Atrial Fibrillation and Acute Myocardial Infarction: Antithrombotic Therapy and Outcomes

Transcription

1 CLINICAL RESEARCH STUDY Atrial and Acute Myocardial Infarction: Antithrombotic Therapy and Outcomes Renato D. Lopes, MD, PhD, a Li Li, MS, a Christopher B. Granger, MD, a Tracy Y. Wang, MD, MHS, MSc, a JoAnne M. Foody, MD, b Marjorie Funk, PhD, RN, c Eric D. Peterson, MD, MPH, a Karen P. Alexander, MD a a Duke Clinical Research Institute, Duke University Medical Center, Durham, NC; b Brigham and Women s Hospital, Boston, Mass; c Yale University School of Nursing, New Haven, Conn. ABSTRACT BACKGROUND: Atrial fibrillation guidelines recommend long-term use of warfarin according to a patient s predicted risk of stroke. After acute myocardial infarction, however, combining warfarin and antiplatelet medications poses challenges. METHODS: By using data from more than 69,255 patients with acute myocardial infarction who were enrolled in the National Cardiovascular Data Registry s Acute Coronary Treatment and Intervention Outcomes Network Registry Get With the Guidelines at 309 hospitals from July 1, 2008, to September 30, 2009, we describe the characteristics and outcomes of the population with myocardial infarction with atrial fibrillation diagnosed within 2 weeks before index myocardial infarction admission (7.1%, N 4947). Use of discharge antithrombotic therapy is described overall and across levels of predicted stroke and bleeding risks. RESULTS: Compared with patients without atrial fibrillation, those with atrial fibrillation before their index myocardial infarction were older and had more comorbidities and worse in-hospital outcomes. Only 32.5% of patients with atrial fibrillation were taking warfarin before their myocardial infarction admission. In these patients, use of warfarin at discharge increased with higher Congestive heart failure, Hypertension, Age, Diabetes, Stroke [Doubled] (CHADS 2 ) risk strata (28.5%, 34.6%, and 43.5% for CHADS 2 scores 0, 1, and 2; P.001) and increased in patients at low, intermediate, and high risk of bleeding (25.4%, 42.3%, and 42.1%; P.004). Triple therapy at discharge (aspirin plus clopidogrel plus warfarin) was used in a minority of this population (14.6%). CONCLUSIONS: Use of warfarin at discharge in patients with atrial fibrillation is greater among those with higher stroke and bleeding risks, but despite higher-risk profiles, less than half received warfarin at discharge. These findings highlight that clarification is needed to guide choice of antithrombotic therapy for patients with both atrial fibrillation and acute myocardial infarction Elsevier Inc. All rights reserved. The American Journal of Medicine (2012) 125, KEYWORDS: Acute myocardial infarction; Antiplatelet medication; Atrial fibrillation; Outcomes; Warfarin Atrial fibrillation is prevalent in clinical practice, with its occurrence overlapping substantially with acute and chronic coronary artery disease. 1 New-onset atrial fibrillation may acutely increase oxygen demand, contributing to myocardial ischemia. Several studies also have shown that atrial fibrillation is a frequent complication after acute myocardial infarction because of alterations in hemodynamics and neurohormonal activation, which often accompany acute cardiac events. 2-5 Its presence in the setting of acute myocardial infarction has been associated with worse short- and long-term outcomes. 6-9 Long-term oral anticoagulation is a pillar of atrial fibrillation treatment for prevention of stroke. Atrial fibrillation Funding: Bristol-Myers Squibb/Sanofi Pharmaceuticals Partnership and Schering-Plough Corporation are founding sponsors of the ACTION Registry-GWTG. Conflicts of Interest: Renato D. Lopes reports receiving research grants and consulting honoraria from Bristol-Myers Squibb. Christopher B. Granger, Tracy Y. Wang, and Eric D. Peterson have posted their conflict of interest information online at Karen P. Alexander, JoAnne M. Foody, Marjorie Funk, and Li Li have no conflicts to report. Authorship: All authors had access to the data and played a role in writing this manuscript. Requests for reprints should be addressed to Renato D. Lopes, MD, PhD, Box 3850, 2400 Pratt St, Room 0311 Terrace Level, Durham, NC address: renato.lopes@duke.edu /$ -see front matter 2012 Elsevier Inc. All rights reserved.

2 898 The American Journal of Medicine, Vol 125, No 9, September 2012 guidelines provide graded recommendations for the use of warfarin according to patients risk of stroke assessed by the Congestive heart failure, Hypertension, Age, Diabetes, Stroke [Doubled] (CHADS 2 ) score. 10 The combination of warfarin and antiplatelet medications after acute myocardial infarction is associated with bleeding and poses risk-benefit challenges in clinical practice Although several studies have reported the use of antithrombotic agents and outcomes in patients with atrial fibrillation after acute myocardial infarction, little is known about the use of antithrombotic agents and outcomes in patients with prior atrial fibrillation and acute myocardial infarction. Thus, we report the incidence and associated outcomes of patients with prior atrial fibrillation who present with acute myocardial infarction. In this population, we also describe the use of warfarin and other antithrombotic therapies overall and as a function of ischemic and bleeding risk. CLINICAL SIGNIFICANCE MATERIALS AND METHODS The National Cardiovascular Data Registry s Acute Coronary Treatment and Intervention Outcomes Network Registry (ACTION Registry)-Get With the Guidelines (GWTG) is a national quality improvement registry of ST-segment elevation myocardial infarction and non-st-segment elevation myocardial infarction that began enrolling patients on January 1, 2007 ( Patients are eligible for the ACTION Registry-GWTG if they present within 24 hours from onset of ischemic symptoms and receive a primary diagnosis of non-st-segment elevation myocardial infarction or ST-segment elevation myocardial infarction. Deidentified data are extracted from existing medical records on a web-based case report form by trained data collectors at each hospital. Protection of Human Subjects Study participation at all centers was approved by local institutional review boards. A data quality program ensures that consistent and reliable data are submitted and recorded. Quality assurance measures, such as data quality reports and random site audits by trained nurse abstractors, maximize the completeness and accuracy of all records submitted. Study Population The study population includes patients with ST-segment elevation myocardial infarction and patients with non-stsegment elevation myocardial infarction from 309 ACTION Registry-GWTG hospitals from July 1, 2008, to September Atrial fibrillation 2 weeks before hospitalization for acute myocardial infarction occurs in 7.1% of patients. Patients with atrial fibrillation are sicker and have worse outcomes than patients without atrial fibrillation. Less than 50% of these patients receive warfarin at discharge, even those with CHADS 2 scores 2. Triple therapy is used in only 14.6% of patients with atrial fibrillation. Prescribing warfarin is influenced modestly by ischemic and bleeding risks. 30, 2009 (N 69,543), who had data on version 2.0 of the data collection form where the variable of interest (occurrence of atrial fibrillation within 2 weeks before arrival at the hospital for the index myocardial infarction event) was collected. The presence of atrial fibrillation within 2 weeks before the index myocardial infarction was recorded via a check box on the data-collection form. Patients were excluded if atrial fibrillation status was missing (N 288), leaving a total population of 69,255 patients, of whom 7.1% (N 4947) had atrial fibrillation in the 2 weeks before the index myocardial infarction admission. Patients transferred out of participating ACTION hospitals were excluded from analyses related to in-hospital outcomes and discharge therapy (N 245). In addition, patients who died in-hospital or left against medical advice (N 762) were excluded when reporting discharge therapies. Inhospital procedures and discharge medications are reported among those without contraindications for each treatment. Furthermore, antithrombotic therapies used at discharge are shown according to stroke and bleeding risk at discharge. For this analysis, patients with missing data for calculating risk scores were excluded (N 104 missing CHADS 2 and N 200 missing bleeding score). Missing data for each discharge antithrombotic therapy (warfarin, aspirin, or clopidogrel) were less than 1%. Patients with missing information on the use of discharge therapies were excluded from these analyses. In-hospital Outcomes In-hospital clinical outcomes were described according to the presence or absence of atrial fibrillation 2 weeks before admission, and included in-hospital all-cause death, myocardial infarction, death or myocardial infarction, cardiogenic shock, stroke, major bleeding, overall and noncoronary artery bypass grafting related red blood cell transfusion, and hospital length of stay. Major bleeding was defined as an absolute hemoglobin decrease of 4 g/dl (baseline to nadir), intracranial hemorrhage, documented or suspected retroperitoneal bleed, any red blood cell transfusion with a baseline hemoglobin 9 g/dl, or any red blood cell transfusion with a baseline hemoglobin 9 g/dl and a suspected bleeding event. Bleeding in patients undergoing coronary artery bypass grafting was included as a bleed only if it occurred before surgery. Bleeding during or after surgery was not included in the bleeding definition.

3 Lopes et al Atrial and Outcomes in Patients with Acute Myocardial Infarction 899 CHADS 2 Score and Outpatient Bleeding Risk Index Score All patients were risk stratified for ischemic stroke and bleeding risk scores. The CHADS 2 score combines key factors that predict future risk of ischemic stroke in the atrial fibrillation population. 14 These factors include congestive heart failure, hypertension, age 75 years, diabetes, or stroke or transient ischemic attack. The score varies from 0 to 6 points. Patients at high risk (with a score of 2) have an indication for warfarin to prevent subsequent stroke. 15 The bleeding risk score the Outpatient Bleeding Risk Index has been used. 16 In brief, it is based on age 65 years, history of stroke, history of gastrointestinal bleed, plus 1 or more of the following: recent hematocrit 30%, serum creatinine concentration 1.5 mg/dl, history of diabetes mellitus, and recent myocardial infarction. Patients without any risk factors were categorized as low risk, patients with 1 to 2 risk factors were categorized as intermediate risk, and patients with 3 to 4 factors were categorized as high risk. Discharge antithrombotic strategy was shown across ischemic and bleeding risk strata for patients with atrial fibrillation and acute myocardial infarction. Statistical Analyses Counts and percentages were calculated for categorical variables, and medians (25th, 75th percentiles) was calculated for continuous variables. Baseline characteristics, treatment patterns, and in-hospital outcomes were compared according to the presence or absence of atrial fibrillation using the Wilcoxon rank-sum test for continuous variables and Pearson s chi-square test for categorical variables. To test for linear trend for antithrombotic therapies used at discharge by risk scores, logistic regression modeling was used where CHADS 2 risk score and bleeding risk score at discharge were entered separately into the model as ordinal variables. Because the use of antithrombotic therapy may be influenced by age, we performed a sensitivity analysis on the use of antithrombotic therapy at discharge in patients with and without atrial fibrillation, including only patients aged 75 years. A P value.05 was considered significant for all tests. All analyses were performed using SAS software (v 9.2, SAS Institute Inc, Cary, NC). RESULTS Atrial fibrillation occurring 2 weeks before hospitalization was present in 7.1% (N 4947) of the population with myocardial infarction. Compared with patients without atrial fibrillation, those with atrial fibrillation more often had ST-segment depression or transient ST-segment elevation (20.1% vs 14.8%, respectively) and less often had ST-segment elevation myocardial infarction (22.4% vs 41.4%, respectively). Patients with ST-segment elevation myocardial infarction had less atrial fibrillation when compared with patients with non-st-segment elevation myocardial infarction (4.0% vs 9.3%, respectively). Patients with atrial fibrillation also were older and more likely to have comorbidities (eg, previous myocardial infarction, diabetes, and hypertension) compared with patients without atrial fibrillation (Table 1). Patients with atrial fibrillation had more prior cardiac procedures and coronary artery bypass grafting. Of note, patients with atrial fibrillation were more likely to have prior congestive heart failure (34.8% vs 10.7%, P.0001) and prior stroke (16.3% vs 6.9%, P.0001) when compared with patients without atrial fibrillation. In-hospital Procedures Patients with atrial fibrillation generally received fewer cardiac procedures than patients without atrial fibrillation (Table 2), but a slight increase was observed in noninvasive stress testing among patients with atrial fibrillation (6.3%) versus no atrial fibrillation (4.5%) (P.0001). Patients with ST-segment elevation myocardial infarction with atrial fibrillation were less likely to receive reperfusion therapy with percutaneous coronary intervention or thrombolytics. When compared with patients with ST-segment elevation myocardial infarction without atrial fibrillation, those with atrial fibrillation were less likely to be treated with percutaneous coronary intervention (80.0% vs 88.1%). Similar results were seen for patients with non-st-segment elevation myocardial infarction, who received less catheterization and percutaneous coronary intervention. In both those with ST-segment elevation myocardial infarction and non-st-segment elevation myocardial infarction, those who had atrial fibrillation had longer delays from hospital admission to percutaneous coronary intervention and coronary artery bypass grafting than patients without atrial fibrillation. In patients with atrial fibrillation who underwent percutaneous coronary intervention with stent implantation, a bare mental stent was used more often than in patients without atrial fibrillation. These findings were observed in patients with ST-segment elevation myocardial infarction and patients with non-st-segment elevation myocardial infarction (Table 2). In-hospital Outcomes Unadjusted in-hospital outcomes are shown in Table 3. In general, when compared with patients without atrial fibrillation, patients with atrial fibrillation had worse in-hospital outcomes, such as death, cardiogenic shock, and major bleeding. Similar rates of myocardial infarction were seen among patients with and without atrial fibrillation. Patients with atrial fibrillation had a longer median hospital length of stay than patients without atrial fibrillation. Admission Medications Patients with atrial fibrillation were more likely to be taking evidence-based medication before hospitalization for acute myocardial infarction than were patients without atrial fi-

4 900 The American Journal of Medicine, Vol 125, No 9, September 2012 Table 1 Baseline Characteristics According to Presence of Atrial Characteristics No Atrial (N 64,308) Atrial (N 4947) P Value Age, median (25th, 75th), y 63 (53, 74) 78 (68, 84).0001 BMI, median (25th, 75th), kg/m (25.0, 32.5) 27.2 (23.6, 31.6).0001 Female sex Race.0001 White Black Asian Hispanic Medical history Current smoker Hypertension Dyslipidemia Diabetes Currently on dialysis Myocardial infarction Congestive heart failure Percutaneous coronary intervention Coronary artery bypass grafting Stroke Peripheral artery disease CHADS 2 score Bleeding score.0001 Low Intermediate High Signs and symptoms at presentation Heart failure Shock Heart rate, median (25th, 75th), bpm 80 (68, 95) 90 (74, 117).0001 SBP, median (25th, 75th), mm Hg 144 (123, 164) 134 (113, 156).0001 ECG findings STEMI ST-segment depression or transient ST-segment elevation BMI body mass index; bpm beats per minute; CHADS 2 Congestive heart failure, Hypertension, Age, Diabetes, Stroke (Doubled); SBP systolic blood pressure; ECG electrocardiogram; STEMI ST-segment elevation myocardial infarction. Values presented as percentages, unless otherwise indicated. brillation (Table 4). Before admission, only 32.5% of patients with atrial fibrillation were taking warfarin. Discharge Medications Overall, in patients with atrial fibrillation, warfarin at discharge was used in 41% (N 1526), aspirin was used in 95% (N 3389), and clopidogrel was used in 63.5% (N 2273). At discharge, patients with atrial fibrillation were less likely to receive evidence-based medication, such as aspirin, clopidogrel, betablockers, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and statins, than patients without atrial fibrillation (Table 4). The use of clopidogrel was significantly lower in patients with atrial fibrillation (63.5%) than in patients without atrial fibrillation (82.7%), and although it was used in 70.5% of patients with atrial fibrillation discharged without warfarin, it was used in only 50.5% of patients with atrial fibrillation receiving discharge warfarin. The use of antithrombotic therapy at discharge among patients with and without atrial fibrillation in the overall population was nearly identical to the use of antithrombotic therapy among patients with and without atrial fibrillation when only the older subgroup was analyzed (age 75 years, 27% of the overall population) (data not shown).

5 Lopes et al Atrial and Outcomes in Patients with Acute Myocardial Infarction 901 Table 2 In-hospital Procedures According to Presence of Atrial Procedures No Atrial (N 64, 308) Atrial (N 4947) P Value Noninvasive stress testing Diagnosis catheterization Catheterization within 24 h Catheterization within 48 h LVEF evaluated Coronary artery bypass grafting Time to catheterization (h), median (25th, 75th) 3.4 (0.9, 23.0) 20.0 (1.5, 51.6).0001 Time to coronary artery bypass grafting (h), median (25th, 75th) 69.4 (33.3, 116.7) (57.5, 162.0).0001 STEMI cases (N 26,651) (N 1106) Overall reperfusion Thrombolytic therapy Primary percutaneous coronary intervention Bare metal stent* Drug-eluting stent* Time to primary percutaneous coronary intervention (min), median (25th, 75th) 63 (49, 79) 68 (51, 83).003 NSTEMI cases (N 37,657) (N 3841) Overall revascularization therapy Percutaneous coronary intervention Bare metal stent* Drug-eluting stent* Time to percutaneous coronary intervention (h), median (25th, 75th) 19.2 (6.7, 36.9) 33.3 (15.7, 67.7).0001 LVEF left ventricular ejection fraction; STEMI ST-segment elevation myocardial infarction; NSTEMI non ST-segment elevation myocardial infarction. Values presented as percentages, unless otherwise indicated. *Percentage of patients receiving stents and not from the overall population. Stroke and Bleeding Risks Of the patients with atrial fibrillation and myocardial infarction, 3.7% were at low risk (CHADS 2 score 0), 12.9% were at moderate risk (CHADS 2 score 1), and 80.9% were at high risk (CHADS 2 score 2) for stroke. In terms of bleeding, 4.5% of the patients with atrial fibrillation and Table 3 In-hospital Clinical Events According to Presence of Atrial Events No Atrial (N 64,308) Atrial (N 4947) P Value Death Myocardial infarction Death or myocardial infarction Cardiogenic shock Stroke Major bleeding* Any red blood cell transfusion Red blood cell transfusion (non-cabg) Length of stay, median (25th, 75th), d 3 (2, 5) 5 (3, 8).0001 CABG coronary artery bypass grafting. Values presented as percentages, unless otherwise indicated. *Major bleeding was defined as an absolute hemoglobin decrease of 4 g/dl (baseline to nadir), intracranial hemorrhage, documented or suspected retroperitoneal bleed, any red blood cell transfusion with a baseline hemoglobin 9 g/dl, or any red blood cell transfusion with a baseline hemoglobin 9 g/dl and a suspected bleeding event.

6 902 The American Journal of Medicine, Vol 125, No 9, September 2012 Table 4 Home and Discharge Medications According to Presence of Atrial Medications No Atrial (N 64,308) Atrial (N 4947) P Value At-home medications Aspirin Clopidogrel Ticlopidine Beta-blocker ACEI or ARB Statin Warfarin Discharge medications Aspirin Clopidogrel Ticlopidine Prasugrel Beta-blocker ACEI or ARB Statin Warfarin ACEI angiotensin-converting enzyme inhibitor; ARB angiotensin receptor blocker. Values presented as percentages. myocardial infarction were at low risk, 75.9% were at moderate risk, and 14.4% were at high risk for bleeding. At discharge, warfarin use increased with higher CHADS 2 scores (P.0001); however, warfarin was still used in only 41.0% of all patients with atrial fibrillation, approximately a 10% increase compared with use before admission (32.5%). Of note, less than half of patients with CHADS 2 scores 2 (43.5%) were given warfarin at discharge (Figure 1). At discharge, warfarin use increased in patients with lowto-intermediate and high bleeding risk scores, but not in patients from intermediate-to-high risk. Warfarin was given to 25.4%, 42.3%, and 42.1% of patients at low, intermediate, and high risk of bleeding, respectively (P.004) (Figure 2). Triple therapy, defined as the combination of aspirin plus clopidogrel plus warfarin, was used in a small minority (14.6%) of patients. There was a trivial-to-nonexistent increase in the use of triple therapy in patients with higher CHADS 2 scores (Figure 3). In general, the use of triple therapy did not vary with bleeding risk, whereas warfarin increased modestly from low-to-intermediate risk or lowto-high risk for bleeding. The median CHADS 2 score across the bleeding categories was 1 (in the low category), 2 (in the intermediate category), and 5 (in the high category). DISCUSSION The use of postmyocardial infarction antithrombotic therapy in patients with atrial fibrillation is cautious. Less than half of patients with myocardial infarction with atrial fibrillation received warfarin at discharge, even those with CHADS 2 scores 2. In addition, clopidogrel was given at discharge to 63.5% of all patients with atrial fibrillation, to 70.5% of patients with atrial fibrillation who did not receive warfarin, and only to 50.5% of patients discharged with warfarin. Figure 1 Use of discharge antithrombotic therapy among patients with atrial fibrillation by CHADS 2 score. CHADS 2 Congestive heart failure, Hypertension, Age, Diabetes, Stroke (Doubled).

7 Lopes et al Atrial and Outcomes in Patients with Acute Myocardial Infarction 903 Figure 2 Use of discharge antithrombotic therapy among patients with atrial fibrillation by bleeding score at discharge. This was surprising, because clopidogrel is recommended after acute myocardial infarction and provides additional benefits in a population with atrial fibrillation not receiving warfarin. 17,18 Patients with acute myocardial infarction and atrial fibrillation are a particularly high-risk group, spurring speculation regarding the drivers of an antithrombotic treatment pattern for which risk of bleeding or ischemic events does not adequately explain observed care patterns. Previous studies have demonstrated that new-onset atrial fibrillation after an acute myocardial infarction is associated with worse outcomes. 6-9 In the present study, we showed that atrial fibrillation before an acute myocardial infarction also is associated with more in-hospital complications. Antiplatelet therapy, usually aspirin plus clopidogrel, in the setting of acute myocardial infarction improves clinical outcomes. 17,18 The benefit of warfarin in patients with atrial fibrillation for stroke prevention is similarly well established. 19 For patients with atrial fibrillation who are unable to take warfarin, aspirin plus clopidogrel has been shown to reduce the risk of stroke by 28%. 20 In our study, less than Figure 3 Use of discharge warfarin and triple therapy among patients with atrial fibrillation according to CHADS 2 score. CHADS2 Congestive heart failure, Hypertension, Age, Diabetes, Stroke (Doubled).

8 904 The American Journal of Medicine, Vol 125, No 9, September % of the patients who were not discharged with warfarin received clopidogrel, even when clopidogrel also is recommended postmyocardial infarction. 18,21 This illustrates the lack of a clear hierarchy in the management of antithrombotic therapy for patients with both atrial fibrillation and acute myocardial infarction in clinical practice. The in-hospital risk of bleeding increases with the number of antithrombotic agents used in patients with acute myocardial infarction. 12 By using aspirin as a reference, the use of clopidogrel, warfarin, aspirin plus clopidogrel, aspirin plus warfarin, clopidogrel plus warfarin, and triple therapy (aspirin plus clopidogrel plus warfarin) is associated with a progressively increased risk of bleeding (33%, 23%, 47%, 84%, a 3-fold and a 4-fold increase, respectively). Similar results were described in another large study of patients with chronic atrial fibrillation. 13 In this study, triple therapy was associated with a 3-fold increase in the risk of bleeding when compared with risk in patients using warfarin only. In patients with atrial fibrillation after non-stsegment elevation myocardial infarction, it was shown that warfarin was used only 13.5% of the time at discharge. 22 It is understandable for providers and patients to be cautious in light of such data because bleeding events in the setting of acute myocardial infarction are associated with worse short- and long-term outcomes Yet, similar to our study, in these pooled analyses from some clinical trials, the use of warfarin was independent of patients risk of stroke or bleeding. 22 This is somewhat different from what has been observed in a large cohort of patients with non-st-segment elevation myocardial infarction undergoing percutaneous coronary intervention, where the choice of antithrombotic strategy at the time of percutaneous coronary intervention was driven primarily by the patient s risk of bleeding. 29 In the situation in which patients had atrial fibrillation predating the index event, the choice of warfarin was driven more by the stroke risk than by the bleeding score, because patients were likely to use warfarin even with a higher bleeding score. Different clinical settings seem to variably influence perceived risks, as well as choice of therapy. With atrial fibrillation, only 41% of patients received warfarin and only 63.5% of patients received clopidogrel at discharge, resulting in undertreatment for both atrial fibrillation and acute myocardial infarction. Another important issue in the management of these patients is the choice of stent, which determines the use and duration of clopidogrel. In a subgroup analysis of the Assessment of Pexelizumab in Acute Myocardial Infarction trial, it was shown that among approximately 6000 patients with ST-segment elevation myocardial infarction treated with primary percutaneous coronary intervention, new-onset atrial fibrillation was common and associated with worse clinical outcomes. 8 In addition, among patients with atrial fibrillation and coronary stents, only 5.1% were discharged on triple therapy, indicating a strong preference toward stent protection and away from triple therapy in this setting. A bare metal stent also was more commonly used in patients with atrial fibrillation than in those without atrial fibrillation. We found similar results for both patients with STsegment elevation myocardial infarction and non-st-segment elevation myocardial infarction, when a bare metal stent was more often used in patients with atrial fibrillation than in patients without atrial fibrillation. 18,21 Thus, the use of bare metal stents should be strongly considered in patients with atrial fibrillation and acute myocardial infarction, particularly if they have a CHADS 2 score 2. The longterm efficacy and safety of triple therapy need further evaluation, especially in light of new antithrombotic agents under development that might change the current constituents of triple therapy. Finally, the average age of patients with atrial fibrillation is substantially higher than those without atrial fibrillation, as would be expected. This difference in age might have influenced observed outcomes in the atrial fibrillation group. Despite the observed similar pattern of underuse of antithrombotic therapy overall and among patients aged 75 years with atrial fibrillation, additional factors, such as concerns about safety in this population, may have played a role in our findings. Despite the potential concerns about age-related risks from antithrombotic therapy, age also remains a powerful risk factor for thrombotic complications and is not a contraindication for antithrombotic treatment. Therefore, CHADS 2 and bleeding risk scores (which take age into account) are important aids in identifying which patients warrant different combinations of antithrombotic therapy, including warfarin in high-risk groups with atrial fibrillation and myocardial infarction. LIMITATIONS Our study has several limitations. First, we did not collect timing, type, and duration of atrial fibrillation. We also did not use an electrocardiogram for the diagnosis of atrial fibrillation. Second, we do not have information on postdischarge outcomes, particularly bleeding events. Third, we did not collect information on long-term antithrombotic therapy adherence. Fourth, because we applied the CHADS 2 and bleeding scores at the time of hospital discharge, congestive heart failure and myocardial infarction during the hospitalization were considered history of these events for the purposes of the CHADS 2 score definition. Fifth, we did not collect information on rationale for treatment selection. This information would be helpful in understanding the decision-making process. Finally, although the difference in age of patients with and without atrial fibrillation is notable, management is better described by comparing risk profiles rather than age alone. CONCLUSIONS In this study, patients with atrial fibrillation and acute myocardial infarction were at higher risk for adverse outcomes; less than half received warfarin at discharge, and only approximately two thirds received clopidogrel. The prescribing of warfarin was influenced modestly by ischemic risk

9 Lopes et al Atrial and Outcomes in Patients with Acute Myocardial Infarction 905 (CHADS 2 score) and bleeding risk, so this undertreatment was not well explained. These findings highlight uncertainty in practice and the need for greater clarity about antithrombotic strategies for patients with the combined cardiovascular conditions of atrial fibrillation and acute myocardial infarction. References 1. Lloyd-Jones D, Adams RJ, Brown TM, et al. Heart disease and stroke statistics 2010 update. Circulation. 2010;121:e Pedersen OD, Baggar H, Kober L, Torp-Perdersen C. The occurrence and prognostic significance of atrial fibrillation/flutter following AMI. Eur Heart J. 1999;20: Pizzetti F, Turazza FM, Franzosi MG, et al. Incidence and prognostic significance of atrial fibrillation in AMI. Heart. 2001;86: Al-Khatib SM, Pieper KS, Lee KL, et al. Atrial fibrillation and mortality among patients with ACS without ST-segment elevation. Am J Cardiol. 2001;88:A7, Wong CK, White HD, Wilcox RG, et al. Significance of atrial fibrillation during AMI and its current management. Card Electrophysiol Rev. 2003;7: Mehta RH, Dabbous OH, Granger CB, et al. Comparison of outcomes of patients with ACS with and without atrial fibrillation. Am J Cardiol. 2003;92: Lopes RD, Pieper KS, Horton JR, et al. Short- and long-term outcomes following atrial fibrillation in patients with ACS with or without ST-segment elevation. Heart. 2008;94: Lopes RD, Elliott LE, White HD, et al. Antithrombotic therapy and outcomes of patients with atrial fibrillation following primary PCI. Eur Heart J. 2009;30: Jabre P, Jouven X, Adnet F, et al. Atrial fibrillation and death after MI. Circulation. 2011;123: Fuster V, Rydén LE, Cannom DS, et al ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation. Circulation. 2011;123:e Rothberg MB, Celestin C, Fiore LD, et al. Warfarin plus aspirin after MI or ACS: meta-analysis with estimates of risk and benefit. Ann Intern Med. 2005;143: Sørensen R, Hansen ML, Abildstrom SZ, et al. Risk of bleeding in patients with AMI treated with different combinations of aspirin, clopidogrel, and vitamin K antagonists in Denmark. Lancet. 2009;374: Hansen ML, Sørensen R, Clausen MT, et al. Risk of bleeding with single, dual, or triple therapy with warfarin, aspirin, and clopidogrel in patients with atrial fibrillation. Arch Intern Med. 2010;170: Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke. JAMA. 2001;285: Fuster V, Rydén LE, Cannom DS, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation. Circulation. 2006;114:e Beyth RJ, Quinn LM, Landefeld CS. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am J Med. 1998;105: Yusuf S, Zhao F, Mehta SR, et al. Effects of clopidogrel in addition to aspirin in patients with ACS without ST-segment elevation. N Engl J Med. 2001;345: Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/nstemi. J Am Coll Cardiol. 2007;50:e Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med. 2007;146: Connolly SJ, Pogue J, Hart RG, et al. Effect of clopidogrel added to aspirin in patients with atrial fibrillation. N Engl J Med. 2009;360: Antman EM, Hand M, Armstrong PW, et al Focused update of the ACC/AHA 2004 guidelines for the management of patients with STEMI. Circulation. 2008;117: Lopes RD, Starr A, Pieper CF, et al. Warfarin use and outcomes in patients with atrial fibrillation complicating ACS. Am J Med. 2010; 123: Spencer FA, Moscucci M, Granger CB, et al. Does comorbidity account for the excess mortality in patients with major bleeding in AMI? Circulation. 2007;116: Wang TY, Xiao L, Alexander KP, et al. Antiplatelet therapy use after discharge among AMI patients with in-hospital bleeding. Circulation. 2008;118: Eikelboom JW, Mehta SR, Anand SS, et al. Adverse impact of bleeding on prognosis in patients with ACS. Circulation. 2006;114: Mehran R, Pocock SJ, Nikolsky E, et al. A risk score to predict bleeding in patients with ACS. J Am Coll Cardiol. 2010;55: Manoukian SV, Feit F, Mehran R, et al. Impact of major bleeding on 30-day mortality and clinical outcomes in patients with ACS. JAm Coll Cardiol. 2007;49: Lopes RD, Alexander KP, Chen AY, et al. Major bleeding is an independent predictor of 1-year post-discharge mortality in patients with NSTE ACS. [ACC abstract ]. J Am Coll Cardiol. 2010; 55:A124.E Lopes RD, Peterson ED, Chen AY, et al. Antithrombotic strategy in NSTEMI patients undergoing PCI. JACC Cardiovasc Interv. 2010;3: