medication changes, or use AI-ECG results to guide clinical care. No individual-level data will be combined across Phase 1 and Phase 2.
hegletter@hanmail.net+82-31-8086-2492
yoonique555@naver.com+82-31-8086-3491