Original Article

No Benefit of Ticagrelor Pretreatment Compared With Treatment During Percutaneous Coronary Intervention in Patients With ST-Segment–Elevation Myocardial Infarction Undergoing Primary Percutaneous Coronary Intervention

Sasha Koul, J. Gustav Smith, Matthias Götberg, Elmir Omerovic, Joakim Alfredsson, Dimitrios Venetsanos, Jonas Persson, Jens Jensen, Bo Lagerqvist, Björn Redfors, Stefan James, David Erlinge
https://doi.org/10.1161/CIRCINTERVENTIONS.117.005528
Circulation: Cardiovascular Interventions. 2018;11:e005528
Originally published March 16, 2018

Jump to

Abstract

Background—The effects of ticagrelor pretreatment in patients with ST-segment–elevation myocardial infarction undergoing primary percutaneous coronary intervention (PCI) is debated. This study investigated the effects of ticagrelor pretreatment on clinical outcomes in this patient group.

Methods and Results—Patients with ST-segment–elevation myocardial infarction undergoing primary PCI were included from October 2010 to October 2014 in Sweden. Screening was done using the SWEDEHEART register (Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies). A total of 7433 patients were included for analysis with 5438 patients receiving ticagrelor pretreatment and 1995 patients with ticagrelor given only in the catheterization laboratory. The primary end point of the study was 30-day event rates of a composite of all-cause mortality, myocardial infarction (MI), and stent thrombosis. Secondary end points were mortality, MI, or stent thrombosis alone and major in-hospital bleeding. Crude event rates showed no difference in 30-day composite end point (6.2% versus 6.5%; P=0.69), mortality (4.5% versus 4.7%; P=0.86), MI (1.6% versus 1.7%; P=0.72), or stent thrombosis (0.5% versus 0.4%; P=0.80) with ticagrelor pretreatment. Three different statistical models were used to correct for baseline differences. No difference in the composite end point, mortality, MI, or stent thrombosis was seen between the 2 groups after statistical adjustment. No increase in in-hospital major bleeding rate was observed with ticagrelor pretreatment.

Conclusions—Ticagrelor pretreatment versus ticagrelor given in the catheterization laboratory in patients with ST-segment–elevation myocardial infarction undergoing primary PCI did not improve the composite end point of all-cause mortality or MI or stent thrombosis or its individual components at 30 days.

Introduction

WHAT IS KNOWN

  • Chronic P2Y12 inhibition with ticagrelor decreases mortality and reinfarction in patients with acute coronary syndromes.

  • However, pretreatment with ticagrelor in patients with ST-segment–elevation myocardial infarction is debated.

  • In a previous randomized trial (n=1862), pretreatment with ticagrelor in patients with ST-segment–elevation myocardial infarction did not improve myocardial flow at angiography or ST resolution. There was, in addition, a trend toward increased mortality in patients receiving ticagrelor pretreatment. Potential benefits or harm with ticagrelor pretreatment in patients undergoing primary percutaneous coronary intervention is, thus, still unclear.

WHAT THE STUDY ADDS

  • This registry study (n=7433) evaluated ticagrelor pretreatment compared with treatment given in the catheterization laboratory in patients with ST-segment–elevation myocardial infarction undergoing primary percutaneous coronary intervention in the SWEDEHEART register (The Swedish Web-system for Enhancement and Development of Evidence-based care in Heart disease Evaluated According to Recommended Therapies).

  • The results indicated overall no benefit of ticagrelor pretreatment on a composite end point of mortality, myocardial infarction, and stent thrombosis or the individual components of this end point at 30 days.

  • Delaying ticagrelor administration until after the coronary anatomy is known and a diagnosis of ST-segment–elevation myocardial infarction is confirmed could avoid potent platelet inhibition to patients with other life-threatening disorders, such as aortic dissection, where platelet inhibition could be deleterious.

Modern P2Y12 inhibitors constitute a cornerstone in the treatment of ST-segment–elevation myocardial infarction (STEMI).1,2 Early P2Y12 inhibition for patients with STEMI administered before coronary angiography may decrease stent thrombosis and ischemic events but may increase bleeding complications. Furthermore, patients with other complicating diseases (aortic dissection, aortic rupture, cardiac tamponade, etc) could, thus, be subjected to a treatment that might worsen their clinical condition, and administration of irreversible early P2Y12 inhibitors may complicate or delay emergent surgery. Pretreatment with P2Y12 inhibitors in patients with acute coronary syndromes, especially in patients with STEMI, before coronary angiography and percutaneous coronary intervention (PCI) has, thus, been debated.3–8 The 2013 American College of Cardiology Foundation/American Heart Association guidelines for STEMI recommend usage of P2Y12 inhibitors as early as possible in patients presenting with STEMI undergoing primary PCI, which is also endorsed by the 2014 European Society of Cardiology guidelines for myocardial revascularization and is currently practiced in many countries worldwide.1,2 However, there is limited randomized data on this matter.4

See Editorial by Huber

In the ATLANTIC study (Administration of Ticagrelor in the Cath Lab or in the Ambulance for New ST Elevation Myocardial Infarction to Open the Coronary Artery), ticagrelor was randomized in patients with STEMI and planned primary PCI to either administration before arrival at the catheterization laboratory or administration at the coronary catheterization laboratory.4 Ticagrelor pretreatment did not improve TIMI (Thrombolysis in Myocardial Infarction) flow grade 3 or ST resolution before PCI. However, ticagrelor pretreatment decreased the rate of definite stent thrombosis both within 24 hours and within 30 days. Mortality trended unfavorably with ticagrelor pretreatment (3.3% in pretreatment group compared with 2.0% in standard-care group; P=0.08). The absence of benefit of ticagrelor pretreatment in ATLANTIC study has been discussed as maybe attributed to the short time delay between ticagrelor pretreatment and administration in the catheterization laboratory.4

The purpose of this study was to investigate the effects of pretreatment with ticagrelor in patients with STEMI undergoing primary PCI compared with ticagrelor given at the catheterization laboratory in a real-life setting using the SWEDEHEART register (Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart disease Evaluated According to Recommended Therapies).

Methods

The data, analytic methods, and study materials are available to other researchers for purposes of reproducing the results or replicating the procedure and can be acquired by submitting request to the corresponding author. The corresponding author is responsible for maintaining availability of the data. Pretreatment was defined as treatment before arrival at the catheterization laboratory. This includes pretreatment in the ambulance (majority of patients), referring hospital, coronary care unit, or local emergency department. The study design is a retrospective cohort study.

National Registers

The SWEDEHEART register includes all patients with myocardial infarction (MI) and those undergoing coronary angiography, PCI, cardiac surgery, or percutaneous valve implantations in Sweden. Using each patient’s unique personal identification, the register can be merged with other national registers. The SWEDEHEART subregisters used in this study were the SCAAR (Swedish Coronary Angiography and Angioplasty Register) and the RIKS-HIA (Swedish Register of Information and Knowledge about Swedish Heart Intensive care Admissions). These registers were merged with the national Swedish Hospital Discharge Register and the National Population Register. Patient screening was performed using SCAAR. Data on PCI and pretreatment or treatment in the catheterization laboratory with platelet inhibitors or anticoagulants and definite stent thrombosis were also obtained from the SCAAR. Information on discharge medications, in-hospital bleeding events, and ejection fraction was obtained from RIKS-HIA. Data on medical history and new MI were obtained from the Swedish Hospital Discharge Register. Information on death was obtained from the National Population Register.

Study Sample

A total of 105 242 patients were identified from October 24, 2010, to October 10, 2014, with a first-time inclusion in SCAAR during this time period. Data on death and stent thrombosis were collected up to November 14, 2014. Data on myocardial reinfarctions were collected to December 14, 2014 (maximum follow-up time).

Patients with no STEMI were excluded leaving 17 166 patients for further analysis. Patients without PCI treatment were excluded (n=2099), leaving 15 067 patients for analysis. Patients with no information on ticagrelor pretreatment and ticagrelor administration in the catheterization laboratory were excluded (n=2528+257). Patients not having received either ticagrelor pretreatment or ticagrelor in the catheterization laboratory (n=4757) were excluded. A further 68 patients who received both ticagrelor pretreatment and ticagrelor in the catheterization laboratory were excluded, leaving a total of 7457 patients for analysis, who received either ticagrelor pretreatment or ticagrelor in the catheterization laboratory. Out of these, 5457 patients received ticagrelor pretreatment and no ticagrelor in the catheterization laboratory and 2000 patients received no ticagrelor pretreatment but ticagrelor in the catheterization laboratory. Patients undergoing primary PCI in hospitals with <20 patients on ticagrelor were excluded from the analysis leading to final patient pool of 5438 patients with ticagrelor pretreatment and 1995 patients with ticagrelor given only in the catheterization laboratory (ticagrelor no pretreatment group).

End Points

The main end points of the study were 30-day composite cardiovascular outcome (all-cause mortality/MI/stent thrombosis) and the individual components of the composite end point alone. The secondary end point was major in-hospital bleeding rate defined as fatal, cerebral, or requiring surgery/blood transfusion. The definition of MI was rehospitalization for MI at a cardiac intensive care unit. Stent thrombosis was defined as angiographically verified stent thrombosis.

Statistical Methods

Crude survival rate was estimated using the Kaplan–Meier estimator. Unadjusted clinical events were compared using the log-rank test. Adjusted analyses were performed using 3 different statistical models to balance out baseline confounding differences: (1) direct multivariable regression, (2) propensity score covariate adjustment, and (3) propensity score matching. Covariate selection differed a bit between the models (see below). In general, covariates that based on clinical experience were judged a priori to be potentially confounding were screened for inclusion in the models. Furthermore, variables showing a baseline difference of P<0.20 and deemed as potential confounders were also screened for inclusion.

Discharge medications were not used as covariates because that could impart potential immortal time bias. Glycoprotein IIB/IIIA (GPIIB/IIIA) inhibitor pretreatment was not included as a covariate because the usage was low (0.1%–0.2%). Proportionality of hazards assumption was assessed using Schoenfeld residuals for the primary composite end point.

Direct Multivariable Regression Model

Using the criteria above, the following variables were used in the direct multivariable regression model for the composite end point and mortality: age, sex, current smoking status, aspirin pretreatment, aspirin during PCI, clopidogrel pretreatment, heparin pretreatment, heparin during PCI, GPIIB/IIIA inhibitors during PCI, bivalirudin during PCI, diabetes mellitus, hyperlipidemia, hypertension, previous MI, stroke, heart failure, chronic obstructive pulmonary disease, tertiary center versus nontertiary center, usage of drug-eluting stent, inotropic support during hospital stay, usage of intra-aortic balloon pump, usage of intravenous nitroglycerin, time delay from symptom-to-PCI, year of procedure, access site (femoral versus nonfemoral), and number of stents used. These variables were then used in Cox regression analyses.

For the direct multivariable regression model for MI, because of low event rate (n=112), fewer variables could be included in direct multivariable regression. Using a logistic regression analyses, the following factors were deemed of most importance for the development of subsequent MI at 30 days and were included in the model: clopidogrel pretreatment, heparin pretreatment, age, bivalirudin during PCI, GPIIB/IIIA inhibitors during PCI, previous stroke, previous heart failure, usage of drug-eluting stents, usage of intravenous nitroglycerin, usage of inotropic support during hospital stay, tertiary center versus nontertiary center. These variables were then used in Cox regression analyses. Cardiogenic shock was later added in the models or excluded from the patient population as sensitivity analyses.

Propensity Score Covariate Adjustment

Propensity scoring was done by including factors that were deemed most important for the propensity of receiving treatment at the time of treatment decision in a logistic regression model. Propensity scoring was performed for smoking status, aspirin pretreatment, clopidogrel pretreatment, heparin pretreatment, age, sex, diabetes mellitus, hypertension, hyperlipidemia, previous MI, bypass surgery, stroke, heart failure, cancer, chronic obstructive pulmonary disease, year of procedure, time delay from symptom-to-PCI, specific hospital where PCI was performed, access site (femoral versus nonfemoral), and treated vessel. The propensity score was then used in a Cox regression analysis to achieve a propensity score covariate adjustment. In a separate model, cardiogenic shock was also included in the propensity score.

Propensity Score Matching

Propensity score matching was performed in a 1:1 fashion based on smoking status, aspirin pretreatment, heparin pretreatment, age, sex, diabetes mellitus, hypertension, hyperlipidemia, previous MI, bypass surgery, stroke, heart failure, cancer, chronic obstructive pulmonary disease, symptom-to-PCI time delay, specific hospital where PCI was performed, and treated vessel. Majority of variables were balanced after matching. However, additional adjustment using Cox regression analyses was done for variables that were deemed of importance and were residually unmatched for (femoral access versus nonfemoral access, year of procedure, tertiary center versus nontertiary center, symptom-to-PCI, GPIIB/IIIA inhibitors, and heparin during PCI). Matching was performed using a caliper of 0.02 in the propensity score for the maximum allowed difference in propensity score to closest neighbor.

Multiple Imputation

Results are presented for complete case analysis; however, end points were also analyzed by using multiple imputation for missing data. Multiple imputation was done by using 10 imputations with 100 iterations for every imputation.

Other Statistical Methods

For patient demographics, categorical variables were compared, stratified by ticagrelor pretreatment or not, using Pearson χ2 test. For continuous variables with normal distribution, Student t test was used. For continuous variables without normal distribution, groups were compared using the Mann–Whitney U test.

All tests were 2 sided with a P value for significance 0.05. All analyses were performed in SPSS (SPSS version 22; SPSS, Inc, Chicago, IL) and STATA (STATA version 14; StataCorp, TX).

Ethical Approval

This study was ethically approved by the regional scientific ethics committee of Lund University. Since all patients were anonymized in the study (with their social security number substituted by a unique SWEDEHEART specific ID-number) no informed consent was deemed necessary by the scientific ethics committee.

Results

Baseline Demographics

The degree of pretreatment increased gradually with subsequent years with rates of 47% (2011), 64% (2012), 76% (2013), and 78% (2014).

Patients with ticagrelor pretreatment were significantly older than patients with no ticagrelor pretreatment (Table 1). Patients with ticagrelor pretreatment received significantly more heparin and aspirin pretreatment, compared with patients with no ticagrelor pretreatment, who instead received more clopidogrel pretreatment. Consequently, patients with no ticagrelor pretreatment received more aspirin and heparin during the PCI procedure. Ticagrelor pretreatment patients received less GPIIB/IIIA inhibitors during PCI but more drug-eluting stents (Table 1).

View this table:
Table 1.

Baseline Demographics

After propensity score matching, both groups were overall considerably better matched with 1115 patients included in the ticagrelor pretreatment group and 1115 patients in the no ticagrelor pretreatment group.

After propensity score matching, baseline differences were evened out for age, sex, cardiogenic shock at arrival, diabetes mellitus, previous MI, hyperlipidemia, heparin pretreatment, aspirin pretreatment, and usage of drug-eluting stents. One factor was not significantly different before matching but significantly different after matching (PCI performed at tertiary center). This covariate and the remaining few covariates that were still unbalanced after matching (a minority) and deemed of importance to adjust for were fitted in a Cox regression analysis.

Composite End Point (Death/MI/Stent Thrombosis)

A total of 338 events in the composite end point occurred in the ticagrelor pretreatment group and 130 events in the no pretreatment group. Kaplan–Meier curves for unadjusted event rates of the composite end point are presented in Figure 1A with no observed difference at 30 days (6.2% in pretreatment group versus 6.5% in no pretreatment group; Table 2). Direct multivariable Cox regression showed no difference in adjusted rates of the composite end point (hazard ratio [HR], 1.00; 95% confidence interval [CI], 0.71–1.41). Propensity score covariate adjustment and propensity score matching did not either show any difference in the adjusted rates of the composite end point (Table 2).

View this table:
Table 2.

Summary of Clinical Event Rates at 30 Days

Figure 1.
Figure 1.

Unadjusted Kaplan-Meier event curves for 30-day clinical outcomes. A, Kaplan–Meier curve of 30-day event rate of composite end point (all-cause death, myocardial infarction [MI], and stent thrombosis). B, Kaplan–Meier curve of 30-day event rate of all-cause mortality. C, Kaplan–Meier curve of 30-day event rate of MI. D, Kaplan–Meier curve of 30-day event rate of stent thrombosis. E, Kaplan–Meier curve of 30-day event rate of the composite end point in the propensity score–matched population.

Mortality

A total of 247 deaths occurred in the ticagrelor pretreatment group and 93 in the no pretreatment group at 30 days. Kaplan–Meier curves for crude mortality are presented in Figure 1B with no difference observed at 30 days (4.5% in pretreatment group versus 4.7% in no pretreatment group, Table 2). No model showed a statistically significant effect on mortality, but propensity score covariate adjustment yielded a trend with a HR of 0.78 (95% CI, 0.51–1.19) for 30-day mortality. However, the results for 30-day mortality were neutral with either propensity score matching (HR, 0.88; 95% CI, 0.56–1.38) or direct multivariable regression analysis (HR, 0.92; 95% CI, 0.62–1.38; Table 2).

Myocardial Infarction

A total of 83 MIs occurred in the ticagrelor pretreatment group and 33 in the no pretreatment group at 30 days. Kaplan–Meier curves for MI are presented in Figure 1C with no difference observed in new MI at 30 days (1.6% in pretreatment group versus 1.7% in no pretreatment group, log-rank test P=0.72). Direct multivariable Cox regression analysis showed no association between ticagrelor pretreatment and lower rate of MI at 30 days (HR, 0.91; 95% CI, 0.57–1.47; Table 2). Similar results were obtained using propensity score covariate adjustment (HR, 0.94; 95% CI, 0.46–1.92), whereas propensity score matching suggested a trend, however, with wide CIs toward increased MI with ticagrelor pretreatment (HR, 1.39; 95% CI, 0.67–2.9; Table 2).

Stent Thrombosis

A total of 24 stent thromboses occurred in the ticagrelor pretreatment group and 8 cases in the no pretreatment group at 30 days. Kaplan–Meier curves for stent thrombosis are presented in Figure 1D with no difference observed in stent thrombosis at 30 days (0.5% in pretreatment group versus 0.4% in no pretreatment group, log-rank test P=0.80). Direct multivariable Cox regression analysis and analysis using propensity score matching were not performed because of low event rates of stent thrombosis. Propensity score covariate adjustment showed no statistically significant association between ticagrelor pretreatment and decreased rate of definite stent thrombosis at 30 days (HR, 0.90; 95% CI, 0.22–3.69).

Cardiogenic Shock

A lower incidence of cardiogenic shock on arrival to the catheterization laboratory was noted in the pretreatment group (2.4% in no pretreatment group versus 1.5% in pretreatment group; P=0.03; Table 1). When accounting for total cardiogenic shock, including cardiogenic shock during hospital stay, there was a nonsignificant but borderline association between less cardiogenic shock in the pretreatment group compared with the no pretreatment group (3.4% in no pretreatment group versus 3.0% in pretreatment group; P=0.054). When adjusting for cardiogenic shock in our main models, similar results were obtained for 30-day composite end point, mortality, MI, and stent thrombosis as without cardiogenic shock in the models (Table 3). Sensitivity analyses where patients with cardiogenic shock were excluded yielded similar results.

View this table:
Table 3.

Summary of Adjusted Clinical Event Rates at 30 Days With Cardiogenic Shock Included in Statistical Models

Major In-Hospital Bleeding Events

Major in-hospital bleeding events occurred in 1.0% of patients receiving ticagrelor pretreatment and in 0.9% of patients not receiving ticagrelor pretreatment (P=not significant).

Other Subgroups Analyses

χ2 analyses showed no difference between the ticagrelor pretreatment group and no pretreatment group concerning left ventricular ejection fraction at discharge (Pearson χ2 P=0.42). Further subgroup analyses showed no differences in mortality depending on sex, hypertension, diabetes mellitus, high age, low weight, and so on (Figure 2), with no P values for interaction reaching statistical significance. Adjusted analyses were also performed where all patients with clopidogrel pretreatment were removed and showed similar overall neutral results, however, with wider CIs because of smaller sample size, especially in the no ticagrelor pretreatment group (Table I in the Data Supplement).

Figure 2.
Figure 2.

Forrest plot of 30-day mortality stratified by ticagrelor pretreatment or treatment in catheterization laboratory. No P values of interaction <0.05 were noted.

Propensity Scoring and Multiple Imputation

The C statistic for the propensity score was 0.95 both with and without cardiogenic shock in the propensity scoring model. The Nagelkerke R2 was 0.70 without and 0.71 with cardiogenic shock included in the propensity score.

The C statistic for the propensity score matching was 0.90 without and 0.91 with cardiogenic shock in the propensity score matching model.

All results are presented as complete case analyses, with 91.9% % of cases being complete cases. Multiple imputation was performed to correct for variables with >20 missing cases and did not alter any of the results significantly.

Discussion

Main Findings

Our study constitutes, to our knowledge, the largest published study examining the effects of ticagrelor pretreatment compared with ticagrelor given in the coronary catheterization laboratory in the setting of STEMI and primary PCI. Our findings suggest no significant difference in a composite end point of all-cause mortality/MI/stent thrombosis and mortality, MI, or stent thrombosis alone with ticagrelor pretreatment in patients with STEMI undergoing primary PCI. These findings, in conjunction with the neutral results of the previously published randomized ATLANTIC trial,4 suggest that routine administration of ticagrelor pretreatment should be reevaluated in this patient group.

Previous Studies on Antiplatelet Pretreatment Strategies in STEMI and Primary PCI

Pretreatment with antiplatelet agents for patients with STEMI scheduled for acute PCI has been debated. Several randomized trials using GPIIB/IIIA-inhibitor pretreatment have shown conflicting results concerning hard clinical end points, where some studies have suggested benefit and others not.9–11 Although some data support usage of GPIIB/IIIA inhibitors in patients with STEMI that are early presenters,11 many centers have switched from GPIIB/IIIA-inhibitor–based strategies because of excess bleeding risk.12 Clopidogrel pretreatment has in several registry studies showed promise as a pretreatment option in STEMI patients undergoing primary PCI.13,14 The only prospective and randomized trial evaluating clopidogrel pretreatment in this setting was vastly underpowered because of slow inclusion rate (n=337), however, showed strong trends toward reduction in major adverse clinical events with clopidogrel pretreatment.15

Although both American and European guidelines recommend as early P2Y12 inhibition as possible in patients with STEMI undergoing primary PCI, there is limited data on this matter using modern P2Y12 inhibitors. Ticagrelor is the only modern P2Y12 inhibitor to have been evaluated in this setting in a prospective and randomized fashion in the ATLANTIC study.4

Comparisons With ATLANTIC

Our study shared the same purpose as the ATLANTIC study. Although this study was an observational study, the patients included were all-comer patients, representative of a real-life population and with a considerably larger sample size. Three statistical models were used, each with its advantage and disadvantages, to more adequately shed light on this issue. The choice of usage of multiple statistical models was based on a stricter epidemiological approach; rather than publish a single model that could deviate in any untoward direction, all 3 models were a priori required to show similar findings in order for the data to be considered credible.

Our study indicated no benefit of ticagrelor pretreatment in a 30-day composite cardiovascular end point (mortality/MI/stent thrombosis) or its individual components. We did not see any trend toward increased mortality or a decrease in stent thrombosis at 30 days, in contrast to the ATLANTIC study. However, the overall rate of stent thrombosis was lower in our study than in the ATLANTIC study, more drug-eluting stents were used in our study, and other concomitantly used antiplatelet and anticoagulant agents differed between the studies as well.4

The exact timing of ticagrelor administration is not known in the SCAAR. However, practice in Sweden has been to give ticagrelor pretreatment at first medical contact, after a diagnosis of STEMI has been ascertained by 12-lead ECG. In the recently published prospective and randomized VALIDATE-SWEDEHEART (Bivalirudin versus Heparin in ST-Segment and Non–ST-Segment Elevation Myocardial Infarction in Patients on Modern Antiplatelet Therapy in the Swedish Web-System for Enhancement and Development of Evidence-based Care in Heart Disease Evaluated according to Recommended Therapies Registry Trial), randomization and background demographics were taken from the SCAAR, where additional trial-specific information was added (that is not normally part of the registry). This included time delay from ticagrelor administration to angiography.16 In the VALIDATE trial, representing an all-comer population in Sweden and also using the SWEDEHEART register/SCAAR on which this study is based, >40% of STEMI patients received ticagrelor >1 hour before PCI.16 This time delay in ticagrelor pretreatment is considerably longer than the median of 31 minutes between the 2 ticagrelor loading doses in the ATLANTIC study.4

Cardiogenic shock represents a special case of cause and effect. Antiplatelet agents could potentially have an effect on the hemodynamics, and it would, thus, constitute a part of the causality chain and should not be adjusted for. In fact, in the ATLANTIC study, the trend toward excess death in the pretreatment group consisted of cardiogenic shock, cardiac arrest, mechanical complications, and heart failure.4 However, cardiogenic shock could also represent a potential confounder. Therefore, cardiogenic shock was analyzed in separate models. The results were similar with or without cardiogenic shock in our statistical models and when removing cardiogenic shock patients from the models.

Low rates of bleeding in general were observed, and no increased bleeding rate was noted in the ticagrelor pretreatment group. This could be because of the fact that a majority of patients with STEMI with ticagrelor pretreatment do not have a substantial ticagrelor effect at the initiation of PCI and no greater risk of bleeding during PCI (access-site bleeding, cardiac tamponade, etc).17,18 Subgroup analyses did not show any significant P values for interaction in any subgroup, suggestive of uniform results across multiple subgroups.

Causality

Physiological reasons for the neutral results in our study are purely speculative but deserve mention. The time delay between ticagrelor administration and PCI could be insufficient for ticagrelor effect, in particular when taking into account the morphine interaction with a slower uptake of ticagrelor in STEMI patients compared with healthy volunteers. We have previously published STEMI data (where a majority of patients have received morphine) indicating that only 50% of ticagrelor pretreatment patients have sufficient platelet inhibition at completion of PCI, leaving a majority of patients not sufficiently inhibited during PCI.17 Furthermore, pretreatment perhaps might not be necessary because the cause of mortality and morbidity could be postprocedural thrombotic problems rather than intraprocedural. Patients often receive peri-procedural antithrombotic treatments like heparin, bivalirudin, or GPIIB/IIIA inhibitors, and thus in-catheterization laboratory administration of ticagrelor might be sufficient.

Limitations

We do not have data on patients undergoing coronary angiography because of an indication of STEMI but with no PCI performed. In the SCAAR, information on pretreatment is available only when PCI is performed. Patients receiving ticagrelor pretreatment but with other diagnoses such as aortic dissection are, thus, likely to not be represented in our study material. Furthermore, the exact timing of ticagrelor administration is not known. Although 40% of patients in the VALIDATE-SWEDEHEART trial had more than 1 hour of time delay between ticagrelor and PCI, in 60% of patients, there was <1 hour of time delay between ticagrelor pretreatment to PCI, suggesting that there might not have been adequate time for absorption in several patients.16 Morphine intake and dosage is not known because morphine has been shown to interact with ticagrelor in subgroup analyses of the ATLANTIC study.4

Our study is an observational study, and all results have, thus, to be interpreted with care and serve only as hypothesis generating and cannot provide definite evidence of causality. Although multiple adjusted modeling was performed to decrease confounding, a certain degree of residual confounding cannot be ruled out. Clinical events were registered according to treating physician discretion and not centrally adjudicated.

Our study represents, to our knowledge, the largest study to date examining the effects of ticagrelor pretreatment in patients with STEMI undergoing primary PCI. However, a larger sample size would have increased the statistical power of our observations. Information on TIMI flow grade post PCI was not available and would have added more information. A total of 969 patients also received clopidogrel pretreatment, with the majority of these patients being in the no ticagrelor pretreatment group. Although we have tried to statistically adjust for this, it remains a limitation.

Conclusions

Our results indicate that ticagrelor pretreatment did not improve cardiovascular composite outcomes (death/MI/stent thrombosis) or the individual components of the composite outcome, compared with ticagrelor administered in the catheterization laboratory in patients with STEMI undergoing primary PCI. Our findings cast a doubt on the regular practice of ticagrelor pretreatment in the setting of STEMI and primary PCI. Delaying ticagrelor administration till after the coronary anatomy is known and a diagnosis of STEMI is confirmed, could avoid potent platelet inhibition to patients with other life-threatening disorders, like aortic dissection, where platelet inhibition could be deleterious.6 Furthermore, it could potentially improve outcomes in the subset of patients that require emergency surgery instead of PCI. Our results reinforce the neutral results of the randomized ATLANTIC study. However, a more definite answer on ticagrelor pretreatment in patients with STEMI and primary PCI would require a larger prospective randomized trial.

Acknowledgments

We would like to greatly thank Rebecca Rylance, MS in statistics, for her invaluable help in statistical revision of this article.

Sources of Funding

This study was supported by unrestricted grants from the Swedish Heart-Lung Foundation and TOTAL-AMI. The funders had no role in any part of the study or in any decision about publication.

Disclosures

Drs Koul, Omerovic, Persson, Alfredsson, and Erlinge have received speakers honoraria from Astra Zeneca. Drs Omerovic, James, and Alfredsson have received research grants from Astra Zeneca for other research projects. The other authors report no conflicts

  • Received May 21, 2017.
  • Accepted February 15, 2018.

References

  1. 1.↵
    1. Windecker S,
    2. Kolh P,
    3. Alfonso F,
    4. Collet JP,
    5. Cremer J,
    6. Falk V,
    7. Filippatos G,
    8. Hamm C,
    9. Head SJ,
    10. Juni P,
    11. Kappetein AP,
    12. Kastrati A,
    13. Knuuti J,
    14. Landmesser U,
    15. Laufer G,
    16. Neumann FJ,
    17. Richter DJ,
    18. Schauerte P,
    19. Sousa Uva M,
    20. Stefanini GG,
    21. Taggart DP,
    22. Torracca L,
    23. Valgimigli M,
    24. Wijns W,,
    25. Witkowski A
    ; Authors/Task Force Members. 2014 ESC/EACTS guidelines on myocardial revascularization: the task force on myocardial revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS) developed with the special contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J. 2014;35:2541–619. doi: 10.1093/eurheartj/ehu278.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. O’Gara PT,
    2. Kushner FG,
    3. Ascheim DD,
    4. Casey DE Jr.,
    5. Chung MK,
    6. de Lemos JA,
    7. Ettinger SM,
    8. Fang JC,
    9. Fesmire FM,
    10. Franklin BA,
    11. Granger CB,
    12. Krumholz HM,
    13. Linderbaum JA,
    14. Morrow DA,
    15. Newby LK,
    16. Ornato JP,
    17. Ou N,
    18. Radford MJ,
    19. Tamis-Holland JE,
    20. Tommaso CL,
    21. Tracy CM,
    22. Woo YJ,
    23. Zhao DX,
    24. Anderson JL,
    25. Jacobs AK,
    26. Halperin JL,
    27. Albert NM,
    28. Brindis RG,
    29. Creager MA,
    30. DeMets D,
    31. Guyton RA,
    32. Hochman JS,
    33. Kovacs RJ,
    34. Kushner FG,
    35. Ohman EM,
    36. Stevenson WG,
    37. Yancy CW
    ; American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation. 2013;127:e362–e425. doi: 10.1161/CIR.0b013e3182742cf6.
    OpenUrlFREE Full Text
  3. 3.↵
    1. Bellemain-Appaix A,
    2. Kerneis M,
    3. O’Connor SA,
    4. Silvain J,
    5. Cucherat M,
    6. Beygui F,
    7. Barthélémy O,
    8. Collet JP,
    9. Jacq L,
    10. Bernasconi F,
    11. Montalescot G
    ; ACTION Study Group. Reappraisal of thienopyridine pretreatment in patients with non-ST elevation acute coronary syndrome: a systematic review and meta-analysis. BMJ. 2014;349:g6269.
    OpenUrlPubMed
  4. 4.↵
    1. Montalescot G,
    2. van ‘t Hof AW,
    3. Lapostolle F,
    4. Silvain J,
    5. Lassen JF,
    6. Bolognese L,
    7. Cantor WJ,
    8. Cequier A,
    9. Chettibi M,
    10. Goodman SG,
    11. Hammett CJ,
    12. Huber K,
    13. Janzon M,
    14. Merkely B,
    15. Storey RF,
    16. Zeymer U,
    17. Stibbe O,
    18. Ecollan P,
    19. Heutz WM,
    20. Swahn E,
    21. Collet JP,
    22. Willems FF,
    23. Baradat C,
    24. Licour M,
    25. Tsatsaris A,
    26. Vicaut E,
    27. Hamm CW
    ; ATLANTIC Investigators. Prehospital ticagrelor in ST-segment elevation myocardial infarction. N Engl J Med. 2014;371:1016–1027. doi: 10.1056/NEJMoa1407024.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Montalescot G,
    2. Bolognese L,
    3. Dudek D,
    4. Goldstein P,
    5. Hamm C,
    6. Tanguay JF,
    7. ten Berg JM,
    8. Miller DL,
    9. Costigan TM,
    10. Goedicke J,
    11. Silvain J,
    12. Angioli P,
    13. Legutko J,
    14. Niethammer M,
    15. Motovska Z,
    16. Jakubowski JA,
    17. Cayla G,
    18. Visconti LO,
    19. Vicaut E,
    20. Widimsky P
    ; ACCOAST Investigators. Pretreatment with prasugrel in non-ST-segment elevation acute coronary syndromes. N Engl J Med. 2013;369:999–1010. doi: 10.1056/NEJMoa1308075.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Hansson EC,
    2. Dellborg M,
    3. Lepore V,
    4. Jeppsson A
    . Prevalence, indications and appropriateness of antiplatelet therapy in patients operated for acute aortic dissection: associations with bleeding complications and mortality. Heart. 2013;99:116–121. doi: 10.1136/heartjnl-2012-302717.
    OpenUrlAbstract/FREE Full Text
  7. 7.↵
    1. Redfors B,
    2. Angerås O,
    3. Petursson P,
    4. Råmunddal T,
    5. Omerovic E
    . The ATLANTIC trial does not support the safety of prehospital ticagrelor treatment for patients with ST-elevation myocardial infarction. Int J Cardiol. 2015;190:157–158. doi: 10.1016/j.ijcard.2015.04.135.
    OpenUrl
  8. 8.↵
    1. Capodanno D,
    2. Angiolillo DJ
    . Pretreatment with antiplatelet drugs in invasively managed patients with coronary artery disease in the contemporary era: review of the evidence and practice guidelines. Circ Cardiovasc Interv. 2015;8:e002301. doi: 10.1161/CIRCINTERVENTIONS.114.002301.
    OpenUrlFREE Full Text
  9. 9.↵
    1. Mehilli J,
    2. Kastrati A,
    3. Schulz S,
    4. Früngel S,
    5. Nekolla SG,
    6. Moshage W,
    7. Dotzer F,
    8. Huber K,
    9. Pache J,
    10. Dirschinger J,
    11. Seyfarth M,
    12. Martinoff S,
    13. Schwaiger M,
    14. Schömig A
    ; Bavarian Reperfusion Alternatives Evaluation-3 (BRAVE-3) Study Investigators. Abciximab in patients with acute ST-segment-elevation myocardial infarction undergoing primary percutaneous coronary intervention after clopidogrel loading: a randomized double-blind trial. Circulation. 2009;119:1933–1940. doi: 10.1161/CIRCULATIONAHA.108.818617.
    OpenUrlAbstract/FREE Full Text
  10. 10.↵
    1. Ellis SG,
    2. Tendera M,
    3. de Belder MA,
    4. van Boven AJ,
    5. Widimsky P,
    6. Janssens L,
    7. Andersen HR,
    8. Betriu A,
    9. Savonitto S,
    10. Adamus J,
    11. Peruga JZ,
    12. Kosmider M,
    13. Katz O,
    14. Neunteufl T,
    15. Jorgova J,
    16. Dorobantu M,
    17. Grinfeld L,
    18. Armstrong P,
    19. Brodie BR,
    20. Herrmann HC,
    21. Montalescot G,
    22. Neumann FJ,
    23. Effron MB,
    24. Barnathan ES,
    25. Topol EJ
    ; FINESSE Investigators. Facilitated PCI in patients with ST-elevation myocardial infarction. N Engl J Med. 2008;358:2205–2217. doi: 10.1056/NEJMoa0706816.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Van’t Hof AW,
    2. Ten Berg J,
    3. Heestermans T,
    4. Dill T,
    5. Funck RC,
    6. van Werkum W,
    7. Dambrink JH,
    8. Suryapranata H,
    9. van Houwelingen G,
    10. Ottervanger JP,
    11. Stella P,
    12. Giannitsis E,
    13. Hamm C
    ; Ongoing Tirofiban In Myocardial Infarction Evaluation (On-TIME) 2 Study Group. Prehospital initiation of tirofiban in patients with ST-elevation myocardial infarction undergoing primary angioplasty (On-TIME 2): a multicentre, double-blind, randomised controlled trial. Lancet. 2008;372:537–546. doi: 10.1016/S0140-6736(08)61235-0.
    OpenUrlCrossRefPubMed
  12. 12.↵
    1. Stone GW,
    2. Witzenbichler B,
    3. Guagliumi G,
    4. Peruga JZ,
    5. Brodie BR,
    6. Dudek D,
    7. Kornowski R,
    8. Hartmann F,
    9. Gersh BJ,
    10. Pocock SJ,
    11. Dangas G,
    12. Wong SC,
    13. Kirtane AJ,
    14. Parise H,
    15. Mehran R
    ; HORIZONS-AMI Trial Investigators. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med. 2008;358:2218–2230. doi: 10.1056/NEJMoa0708191.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Koul S,
    2. Smith JG,
    3. Scherstén F,
    4. James S,
    5. Lagerqvist B,
    6. Erlinge D
    . Effect of upstream clopidogrel treatment in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention. Eur Heart J. 2011;32:2989–2997. doi: 10.1093/eurheartj/ehr202.
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Dörler J,
    2. Edlinger M,
    3. Alber HF,
    4. Altenberger J,
    5. Benzer W,
    6. Grimm G,
    7. Huber K,
    8. Pachinger O,
    9. Schuchlenz H,
    10. Siostrzonek P,
    11. Zenker G,
    12. Weidinger F
    ; Austrian Acute PCI Investigators. Clopidogrel pre-treatment is associated with reduced in-hospital mortality in primary percutaneous coronary intervention for acute ST-elevation myocardial infarction. Eur Heart J. 2011;32:2954–2961. doi: 10.1093/eurheartj/ehr360.
    OpenUrlCrossRefPubMed
  15. 15.↵
    1. Zeymer U,
    2. Arntz HR,
    3. Mark B,
    4. Fichtlscherer S,
    5. Werner G,
    6. Schöller R,
    7. Zahn R,
    8. Diller F,
    9. Darius H,
    10. Dill T,
    11. Huber K
    . Efficacy and safety of a high loading dose of clopidogrel administered prehospitally to improve primary percutaneous coronary intervention in acute myocardial infarction: the randomized CIPAMI trial. Clin Res Cardiol. 2012;101:305–312. doi: 10.1007/s00392-011-0393-1.
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. Erlinge D,
    2. Omerovic E,
    3. Fröbert O,
    4. Linder R,
    5. Danielewicz M,
    6. Hamid M,
    7. Swahn E,
    8. Henareh L,
    9. Wagner H,
    10. Hårdhammar P,
    11. Sjögren I,
    12. Stewart J,
    13. Grimfjärd P,
    14. Jensen J,
    15. Aasa M,
    16. Robertsson L,
    17. Lindroos P,
    18. Haupt J,
    19. Wikström H,
    20. Ulvenstam A,
    21. Bhiladvala P,
    22. Lindvall B,
    23. Lundin A,
    24. Tödt T,
    25. Ioanes D,
    26. Råmunddal T,
    27. Kellerth T,
    28. Zagozdzon L,
    29. Götberg M,
    30. Andersson J,
    31. Angerås O,
    32. Östlund O,
    33. Lagerqvist B,
    34. Held C,
    35. Wallentin L,
    36. Scherstén F,
    37. Eriksson P,
    38. Koul S,
    39. James S
    . Bivalirudin versus heparin monotherapy in myocardial infarction. N Engl J Med. 2017;377:1132–1142. doi: 10.1056/NEJMoa1706443.
    OpenUrl
  17. 17.↵
    1. Koul S,
    2. Andell P,
    3. Martinsson A,
    4. Smith JG,
    5. Scherstén F,
    6. Harnek J,
    7. Götberg M,
    8. Norström E,
    9. Björnsson S,
    10. Erlinge D
    . A pharmacodynamic comparison of 5 anti-platelet protocols in patients with ST-elevation myocardial infarction undergoing primary PCI. BMC Cardiovasc Disord. 2014;14:189. doi: 10.1186/1471-2261-14-189.
    OpenUrlPubMed
  18. 18.↵
    1. Franchi F,
    2. Rollini F,
    3. Cho JR,
    4. Bhatti M,
    5. DeGroat C,
    6. Ferrante E,
    7. Dunn EC,
    8. Nanavati A,
    9. Carraway E,
    10. Suryadevara S,
    11. Zenni MM,
    12. Guzman LA,
    13. Bass TA,
    14. Angiolillo DJ
    . Impact of escalating loading dose regimens of ticagrelor in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention: results of a prospective randomized pharmacokinetic and pharmacodynamic investigation. JACC Cardiovasc Interv. 2015;8:1457–1467. doi: 10.1016/j.jcin.2015.02.030.
    OpenUrlAbstract/FREE Full Text
View Abstract

This Issue

Jump to

Article Tools

  • Print
  • Citation Tools
  • Article Alerts
  • Save to my folders

Share this Article

  • Email
  • No Benefit of Ticagrelor Pretreatment Compared With Treatment During Percutaneous Coronary Intervention in Patients With ST-Segment–Elevation Myocardial Infarction Undergoing Primary Percutaneous Coronary Intervention
    Sasha Koul, J. Gustav Smith, Matthias Götberg, Elmir Omerovic, Joakim Alfredsson, Dimitrios Venetsanos, Jonas Persson, Jens Jensen, Bo Lagerqvist, Björn Redfors, Stefan James and David Erlinge
    Circulation: Cardiovascular Interventions. 2018;11:e005528, originally published March 16, 2018
    https://doi.org/10.1161/CIRCINTERVENTIONS.117.005528
    del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Related Articles

Cited By...

Subjects