Duration of Dual Antiplatelets in Patients With ACS
Duration of Dual Antiplatelets in Patients With ACS
This study was a prospective observational cohort study using data from the SWEDEHEART and the Swedish National Board of Healthcare registries. The study was approved by the local Ethics Board and registered with clinicaltrials.gov (NCT01623700).
The national SWEDEHEART register includes patients with ACS from all Swedish hospitals and has since 2009 been integrated with the RIKS-HIA (Register of Information and Knowledge about Swedish Heart Intensive care Admissions); SCAAR (Swedish Coronary Angiography and Angioplasty Register); the Swedish Heart Surgery Register, and the National Register of Secondary Prevention (SEPHIA). Details about the register were described previously. For vital status, date of death, incidence of myocardial infarction and stroke, and prescribed and dispensed medications, SWEDEHEART was then further merged with the National Cause of Death Register, National Patient Register, and the Prescribed Drug Register. The Swedish Prescribed Drug register holds information on each and every filled prescription at a Swedish pharmacy.
Merging of the registries was performed by the Department of Statistics, Monitoring, and Evaluation, Epidemiologic Centre of the Swedish National Board of Health and Welfare and was approved by the Ethics Committee at Uppsala University.
Patients hospitalized for ST-elevation ACS (ST-ACS) or non-ST-elevation ACS (NST-ACS) between 1 January 2006 and 1 July 2010 were included in this study if they had filled prescriptions for DAPT with clopidogrel (75 mg once daily) and aspirin (75 mg once daily) after the event and if follow-up data were available for at least 1 year (Figure 1). Patients with missing values for body mass index (BMI) and creatinine levels were excluded. According to local Swedish guidelines regulating the treatment duration, clopidogrel was generally prescribed for 3, 6, or 12 months. For the combined primary endpoint, treatment groups were defined according to dispensed clopidogrel tablets (t) and divided into 3 months' treatment (84–100 clopidogrel t) and >3 months' treatment (>100 t) groups. For the secondary endpoint analyses, the >3 month group was further divided into 6 months (168–200 t) or >6 months (>200 t) groups. Only patients naïve to clopidogrel, defined as no clopidogrel 180 days prior to the ACS event, and with clopidogrel dispensed from the pharmacy within 30 days of the ACS event were included. For the >3 vs. 3 months' DAPT outcome comparison, patients with any event of death, re-infarction, ischaemic or non-ischaemic stroke, stent thrombosis, coronary revascularization, or any clinically relevant bleeding occurring within the first 3 months after the index ACS event were excluded. This was because both ischaemic and bleeding events during this period would potentially interfere with the treating physician's initially determined antiplatelet treatment strategy. Correspondingly, for the >6 vs. 6 months comparison, patients with an event during the first 6 months were excluded.
(Enlarge Image)
Figure 1.
Flow diagram of study population. ACS, acute coronary syndrome; NST-ACS, non-ST-elevation acute coronary syndrome; ST-ACS, ST-elevation acute coronary syndrome; ST, stent thrombosis; t, tablets; DAPT, dual antiplatlet treatment with clopidogrel and asprin.
The pre-defined primary endpoint was to analyse the effect of >3 months' vs. 3 months' DAPT duration on the combined clinical outcome: all-cause death, re-infarction, ischaemic or non-ischaemic stroke. Secondary endpoints included analyses of the effect of >6 months' vs. 6 months' DAPT on the combined endpoint as well as individual endpoints such as all-cause death, re-infarction, all coronary revascularizations, stroke, and bleeding. Appendix 1 lists the definitions of the pre-specified endpoint variables.
Univariate effects of DAPT duration groups (>3 months vs. 3 months and >6 months vs. 6 months) on the combined clinical endpoint of all-cause death, re-infarction, ischaemic or non-ischaemic stroke were evaluated using the Cox proportional hazard regression model. For the survival analyses, only the first occurrence of any of these events were counted. Logistic regression models with DAPT duration groups as response variables and age, sex, BMI, smoking status, ST-ACS/NST-ACS, number of days in hospital 3 years before index ACS, previous thrombolysis, medical history (cancer, dementia, chronic obstructive lung disease, bleeding, heart failure, ischaemic heart disease, atrial fibrillation, dialysis, renal failure, stroke, diabetes, intestinal disease, peripheral vascular disease, hypertension, hyperlipidaemia, revascularization, angiography, stent thrombosis) and drug history (aspirin, warfarin, statin, ACE-inhibitor, beta-blocker, proton pump inhibitor, calcium channel blocker) as covariates were used to estimate two different propensity scores, one for the >3 months vs. 3 months comparison and another for the >6 months vs. 6 months comparison. The estimated propensity scores were normalized and used as inverse probability of treatment weights in the adjusted models. The adjusted ps-weighted Cox proportional hazard regression models included the covariates of revascularization during the acute phase, number of implanted stents, drug-eluting stent, year of enrolment, creatinine clearance, hospital, insulin (5 years before MI), and on-going medication (ACE-inhibitor, statin, warfarin, beta-blocker, calcium channel blocker, oral diabetic, proton pump inhibitor).
In the survival analyses of the combined primary endpoint (death/stroke or re-infarction), patients were censored during the follow-up for bleeding, stent thrombosis, and revascularization events. For the survival analyses of the individual endpoints of bleeding, re-infarction, revascularization, and stroke, patients who died during follow-up were censored and for the analyses of all-cause death, patients were only censored for end of follow-up.
We used the Charlson Index to measure comorbidity. The Charlson Index includes 19 diseases that have been selected and weighted on the basis of the strength of their association with mortality. For the estimation of bleeding risk in treatment groups, the Reduction in Atherothrombosis for Continued Health (REACH) bleeding risk score was calculated. The risk score was developed from the REACH registry, into a nine-item point scheme to evaluate the 2-year risk of serious bleeding. The risk factors included in both scores were included in the statistical models controlling for confounders.
Statistical analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC, USA) and R version 2.14.1.
Methods
This study was a prospective observational cohort study using data from the SWEDEHEART and the Swedish National Board of Healthcare registries. The study was approved by the local Ethics Board and registered with clinicaltrials.gov (NCT01623700).
Patient Selection and Drug Use
The national SWEDEHEART register includes patients with ACS from all Swedish hospitals and has since 2009 been integrated with the RIKS-HIA (Register of Information and Knowledge about Swedish Heart Intensive care Admissions); SCAAR (Swedish Coronary Angiography and Angioplasty Register); the Swedish Heart Surgery Register, and the National Register of Secondary Prevention (SEPHIA). Details about the register were described previously. For vital status, date of death, incidence of myocardial infarction and stroke, and prescribed and dispensed medications, SWEDEHEART was then further merged with the National Cause of Death Register, National Patient Register, and the Prescribed Drug Register. The Swedish Prescribed Drug register holds information on each and every filled prescription at a Swedish pharmacy.
Merging of the registries was performed by the Department of Statistics, Monitoring, and Evaluation, Epidemiologic Centre of the Swedish National Board of Health and Welfare and was approved by the Ethics Committee at Uppsala University.
Patients hospitalized for ST-elevation ACS (ST-ACS) or non-ST-elevation ACS (NST-ACS) between 1 January 2006 and 1 July 2010 were included in this study if they had filled prescriptions for DAPT with clopidogrel (75 mg once daily) and aspirin (75 mg once daily) after the event and if follow-up data were available for at least 1 year (Figure 1). Patients with missing values for body mass index (BMI) and creatinine levels were excluded. According to local Swedish guidelines regulating the treatment duration, clopidogrel was generally prescribed for 3, 6, or 12 months. For the combined primary endpoint, treatment groups were defined according to dispensed clopidogrel tablets (t) and divided into 3 months' treatment (84–100 clopidogrel t) and >3 months' treatment (>100 t) groups. For the secondary endpoint analyses, the >3 month group was further divided into 6 months (168–200 t) or >6 months (>200 t) groups. Only patients naïve to clopidogrel, defined as no clopidogrel 180 days prior to the ACS event, and with clopidogrel dispensed from the pharmacy within 30 days of the ACS event were included. For the >3 vs. 3 months' DAPT outcome comparison, patients with any event of death, re-infarction, ischaemic or non-ischaemic stroke, stent thrombosis, coronary revascularization, or any clinically relevant bleeding occurring within the first 3 months after the index ACS event were excluded. This was because both ischaemic and bleeding events during this period would potentially interfere with the treating physician's initially determined antiplatelet treatment strategy. Correspondingly, for the >6 vs. 6 months comparison, patients with an event during the first 6 months were excluded.
(Enlarge Image)
Figure 1.
Flow diagram of study population. ACS, acute coronary syndrome; NST-ACS, non-ST-elevation acute coronary syndrome; ST-ACS, ST-elevation acute coronary syndrome; ST, stent thrombosis; t, tablets; DAPT, dual antiplatlet treatment with clopidogrel and asprin.
Endpoints
The pre-defined primary endpoint was to analyse the effect of >3 months' vs. 3 months' DAPT duration on the combined clinical outcome: all-cause death, re-infarction, ischaemic or non-ischaemic stroke. Secondary endpoints included analyses of the effect of >6 months' vs. 6 months' DAPT on the combined endpoint as well as individual endpoints such as all-cause death, re-infarction, all coronary revascularizations, stroke, and bleeding. Appendix 1 lists the definitions of the pre-specified endpoint variables.
Statistical Analyses
Univariate effects of DAPT duration groups (>3 months vs. 3 months and >6 months vs. 6 months) on the combined clinical endpoint of all-cause death, re-infarction, ischaemic or non-ischaemic stroke were evaluated using the Cox proportional hazard regression model. For the survival analyses, only the first occurrence of any of these events were counted. Logistic regression models with DAPT duration groups as response variables and age, sex, BMI, smoking status, ST-ACS/NST-ACS, number of days in hospital 3 years before index ACS, previous thrombolysis, medical history (cancer, dementia, chronic obstructive lung disease, bleeding, heart failure, ischaemic heart disease, atrial fibrillation, dialysis, renal failure, stroke, diabetes, intestinal disease, peripheral vascular disease, hypertension, hyperlipidaemia, revascularization, angiography, stent thrombosis) and drug history (aspirin, warfarin, statin, ACE-inhibitor, beta-blocker, proton pump inhibitor, calcium channel blocker) as covariates were used to estimate two different propensity scores, one for the >3 months vs. 3 months comparison and another for the >6 months vs. 6 months comparison. The estimated propensity scores were normalized and used as inverse probability of treatment weights in the adjusted models. The adjusted ps-weighted Cox proportional hazard regression models included the covariates of revascularization during the acute phase, number of implanted stents, drug-eluting stent, year of enrolment, creatinine clearance, hospital, insulin (5 years before MI), and on-going medication (ACE-inhibitor, statin, warfarin, beta-blocker, calcium channel blocker, oral diabetic, proton pump inhibitor).
In the survival analyses of the combined primary endpoint (death/stroke or re-infarction), patients were censored during the follow-up for bleeding, stent thrombosis, and revascularization events. For the survival analyses of the individual endpoints of bleeding, re-infarction, revascularization, and stroke, patients who died during follow-up were censored and for the analyses of all-cause death, patients were only censored for end of follow-up.
We used the Charlson Index to measure comorbidity. The Charlson Index includes 19 diseases that have been selected and weighted on the basis of the strength of their association with mortality. For the estimation of bleeding risk in treatment groups, the Reduction in Atherothrombosis for Continued Health (REACH) bleeding risk score was calculated. The risk score was developed from the REACH registry, into a nine-item point scheme to evaluate the 2-year risk of serious bleeding. The risk factors included in both scores were included in the statistical models controlling for confounders.
Statistical analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC, USA) and R version 2.14.1.