Benefits of Icosapent Ethyl Across the Range of Kidney Function in Patients with Established Cardiovascular Disease or Diabetes: REDUCE-IT RENAL

Background: Chronic kidney disease is associated with adverse outcomes among patients with established cardiovascular disease (CVD) or diabetes. Commonly used medications to treat CVD are less effective among patients with reduced kidney function. Methods: The Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial (REDUCE-IT) was a multicenter, double-blind, placebo-controlled trial that randomized statin-treated patients with elevated triglycerides (135-499 mg/dL) who had CVD or diabetes and one additional risk factor to treatment with icosapent ethyl (4 grams daily) or placebo. Patients from REDUCE-IT were categorized by prespecified eGFR categories to analyze the effect of icosapent ethyl on the primary endpoint (composite of cardiovascular [CV] death, nonfatal myocardial infarction nonfatal stroke, coronary revascularization, or unstable angina) and key secondary endpoint (a composite of CV death, nonfatal myocardial infarction, or nonfatal stroke). Results: Among the 8179 REDUCE-IT patients, median baseline eGFR was 75 mL/min/1.73m 2 (range: 17 to 123 mL/min/1.73m 2 ). There were no meaningful changes in median eGFR for icosapent ethyl versus placebo across study visits. Treatment with icosapent ethyl led to consistent reduction in both the primary and secondary composite endpoints across baseline eGFR categories. Patients with eGFR<60 mL/min/1.73m 2 treated with icosapent ethyl had the largest absolute and similar relative risk reduction for the primary composite endpoint (icosapent ethyl versus placebo, 21.8% versus 28.9%, hazard ratio [HR] 0.71, 95% confidence interval [CI] 0.59-0.85, P=0.0002) and key secondary composite endpoints (16.8% versus 22.5%, HR 0.71, 95% CI 0.57-0.88, p=0.001). The numerical reduction in CV death was greatest in the eGFR <60 mL/min/1.73m 2 group (icosapent ethyl: 7.6%; placebo: 10.6%; HR 0.70, 95% CI 0.51-0.95, P=0.02). While patients with eGFR <60 mL/min/1.73m 2 treated with icosapent ethyl had the highest numerical rates of atrial fibrillation/flutter (icosapent ethyl: 4.2%; placebo 3.0%; HR 1.42, 95% CI 0.86-2.32, P=0.17) and serious bleeding (icosapent ethyl: 5.4%; placebo 3.6%; HR 1.40, 95% CI 0.90-2.18, P=0.13), hazard ratios for atrial fibrillation/flutter and serious bleeding were similar across eGFR categories (P-interaction for atrial fibrillation/flutter = 0.92; P-interaction for serious bleeding = 0.76). Conclusions: In REDUCE-IT, icosapent ethyl reduced fatal and nonfatal ischemic events across the broad range of baseline eGFR categories. Clinical Trial Registration: URL: https://clinicaltrials.gov Unique Identifier: NCT01492361

FC 089EFFECTS OF CANAGLIFLOZIN ON MAJOR ADVERSE CARDIOVASCULAR EVENTS BY BASELINE ALBUMINURIA: INTEGRATED ANALYSES FROM THE CANVAS PROGRAM AND CREDENCE TRIAL

Background and Aims People with type 2 diabetes mellitus (T2DM) have a greater risk of cardiovascular (CV) disease and major adverse CV events (MACE) that is more common as renal function declines. The sodium glucose co-transporter 2 (SGLT2) inhibitor canagliflozin reduced the risk of MACE (CV death, nonfatal myocardial infarction [MI], and nonfatal stroke) in patients with T2DM and high CV risk or nephropathy in the CANVAS Program and CREDENCE trials, respectively. Method This post hoc analysis included integrated, pooled data from the CANVAS Program and the CREDENCE trial. The effects of canagliflozin compared with placebo on MACE were assessed in subgroups defined by baseline urinary albumin:creatinine ratio (UACR; &lt;30, 30-300, and &gt;300 mg/g). Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using stratified (by study) Cox regression models, with subgroup by treatment interaction terms added to test for heterogeneity. Interaction P values were calculated by including the terms of treatment group, baseline UACR, and their interaction in the model. Results A total of 14,543 participants from the CANVAS Program (N = 10,142) and CREDENCE (N = 4,401) were included, with mean estimated glomerular filtration rate of 70.3 mL/min/1.73 m2 and median (interquartile range) UACR of 501.0 (8.4-523.6) mg/g. Among participants with baseline UACR measurements, 7038 (48.8%), 2762 (19.1%), and 4634 (32.1%) participants had baseline UACR &lt;30, 30-300, and &gt;300 mg/g, respectively. Rates of MACE and its components increased as UACR increased (Figure). Canagliflozin reduced the risk of MACE compared with placebo in the overall population (HR, 0.83; 95% CI, 0.75, 0.92), with consistent effects observed across UACR subgroups (interaction P value = 0.42). Canagliflozin also reduced the risk of the individual components of CV death (HR, 0.84; 95% CI, 0.72, 0.97), nonfatal MI (HR, 0.83; 95% CI, 0.70, 0.99), and nonfatal stroke (HR, 0.84; 95% CI, 0.69, 1.03), independent of baseline UACR (interaction P values = 0.40, 0.88, and 0.69, respectively). Canagliflozin was generally well tolerated in the CANVAS Program and the CREDENCE trial, with consistent results on safety outcomes across UACR subgroups. Conclusion Event rates of MACE and its components increased with higher UACR. Canagliflozin reduced the risk of MACE and its components in participants with T2DM and high CV risk or CKD in the CANVAS Program and CREDENCE trial, with consistent benefits observed regardless of baseline UACR.

Sex, Age, and Socioeconomic Differences in Nonfatal Stroke Incidence and Subsequent Major Adverse Outcomes

Background and Purpose: Data about variations in stroke incidence and subsequent major adverse outcomes are essential to inform secondary prevention and prioritizing resources to those at the greatest risk of major adverse end points. We aimed to describe the age, sex, and socioeconomic differences in the rates of first nonfatal stroke and subsequent major adverse outcomes. Methods: The cohort study used linked Clinical Practice Research Datalink and Hospital Episode Statistics data from the United Kingdom. The incidence rate (IR) ratio of first nonfatal stroke and subsequent major adverse outcomes (composite major adverse cardiovascular events, recurrent stroke, cardiovascular disease-related, and all-cause mortality) were calculated and presented by year, sex, age group, and socioeconomic status based on an individual’s location of residence, in adults with incident nonfatal stroke diagnosis between 1998 and 2017. Results: A total of 82 774 first nonfatal stroke events were recorded in either primary care or hospital data—an IR of 109.20 per 100 000 person-years (95% CI, 108.46–109.95). Incidence was significantly higher in women compared with men (IR ratio, 1.13 [95% CI, 1.12–1.15]; P <0.001). Rates adjusted for age and sex were higher in the lowest compared with the highest socioeconomic status group (IR ratio, 1.10 [95% CI, 1.08–1.13]; P <0.001). For subsequent major adverse outcomes, the overall incidence for major adverse cardiovascular event was 38.05 per 100 person-years (95% CI, 37.71–38.39) with a slightly higher incidence in women compared with men (38.42 versus 37.62; IR ratio, 1.02 [95% CI, 1.00–1.04]; P =0.0229). Age and socioeconomic status largely accounted for the observed higher incidence of adverse outcomes in women. Conclusions: In the United Kingdom, incidence of initial stroke and subsequent major adverse outcomes are higher in women, older populations, and people living in socially deprived areas.

Cardiovascular outcomes and healthcare costs of liraglutide versus basal insulin for type 2 diabetes patients at high cardiovascular risk

We aimed to compare the (1) clinical outcomes including composite cardiovascular outcomes, cardiovascular death, and all-cause death, and (2) healthcare costs of using liraglutide and basal insulin as an initial treatment for patients with type 2 diabetes mellitus (T2DM) and high cardiovascular diseases (CVD) risk. This is a retrospective cohort study using Taiwan’s Health and Welfare Database. A total of 1057 patients treated with liraglutide were identified and matched with 4600 patients treated with basal insulin. The liraglutide group had a lower risk of a composite CVD outcome (hazard ratio (HR) 0.65; 95% confidence interval (CI) 0.50–0.85; p < 0.01), all-cause mortality (HR 0.40; 95% CI 0.28–0.59; p < 0.0001), and nonfatal stroke (HR 0.54; 95% CI 0.34–0.87; p = 0.01). Compared to the basal insulin group, the liraglutide group had lower median per-patient-per-month (PPPM) inpatient, emergency room (ER), and total medical costs, but higher median PPPM outpatient, total pharmacy, and total costs (all p < 0.0001). In conclusion, compared to basal insulin, liraglutide was found to be associated with reduced risk of a composite CVD outcome, nonfatal stroke, and all-cause mortality among high CVD risk patients with T2DM. In addition, liraglutide users had lower inpatient, ER, and total medical costs, but they had higher outpatient and total pharmacy costs.

Abstract 15526: Glucagon-like Peptide-1 Receptor Agonists Reduce Adverse Cardiovascular Disease (CVD) Events in Patients With Type 2 Diabetes: An Assessment of Heterogeneity Among CVD Outcomes and a Meta-regression Analysis

Introduction: Recent evidence suggests glucagon-like peptide-1 receptor agonists (GLP-1 RA) reduce adverse cardiovascular disease (CVD) events. The objective of this study is to analyze randomized controlled trials (RCTs) testing GLP-1 RA’s effects on CVD among participants with type 2 diabetes (T2DM). Methods: RCTs comparing GLP-1 RA versus placebo among participants with T2DM were identified using PubMed and Cochrane databases. The endpoints of this analysis included major adverse cardiovascular events (MACE; a composite of cardiovascular death, nonfatal myocardial infarction (MI), and nonfatal stroke), and the individual components of MACE. The primary analysis calculated risk ratios (RR) and 95% confidence intervals (CI) for each endpoint. Heterogeneity for each endpoint was calculated using Chi 2 and I 2 tests. For any endpoint with significant heterogeneity, a meta-regression was performed using baseline hemoglobin A1C (A1C) in each RCT as the moderator and a R 2 value was calculated. Results: 7 RCTs (N = 56,004) were identified with 174,163 patient years of follow-up. GLP-1 RA reduced MACE [RR 0.89, 95% CI 0.83-0.95], cardiovascular death [RR 0.88, 95% CI 0.81-0.95], and nonfatal stroke [RR 0.85, 95% CI 0.77-0.94]. There was no significant heterogeneity among these RCTs (Figure 1). GLP-1 RA did not reduce nonfatal MI [RR 0.91, 95% CI 0.82-1.02]. However, there was significant heterogeneity among these RCTS (Chi 2 =12.94, p=0.04, I 2 =54%). When accounting for A1C in the regression model, there was no longer significant heterogeneity for this endpoint (p=0.23, I 2 =27%). A relationship between A1C and GLP-1 RA’s effect on nonfatal MI (R 2 =0.64, Figure 1) was observed when performing the meta-regression. Conclusion: GLP-1 RA reduced MACE, cardiovascular death, and nonfatal stroke in patients with T2DM with minimal heterogeneity among RCTs. GLP-1 RA did not reduce nonfatal MI, however there may be an association between A1C and GLP-1 RA’s effect on nonfatal MI.

REDUCE-IT: outcomes by baseline statin type

Background REDUCE-IT (Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial) randomized 8,179 statin-treated patients with elevated triglycerides and increased cardiovascular (CV) risk to either icosapent ethyl (IPE), a pure, stable prescription form of eicosapentaenoic acid, 4g/day or placebo. IPE significantly reduced time to first occurrence of the primary composite endpoint of major adverse CV events (CV death, nonfatal myocardial infarction [MI], nonfatal stroke, coronary revascularization, or hospitalization for unstable angina) (HR 0.75, CI 0.68–0.83) and key secondary endpoint events (composite of CV death, nonfatal MI, or nonfatal stroke) (HR 0.74, CI 0.65–0.83) versus placebo (all p&lt;0.0001). A modest reduction in placebo-corrected LDL-C was observed (−6.6%; p&lt;0.0001). The mechanisms for the CV benefit of icosapent ethyl are not fully understood. Purpose Explore the impact of statin type and lipophilic/lipophobic category on outcomes, and on LDL-C, to further consider the possible relevance of LDL-C pathways to the observed CV benefit of icosapent ethyl. Methods Primary and key secondary endpoint analyses and LDL-C changes from baseline were explored by individual statin type (atorvastatin, simvastatin, rosuvastatin, or pravastatin) at baseline, and then by categorizing these statins into lipophilic (i.e., hydrophobic: atorvastatin, simvastatin) and lipophobic (i.e., hydrophilic: rosuvastatin, pravastatin) statin groups; 96.1% of patients fell within these individual statin groups. Results CV outcomes were similar across statin types (interaction p=0.61) and lipophilic/lipophobic categories (interaction p=0.51) (Figure). Statin type and category had a similar lack of meaningful impact on the modest placebo-corrected median LDL-C changes from baseline to one year, which ranged from −5.8 to −8.4% (all p≤0.0003). Conclusion No meaningful treatment differences in the primary or key secondary endpoints across statin type or lipophilic/lipophobic category were observed. A similar lack of treatment difference was observed in LDL-C changes from baseline to one year. Therefore, the LDL-C changes and CV risk reduction in REDUCE-IT appear independent of the type of concomitant statin therapy. These data provide clinicians with additional insight regarding concomitant statin therapy considerations when prescribing icosapent ethyl and suggest there are important mechanisms of action for the substantial CV risk reduction observed with icosapent ethyl that are distinct from the LDL receptor pathway. Funding Acknowledgement Type of funding source: Other. Main funding source(s): The study was funded by Amarin Pharma, Inc.

REDUCE-IT: total ischemic events reduced across the full range of baseline LDL cholesterol and other key subgroups

Background REDUCE-IT (Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial), a study of 8,179 randomized statin-treated patients with elevated triglycerides (TG) and increased cardiovascular (CV) risk followed for a median of 4.9 years, demonstrated robust results. Icosapent ethyl (IPE), a pure and stable prescription form of eicosapentaenoic acid, 4g/day reduced both time-to-first and total primary endpoint ischemic events (CV death, nonfatal myocardial infarction [MI], nonfatal stroke, coronary revascularization, or hospitalization for unstable angina) by 25% (HR 0.75; 95% CI 0.68–0.83; p&lt;0.0001) and 30% (rate ratio 0.70; 95% CI 0.62–0.78; p&lt;0.0001), respectively. Similar substantial reductions in first and total key secondary endpoint ischemic events (composite of CV death, nonfatal MI, or nonfatal stroke) were also observed. Demographic and baseline disease characteristics were generally balanced across treatment groups. Time-to-first event analyses showed robust and generally consistent benefit across subgroups. Previous total event analyses by baseline TG demonstrated large, consistent, statistically significant reductions across tertiles, suggesting the CV benefit of IPE is tied primarily to non-TG factors. Purpose Further explore the extent to which IPE reduced total primary and key secondary events across prespecified baseline demographic, disease, treatment, and lipid/lipoprotein/inflammatory biomarker subgroups. Methods Total events across subgroups were assessed with the prespecified negative binomial regression method. Main outcomes were total (first and subsequent) primary and key secondary composite endpoint events. Results Median baseline LDL-C levels in ascending tertiles were 58, 76, and 96 mg/dL; there were large, significant relative reductions in total primary endpoint events with IPE across tertiles (35%, 28%, and 27%, respectively; interaction p=0.62), with parallel substantial absolute risk reductions. Similar, significant relative reductions of 33%, 28%, and 24% in total key secondary endpoint events were observed, along with substantial absolute risk reductions. Total events analyses of prespecified subgroups also demonstrated robust and generally consistent findings for the primary and key secondary composite endpoints. Conclusion REDUCE-IT demonstrated substantial reductions in first and total primary and key secondary endpoint ischemic events, with robust and generally consistent results across baseline TG and LDL-C levels, as well as other prespecified baseline biomarker, demographic, disease, and treatment subgroups. These analyses provide useful insights for clinicians considering the range of patients who may benefit from IPE therapy and suggest that mechanisms beyond the lipid/lipoprotein/inflammatory pathways tested, including mechanisms beyond the LDL receptor pathways, may contribute to the observed substantial reductions in total ischemic burden with IPE therapy. Funding Acknowledgement Type of funding source: Other. Main funding source(s): The study was funded by Amarin Pharma, Inc.

Are the results of clinical trials relevant in the real world? The applicability of REDUCE-IT to the FAST-MI Registry

Background Large clinical trials are often criticized for testing therapeutic strategies in selected populations. The REDUCE-IT data revealed robust atherosclerotic cardiovascular risk reduction with a strategy comprising of high dose omega-3 icosapent ethyl versus placebo in statin treated patients with triglyceride values (TG) between 150 and 499 mg/dl and controlled LDL-C (41–100 mg/ dl). Purpose In order to evaluate the applicability of REDUCE-IT in a French population, we applied the inclusion and exclusion criteria in the French Registry on Acute ST- elevation and non-ST-elevation Myocardial Infarction (FAST-MI) [2010 and 2015]. Methods From the FAST-MI registry, we included patients over 45 years, who had detailed lipid values post-acute hospitalization [11.1 months (median) after hospitalization for myocardial infarction]. We thus compared the applicability of REDUCE-IT [in patients with TG between 150 (or 200) and 500 mg/dl and an LDL-C between 40 and 100 mg/dl while treated with statin therapy] in the FAST-MI registry focusing on the general characteristics, the risk factors, and the cardiovascular prognosis i.e. the rate of total mortality, myocardial infarction and stroke. Results 12.5% of the patients (5.5% if TG between 200 and 500 mg/dl) met the eligibility criteria for REDUCE-IT. The differences between the REDUCE-IT like (n=472) and REDUCE-IT excluded (n=3317) samples were related to age (61 vs 65, NS), male gender (79.5% vs 73.8%, p&lt;0.01), body mass index in kg/m2 (28 vs 26.2, p&lt;0.001), current smoking (45% vs 31.4%, p&lt;0.001), hypertension (56.5% vs 50.2%, p&lt;0.01), and the percentage of diabetic patients (29.5% vs 15.6%, p&lt;0.001). In the REDUCE-IT like sample, the mean values of total cholesterol, TG (median) and the HDL-C were respectively 159, 192, 43 and the LDL-C value was 72 mg/dl. The distribution of the statin regimens in the REDUCE-IT like sample was as follows: 65.3%, 32.4% and 2.3% had high, moderate and low intensity statin therapy. The distribution of TG categories in the REDUCE-IT like sample was as follows: 55.3% (150–199), 36.0% (200–299) and 8.7% (300–500 mg/dl). The subgroup with TGs greater than 200 mg/dl and HDL-C less than 35 mg/dl was 16.1% in the REDUCE-IT like group and 2.1% in the REDUCE-IT excluded group (p&lt;0.001). The cardiovascular event rate (death, nonfatal MI, nonfatal stroke) was respectively 36.7 for the FAST-MI REDUCE-IT like group and 36.9 persons-years (CVD death, nonfatal MI, nonfatal stroke) for the REDUCE-IT trial. Conclusion The cardiovascular residual risk related to elevated TG in the applicable patient population in the FAST-MI registry was similar to the risk in REDUCE-IT. If the results of REDUCE-IT are applied to patients hospitalized for a myocardial infarction in France, on top of statins, 12.5% of these patients could benefit from a strategy of high dose omega-3 icosapent ethyl on top of contemporary medical therapy to improve their future cardiovascular health. Funding Acknowledgement Type of funding source: None

REDUCE-IT: accumulation of data across prespecified interim analyses to final results

Background REDUCE-IT (Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial), an event-driven trial, randomized 8,179 statin-treated patients with elevated triglycerides (TGs) and increased cardiovascular (CV) risk to icosapent ethyl (IPE); pure, stable prescription eicosapentaenoic acid, 4g/day or placebo. 1,612 primary endpoint events (CV death, nonfatal myocardial infarction [MI], nonfatal stroke, coronary revascularization, or hospitalization for unstable angina) projected 90% power to detect 15% relative risk reduction (5% 2-sided alpha). The key secondary composite endpoint was CV death, nonfatal MI, or nonfatal stroke. An independent data and safety monitoring committee (DMC) performed prespecified interim analyses (IAs) at ∼60% (IA1 31 May 2016 data cutoff; 2.9 y median primary endpoint follow-up) and ∼80% (IA2 01 May 2017; 3.7 y) of events; final analysis included 1,606 events (06 Sep 2018; 4.9 y median study follow-up). Purpose Explore REDUCE-IT efficacy and safety across prespecified IAs for insight into progression of robustness and consistency of conclusions. Methods The interim statistical analysis plan guided study continuation decisions by a prespecified decision-making process, including assessment of safety, treatment arm performance, primary composite endpoint formal analyses, and informal robustness analyses, with no futility or efficacy stopping requirements. Prior to DMC IA study continuation decisions, the need for a mature dataset to support the robustness of final efficacy and safety findings was discussed. Sponsor, Steering Committee, and Clinical Endpoint Committee were blinded throughout. Results Primary and key secondary endpoints achieved statistical significance at IA1 and IA2 that persisted at final analyses (p-value below final adjusted 2-sided alpha of 0.0437); hazard ratios also remained consistent and similar robustness was observed across individual endpoint components; clarity of findings across endpoints and subgroups improved with more events. Stopping for overwhelming efficacy was discussed at each IA; prior to IA study continuation recommendations, the DMC considered historical examples of failed CV outcome studies for TG-lowering and mixed omega-3 therapies, reflected on the potential for overestimating final demonstrated benefit using incomplete data, and weighed societal impacts of fuller datasets relative to patient therapy access. Conclusions Consistent, potent efficacy emerged early and persisted across the two prespecified interim and final analyses. The mature dataset demonstrated highly statistically significant reductions in the primary (25%; p=0.00000001) and key secondary (26%; p=0.0000006) endpoints and allowed robust analyses to support overall efficacy and safety conclusions. Allowing the REDUCE-IT dataset to fully mature provided clinicians with robust, consistent, and reliable data upon which to base clinical decisions for IPE in CV risk reduction. Funding Acknowledgement Type of funding source: Other. Main funding source(s): The study was funded by Amarin Pharma, Inc.

SIMILAR CARDIOVASCULAR OUTCOMES BETWEEN INSULIN DETEMIR AND INSULIN GLARGINE IN TYPE 2 DIABETIC PATIENTS WITH EXTREMELY ATHEROSCLEROTIC CARDIOVASCULAR DISEASE RISKS

Objective: The cardiovascular outcomes of insulin detemir in patients with type 2 diabetes mellitus (T2DM) after acute coronary syndrome (ACS) or acute ischemic stroke (AIS) are unclear. The aim of our real-life cohort study was to evaluate the cardiovascular outcomes of insulin detemir (IDet) versus insulin glargine (IGlar) in T2DM patients after ACS or AIS. Methods: A retrospective cohort study was conducted between June 1, 2005, and December 31, 2013, utilizing the Taiwan National Health Insurance Research Database. A total of 3,129 ACS or AIS patients were eligible for the analysis. Clinical outcomes were evaluated by comparing 1,043 subjects receiving IDet with 2,086 propensity score-matched subjects who received IGlar. The primary composite outcome included cardiovascular (CV) death, nonfatal myocardial infarction (MI) and nonfatal stroke. Results: The primary composite outcome occurred in 322 patients (30.9%) in the IDet group and 604 patients (29.0%) in the IGlar group (hazard ratio [HR], 1.12; 95% confidence interval [CI], 0.95 to 1.32) with a mean follow-up of 2.4 years. No significant differences were observed for CV death (HR, 1.09; 95% CI, 0.86 to 1.38), nonfatal MI (HR, 0.88; 95% CI, 0.66 to 1.19), and nonfatal stroke (HR, 1.15; 95% CI, 0.97 to 1.35). There were similar risks of all-cause mortality, hospitalization for heart failure and revascularization between the IDet group and the IGlar group ( P = .647, .115, and .390 respectively). Conclusion: Compared with IGlar, in T2DM patients after ACS or AIS, IDet was not associated with increased risks of CV death, nonfatal MI, or nonfatal stroke. Abbreviations: ACS = acute coronary syndrome; AIS = acute ischemic stroke; ASCVD = atherosclerotic cardiovascular disease; CI = confidence interval; CV = cardiovascular; DKA = diabetic ketoacidosis; HHF = hospitalization for heart failure; HHS = hyperosmolar hyperglycemic state; HR = hazard ratio; IDet = insulin detemir; IGlar = insulin glargine; MI = myocardial infarction; NHIRD = National Health Insurance Research Database; PCI = percutaneous coronary intervention; PSM = propensity score matching; T2DM = type 2 diabetes mellitus