Author + information
- Received November 21, 2012
- Revision received January 7, 2013
- Accepted January 15, 2013
- Published online May 7, 2013.
- Jessica L. Mega, MD, MPH⁎,⁎ (, )
- Eugene Braunwald, MD⁎,
- Sabina A. Murphy, MPH⁎,
- Alexei N. Plotnikov, MD†,
- Paul Burton, MD, PhD†,
- Robert Gabor Kiss, MD, PhD‡,
- Alexander Parkhomenko, MD, PhD§,
- Michal Tendera, MD, PhD∥,
- Petr Widimsky, MD, PhD¶ and
- C. Michael Gibson, MS, MD⁎,#
- ↵⁎Reprint requests and correspondence:
Dr. Jessica L. Mega, Brigham and Women's Hospital, 75 Francis Street, Boston, Massachusetts 02115
Objectives The present analysis reports on the pre-specified subgroup of ST-elevation myocardial infarction (STEMI) patients, in whom anticoagulant therapy has been of particular interest.
Background In ATLAS ACS-2–TIMI-51 (Anti-Xa Therapy to Lower Cardiovascular Events in Addition to Standard Therapy in Subjects with Acute Coronary Syndrome–Thrombolysis In Myocardial Infarction-51), rivaroxaban reduced cardiovascular events across the spectrum of acute coronary syndrome (ACS).
Methods Seven thousand eight hundred seventeen patients in ATLAS ACS-2-TIMI 51 presented with a STEMI. After being stabilized (1 to 7 days), they underwent randomization to twice daily rivaroxaban 2.5 mg, rivaroxaban 5 mg, or placebo. Data are presented as 2-year Kaplan-Meier rates, and for intention-to-treat (ITT) and modified ITT (mITT) analyses.
Results Among STEMI patients, rivaroxaban reduced the primary efficacy endpoint of cardiovascular death, myocardial infarction, or stroke, compared with placebo (ITT: 8.4% vs. 10.6%, hazards ratio [HR]: 0.81, 95% confidence interval [CI]: 0.67 to 0.97, p = 0.019; mITT: 8.3% vs. 9.7%, HR: 0.85, 95% CI: 0.70 to 1.03, p = 0.09). This reduction emerged by 30 days (ITT and mITT: 1.7% vs. 2.3%, p = 0.042) and was evident in analyses that included events while patients received background dual antiplatelet therapies (ITT: 7.9% vs. 11.9%, p = 0.010; mITT: 7.7% vs. 10.1%, p = 0.061). In terms of the individual doses, rivaroxaban 2.5 mg reduced cardiovascular death (ITT: 2.5% vs. 4.2%, p = 0.006; mITT: 2.2% vs. 3.9%, p = 0.006), which was not seen with 5 mg of rivaroxaban. Rivaroxaban versus placebo increased non-coronary artery bypass grafting Thrombolysis In Myocardial Infarction major bleeding (2.2% vs. 0.6%, p < 0.001) and intracranial hemorrhage (0.6% vs. 0.1%, p = 0.015) without a significant increase in fatal bleeding (0.2% vs. 0.1%, p = 0.51).
Conclusions In patients with a recent STEMI, rivaroxaban reduced cardiovascular events. This benefit emerged early and persisted during continued treatment with background antiplatelet therapies. Rivaroxaban compared with placebo increased the rate of major bleeding, but there was no significant increase in fatal bleeding. (An Efficacy and Safety Study for Rivaroxaban in Patients With Acute Coronary Syndrome; NCT00809965)
Morbidity and mortality rates following ST-segment elevation myocardial infarction (STEMI) have declined over time due in part to the delivery of timely reperfusion and in-hospital treatment with anticoagulant and antiplatelet therapies (1,2). Nonetheless, the cumulative risk of death and ischemic events persists during the initial treatment period and after the acute STEMI event (3). Thus, ways to prevent future complications in this at-risk patient population continue to be explored.
ATLAS ACS-2–TIMI-51 (Anti-Xa Therapy to Lower Cardiovascular Events in Addition to Standard Therapy in Subjects with Acute Coronary Syndrome–Thrombolysis In Myocardial Infarction-51) is the largest study to date to test a novel Xa inhibitor, rivaroxaban, in patients with an acute coronary syndrome (ACS). In the trial, as a whole, rivaroxaban reduced recurrent cardiovascular events across the spectrum of ACS (4,5). The present analysis focuses on the results of rivaroxaban versus placebo in the pre-specified subgroup of patients following a STEMI, in whom long-term anticoagulant therapy has been of particular interest. In addition, this analysis addresses several other key aspects of the trial: 1) the results of rivaroxaban versus placebo early after initiation of therapy, during a time when the post-STEMI event rates are at their highest; 2) the effect of rivaroxaban among patients who continued their background antiplatelet therapies, thus providing a particularly rigorous test of the additive role of this novel anticoagulant; and 3) further material on data completeness.
ATLAS ACS-2-TIMI-51 was a randomized, double-blind, placebo-controlled, event-driven trial that included 15,526 patients (≥18 years of age without an upper age limit) who presented with symptoms suggestive of ACS and a diagnosis of STEMI, non–ST-elevation MI (NSTEMI), or unstable angina. Patients were enrolled within 1 to 7 days after hospital admission. They needed to be stabilized before enrollment, with the initial management strategies completed. Patients were randomized in a 1:1:1 fashion to twice daily administration of either rivaroxaban 2.5 mg, rivaroxaban 5 mg, or placebo. They received standard care, including low-dose aspirin therapy and a thienopyridine (either clopidogrel or ticlopidine per the national or local guidelines). The key eligibility criteria and definitions of the endpoints have been published (4).
Rates of the endpoints were expressed as Kaplan-Meier event rates through 24 months. Treatment groups were evaluated using hazard ratios (HRs) and 2-sided 95% confidence intervals (CIs) using Cox proportional hazard models. Testing occurred between doses of rivaroxaban and placebo, based on the log-rank test, stratified by the intention to use a thienopyridine. The term rivaroxaban refers to the combined doses unless the individual doses are indicated. As pre-specified, this analysis focused on a subgroup of patients based on the index event (i.e., STEMI). Additionally, analyses were conducted that included events that occurred while patients were continuing on dual antiplatelet therapy. Exploratory analyses were conducted through 30 days. Data for the efficacy endpoints are presented for the intention-to-treat (ITT) analysis to provide complete accounting of events and for the modified intention-to-treat (mITT) analysis. The ITT analysis consists of all randomized patients and all first endpoint events through the global treatment end date. The mITT analysis consists of all randomized patients and the first endpoint events that occurred no later than: 1) the global treatment end date; 2) 30 days following early permanent discontinuation of the study drug; or 3) 30 days following randomization for patients who did not receive study drug. Thus, the mITT events are a subset of the ITT events. Before unblinding, 90 patients within the STEMI subgroup were excluded from the efficacy analyses due to trial misconduct at 3 sites. Data for the safety endpoints corresponds to the safety analysis set (4).
In ATLAS ACS-2–TIMI-51, a total of 7,817 patients presented with a STEMI. The baseline characteristics of the STEMI patients were well matched across the treatment groups (Table 1). Background therapy included aspirin and a thienopyridine in 98.8% and 96.8% of the patients, respectively. Among the STEMI population, the median time from index STEMI to randomization was 4.7 days (interquartile range: 3.3 to 6.0 days).
The rates of withdrawal of consent were 8.5% for rivaroxaban 2.5 mg twice daily, 7.9% for rivaroxaban 5 mg twice daily, and 7.2% for placebo (p = 0.23). Additionally, the observed follow-up years were 3,132 (94.3% of the complete follow-up time) for rivaroxaban 2.5 mg twice daily, 3,114 (94.5%) for rivaroxaban 5 mg twice daily, and 3,155 (94.7%) for placebo (p = 0.82) (Online Fig. 1). The lost to follow-up rates were 0.3% for rivaroxaban 2.5 mg twice daily and 0.4% for rivaroxaban 5 mg twice daily and placebo (p = 0.88).
Among patients with a STEMI, rivaroxaban reduced the primary efficacy endpoint of cardiovascular death, myocardial infarction, or stroke (ischemic, hemorrhagic, or stroke of uncertain cause) versus placebo (ITT: 8.4% vs. 10.6%, HR: 0.81, 95% CI: 0.67 to 0.97, p = 0.019; mITT: 8.3% vs. 9.7%, HR: 0.85, 95% CI: 0.70 to 1.03, p = 0.09) (Fig. 1), and there was no significant heterogeneity based on region (pinteraction = 0.52). This reduction in the primary endpoint began to emerge within the first 30 days (ITT and mITT: 1.7% vs. 2.3%, HR: 0.71, 95% CI: 0.51 to 0.99, p = 0.042) (Fig. 2). During this early time frame, follow-up data were available for more than 99% of the participants in all 3 treatment groups. Because some patients discontinued their antiplatelet agents during the study, sensitivity analyses were conducted for events that occurred while on treatment with aspirin and a thienopyridine, which yielded consistent results (ITT: 7.9% vs. 11.9%, HR; 0.78, 95% CI: 0.64 to 0.94, p = 0.010; mITT: 7.7% vs. 10.1%, HR; 0.82, 95% CI: 0.67 to 1.01, p = 0.061) (Fig. 3).
Assessing the components of the primary efficacy endpoint, rivaroxaban versus placebo yielded ITT and mITT HRs of 0.76 (95% CI: 0.57 to 1.02) and 0.76 (95% CI: 0.56 to 1.03) for cardiovascular death (which included hemorrhage-related deaths) and 0.78 (95% CI: 0.62 to 0.98) and 0.86 (95% CI: 0.67 to 1.09) for MI, respectively. For stroke, the ITT and mITT HRs were 1.42 (95% CI: 0.81 to 2.46) and 1.40 (95% CI: 0.79 to 2.48), respectively.
Examining the 2 active treatment doses, rivaroxaban 2.5 and 5 mg compared with placebo exhibited similar reductions in the primary efficacy endpoint, without significant differences comparing the 2.5- and 5-mg doses (ITT: 8.7% vs. 8.2%, HR: 0.99, 95% CI: 0.80 to 1.24, p = 0.96; mITT: 8.8% vs. 7.9%, HR: 0.99, 95% CI: 0.79 to 1.25, p = 0.95) (Table 2). Rivaroxaban 2.5 mg reduced cardiovascular death (ITT HR: 0.60, 95% CI: 0.42 to 0.87, p = 0.006; mITT HR: 0.58, 95% CI: 0.39 to 0.86, p = 0.006) and all-cause death (ITT HR: 0.63, 95% CI: 0.45 to 0.89, p = 0.008; mITT HR: 0.60, 95% CI: 0.41 to 0.87, p = 0.007) (Fig. 4); this survival benefit was not seen with 5 mg. This observation for the 2.5-mg dose was consistent during treatment with dual antiplatelet therapy (Online Fig. 2) and regardless of MI location and treatment strategy (Online Fig. 3). Table 2 provides data for the other efficacy endpoints.
Rivaroxaban, compared with placebo, increased the rates of non-coronary artery bypass grafting (CABG) Thrombolysis In Myocardial Infarction (TIMI) major bleeding (2.2% vs. 0.6%, HR: 4.52, 95% CI: 2.27 to 9.01, p < 0.001) (Table 3,Figs. 5 and 6)⇓⇓ without evidence of heterogeneity based on age (Online Table 1). There was also an increase in intracranial hemorrhage (0.6% vs. 0.1%, HR: 8.14, 95% CI: 1.08 to 61.37, p = 0.015) without a statistically significant increase in fatal bleeding (0.2% vs. 0.1%, HR: 1.54, 95% CI: 0.42 to 5.70, p = 0.51).
Comparing the 2 active treatment arms, rivaroxaban 2.5 mg versus 5 mg resulted in fewer non-CABG TIMI major or minor bleeding events (2.5% vs. 4.5%, HR: 0.64, 95% CI: 0.45 to 0.93, p = 0.02). Additionally, there were fewer TIMI medical attention bleeds with 2.5 mg versus 5 mg (12.6% vs. 15.7%, HR: 0.76, 95% CI: 0.64 to 0.89, p < 0.001). Fewer fatal bleeding events occurred with 2.5 mg versus 5 mg of rivaroxaban (1 event vs. 8 events, p = 0.018).
Among the patients who presented with a STEMI in ATLAS ACS-2–TIMI-51, rivaroxaban reduced the primary efficacy endpoint of cardiovascular death, MI, or stroke, compared with placebo. In this study, patients were treated for their initial STEMI event and enrolled within 1 week of their presentation. Evaluating the early effects of rivaroxaban versus placebo, the reductions in cardiovascular endpoints emerged shortly after the study initiation. These findings provide additional insight into factor Xa inhibition in patients following a STEMI and help support the overall results of the trial.
The majority of STEMI patients were treated with aspirin and a thienopyridine (predominantly clopidogrel) at baseline. When focusing the analyses on STEMI patients continuing their antiplatelet therapies, the reductions in cardiovascular events with rivaroxaban were maintained. In the present study, patients were not treated with prasugrel or ticagrelor, and rivaroxaban in addition to these agents has not been tested thoroughly. Nonetheless, the cardiovascular benefits associated with the regimen of rivaroxaban 2.5 mg twice daily, aspirin, and clopidogrel versus aspirin and clopidogrel support the results for aspirin and newer adenosine diphosphate receptor blockers versus aspirin and clopidogrel (6,7). In aggregate, these studies underscore the efficacy of greater degrees of antithrombotic therapy beyond aspirin and standard dose clopidogrel following a STEMI. Moving forward, the ideal duration and combination of these treatments will continue to be studied and discussed (8). Additionally, a better understanding of a given patient's underlying pathobiology and predisposition to recurrent events may help in providing the most favorable antithrombotic strategy.
Rivaroxaban compared with placebo also resulted in a dose-dependent increased rate of bleeding among STEMI patients. Both doses of rivaroxaban led to significantly higher rates of major, as well as medical attention, bleeding events. However, there was no significant increase in the risk of fatal bleeding with rivaroxaban compared with placebo. When comparing the 2 active doses in STEMI patients, the rates of bleeding were lower in the 2.5-mg twice daily group compared with the 5-mg twice daily group. Thus, the lower dose regimen resulted in a more favorable safety profile and might explain, in part, the better survival results with the very low dose of rivaroxaban.
Regarding limitations, this analysis presents data from a subgroup of the trial, and therefore, the power to detect treatment differences is reduced, especially among endpoints with low event rates and potentially within the mITT analysis. Specifically, the mITT analysis included fewer events than the ITT analysis, and the 95% CIs were generally wider with the mITT approach, in some cases crossing the line of unity. However, the ITT and mITT analysis sets yielded directionally consistent results throughout. Additionally, the potential of multiple testing can occur within subgroup analyses. Nonetheless, this index event subgroup was pre-specified and provides data relevant to this important group of patients. Information about the use of rivaroxaban in the acute STEMI setting and specific details about the peri-STEMI management (e.g., use of glycoprotein IIb/IIIa inhibitors) were not available. The trial was designed to exclude patients with an increased risk of bleeding and thus would not apply to these patients.
Among patients stabilized after a STEMI, rivaroxaban reduced cardiovascular events compared with placebo. This reduction emerged within the first 30 days of the study. Moreover, the benefit of rivaroxaban was evident despite the use of antiplatelet therapies, as confirmed by analyses that included patients continuing their background medications. Within the STEMI group, the survival benefit with the 2.5-mg twice daily dose persisted and was not seen with the higher dose of rivaroxaban. As a Xa inhibitor, rivaroxaban was associated with higher rates of bleeding than placebo, with the 2.5-mg twice daily dose of rivaroxaban exhibiting a better safety profile than the 5-mg twice daily dose. Thus, the addition of rivaroxaban 2.5-mg twice daily dose may offer an effective strategy to reduce thrombotic events in patients following a STEMI.
For supplemental tables and figures, please see the online version of this article.
This study was supported by research grants from Johnson & Johnson Pharmaceutical Research and Development and Bayer Healthcare. Dr. Mega has received research grant support from Johnson & Johnson, Bayer Healthcare, Bristol-Myers Squibb/sanofi-aventis, Daiichi Sankyo, Eli Lilly; research supplies from Accumetrics and Nanosphere; and honoraria for consulting from American Genomics, Boehringer Ingelheim, Janssen, and Merck. Dr. Braunwald has received research grant support from Abbott, AstraZeneca, Amgen, Bayer Healthcare, Bristol-Myers Squibb, Daiichi Sankyo, Eli Lilly, GlaxoSmithKline, Merck (SPRI), Pfizer, Roche (Diagnostics), sanofi-aventis, Johnson & Johnson; and has been a consultant for Merck (no compensation), Amorcyte, Daiichi Sankyo, The Medicines Co., Ikaria, CardioRentis, sanofi-aventis, and CVRx (no compensation). Ms. Murphy has received research grant support from Bayer Healthcare, Johnson & Johnson; and honoraria for consulting from Amarin Pharmaceuticals, and Eli Lilly and Company. Drs. Kiss, Parkhomenko, Tendera, and Widimsky have received research grant support from Johnson & Johnson and Bayer Healthcare. Drs. Burton and Plotnikov are employed by and own stock in Johnson & Johnson. Dr. Gibson has received research grant support from Johnson & Johnson, Bayer Healthcare, Bristol-Myers Squibb; honoraria for consulting from Portola Pharmaceuticals, sanofi-aventis, The Medicines Co., Daiichi Sankyo Company, Eli Lilly and Company, Biogen IDEC, Bristol-Myers Squibb, Ischemix, Inc., Johnson & Johnson, Bayer Healthcare, GlaxoSmithKline, Merck Schering Plough, Ortho McNeil, Medicure, Inc., Archemix, Inc., Genentech, Inc., and Boeringer Ingelheim; payment for lectures from Daiichi Sankyo Company, Eli Lilly and Company, Medicines Co.; and payment for development of educational presentations from Daiichi Sankyo Company and Eli Lilly and Company.
- Abbreviations and Acronyms
- acute coronary syndrome
- coronary artery bypass grafting
- confidence interval
- hazard ratio
- myocardial infarction
- modified intention-to-treat
- non–ST-elevation MI
- ST-segment elevation myocardial infarction
- Thrombolysis In Myocardial Infarction
- Received November 21, 2012.
- Revision received January 7, 2013.
- Accepted January 15, 2013.
- American College of Cardiology Foundation
- Roger V.L.,
- Go A.S.,
- Lloyd-Jones D.M.,
- et al.
- Gibson C.M.,
- Mega J.L.,
- Burton P.,
- et al.
- Steg P.G.,
- James S.,
- Harrington R.A.,
- et al.
- Lemesle G.,
- Paparoni F.,
- Delhaye C.,
- Bonello L.,
- Lablanche J.M.