Implications of prior myocardial infarction for patients presenting with an acute myocardial infarction Lan Shen, MD, a,b Bimal R. Shah, MD, MBA, a Aro Nam, MS, a DaJuanicia Holmes, MS, a Karen P. Alexander, MD, a Deepak L. Bhatt, MD, MPH, c P. Michael Ho, MD, d Eric D. Peterson, MD, MPH, a Ben He, MD, b and Matthew T. Roe, MD, MHS a Durham, NC; Shanghai, China; Boston, MA; and Denver, CO
Background Prior myocardial infarction (MI) is a known risk factor for long-term mortality among acute MI patients; but its prevalence and implications for the short-term outcomes of patients with a new, acute MI remain uncertain. Methods We studied a total of 319,152 consecutively enrolled ST-segment elevation MI (STEMI) and non-STEMI (NSTEMI) patients in the National Cardiovascular Data Registry Acute Coronary Treatment and Intervention Outcomes Network Registry—Get With The Guidelines (01/2007-03/2012). Baseline characteristics, home and in-hospital treatments, mortality rates, and major bleeding were compared separately for STEMI and NSTEMI by prior MI status, with adjustment for mortality and major bleeding. Results Prior MI was documented in 19% of STEMI (n = 124,535) and 29% of NSTEMI (n = 194,617) patients, who were older, were more likely to have comorbidities or prior revascularization, and were more commonly taking secondary prevention medications at home. Guideline-recommended treatments in-hospital and at discharge did not differ in prior-MI STEMI patients, but invasive management was lower for prior-MI NSTEMI patients. The frequency of in-hospital mortality was higher for prior-MI STEMI (5.9% vs 5.2%) and NSTEMI patients (4.3% vs 3.4%). After adjustment, the excess mortality risk associated with prior MI was no longer present for STEMI (odds ratio = 1.06, 95% CI 0.97-1.15), with only modest excess risk for NSTEMI (odds ratio = 1.10, 95% CI 1.04-1.15). The risk of in-hospital major bleeding was marginally lower for prior-MI NSTEMI. Conclusion More than 20% of patients with acute MI treated in contemporary practice have a history of a prior MI; despite differences in the baseline risk profile, there was little difference in the adjusted risk of in-hospital mortality by prior-MI status. (Am Heart J 2014;0:1-6.)
Despite the use of secondary prevention therapies, the rate of acute myocardial infarction (MI) recurrence among patients with a prior history of MI remains high. Prior studies have shown that N20% of patients presenting with an MI have had a prior MI, 1,2 and these patients tend to have worse long-term outcomes compared with patients without prior MI. 3 As a result, prior MI has been identified as a known risk factor for an increased risk of adverse outcomes and is consequently included as a variable in the Thrombolysis in Myocardial Infarction and Global Registry of Acute Cardiac Events risk scores. 4,5 Nonetheless, the contemporary prevalence of prior MI in
From the aDuke Clinical Research Institute, Durham, NC, bShanghai Renji Hospital, Department of Cardiology, Shanghai, China, cVeterans Affairs Boston Healthcare System, Brigham and Women's Hospital, and Harvard Medical School, Boston, MA, and d Veterans Affairs Eastern Colorado Healthcare System, and University of Colorado, Denver, CO. Submitted September 11, 2013; accepted March 29, 2014. Reprint requests: Matthew T. Roe, MD, MHS, Duke Clinical Research Institute, 2400 Pratt St, Durham, NC 27705. E-mail: [email protected] http://dx.doi.org/10.1016/j.ahj.2014.03.009 0002-8703 © 2014, Mosby, Inc. All rights reserved.
ST-segment elevation MI (STEMI) versus non-STEMI (NSTEMI) and its association with treatments and shortterm outcomes among patients presenting with a new MI event have not been evaluated. Using a large contemporary dataset, we sought to (1) determine the differential prevalence of prior MI by STEMI versus NSTEMI, (2) assess differences in baseline characteristics and treatment patterns by prior-MI status for STEMI versus NSTEMI, and (3) delineate the impact of prior MI on major in-hospital outcomes in STEMI versus NSTEMI. Using clinical data from National Cardiovascular Data Registry (NCDR) Acute Coronary Treatment and Intervention Outcomes Network Registry—Get With The Guidelines (ACTION Registry-GWTG), we examine the association between prior MI and in-hospital outcomes.
Methods Data collection and patient population The ACTION Registry-GWTG is a voluntary registry and quality improvement program that began on January 1, 2007. Details of this registry’s effort have been previously defined. 6 Briefly, patients were eligible for the ACTION Registry-GWTG if they presented within 24 hours of the onset of an ischemic
American Heart Journal Month Year
2 Shen et al
syndrome and if the final primary diagnosis was acute MI (either STEMI or NSTEMI). Data were abstracted by a trained data collector at each hospital. Variables collected included prehospital data, medical history, treatments administered, as well as associated major contraindications to evidence-based therapies and in-hospital outcomes. Prior MI was defined in the data definition dictionary as patients who have had at least one documented previous MI at any time between birth and arrival at the first facility for the index MI event. STEMI, NSTEMI, inhospital mortality, and prior MI were defined by the NCDR ACTION Registry-GWTG v2.2 Data Dictionary Full Specifications. 7 Before data were submitted, all participating sites received local institutional review board approval. The population for the current study was derived from 370,476 patients enrolled in the ACTION Registry-GWTG using the data collection form used from January 1, 2007, to March 31, 2012, at 605 hospitals. Of these patients, we excluded the following: patients from sites without coronary artery bypass graft (CABG) surgery capabilities to fairly compare the use of invasive procedures among the analysis categories (n = 41,360), transferout patients who did not have clinical information collected after transfer because of privacy rules (n = 6,159), patients with cardiac arrest on presentation (this variable did not start to be collected until 2011, so we could not ascertain the frequency of this variable in the earlier time frame of our analysis time period) (n = 3,387), and patients for whom information on previous MI was missing (n = 418). The final population included 319,152 patients (STEMI = 124,535 and NSTEMI = 194,617) from 446 hospitals.
Statistical analysis Patient demographics, cardiac risk factors, and presentation features were described for patients with prior MI and patients without prior MI. Continuous variables were expressed as means (standard deviation) and medians (interquartile range) and compared with the Wilcoxon rank test. Categorical variables were presented as frequencies and percentages and compared with the χ 2 test. The primary outcomes of interest were in-hospital mortality and in-hospital major bleeding. To evaluate the association of prior MI with in-hospital outcomes (mortality and major bleeding), logistic generalized estimating equations 8 method with exchangeable working correlation matrix was used to account for within-hospital clustering, as patients within hospitals tend to have more similar responses relative to patients at other hospitals. This method produces estimates similar to those from ordinary logistic regression, but variances are adjusted for the correlation of outcomes within a hospital. Variables from the ACTION Registry-GWTG in-hospital mortality model 9 (c-index = 0.84) that include patient demographics, medical history, laboratory values, and home medication use were used for in-hospital mortality adjustment. Additionally, variables from the ACTION Registry-GWTG major bleeding model (c-index = 0.71) 10 that include patient demographics, medical history, laboratory values, and home medication use were used for in-hospital major bleeding adjustment. In-hospital major bleeding (defined as an absolute hemoglobin decrease) and home medication use were used for in-hospital major bleeding adjustment. In-hospital major bleeding was defined as an absolute hemoglobin decrease of ≥4 g/dL, intracranial hemorrhage, any red blood cell transfusion with baseline hemoglobin ≥9 g/dL, or red blood cell transfusion
if hemoglobin b9 g/dL and a suspected bleeding event. Adjusted associations for outcomes were displayed as odds ratios (ORs) and 95% CIs. A 2-sided P b .05 was established as the level of statistical significance for all tests. All analyses were performed using SAS software (version 9.2; SAS Institute, Cary, NC). This research was supported by the American College of Cardiology Foundation’s NCDR. The views expressed in this article represent those of the authors and do not necessarily represent the official views of the NCDR or its associated professional societies identified at www.ncdr.com. The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper, and its final contents.
Results Frequency of prior MI The prevalence of prior MI was greater for NSTEMI patients versus STEMI patients (29% vs 19% over the entire time period). Over the 5-year analysis time period, the temporal frequency of prior MI was similar for STEMI patients (trend P value = .32) but increased for NSTEMI patients (27.3% in quarter 1, 2007, to 30.2% in Quarter 1, 2012; trend P value b .0001) (Figure). Baseline characteristics, treatments, and procedures Both for STEMI and NSTEMI patients, prior-MI patients were more likely to be older, be male, and have comorbidities such as hypertension, hyperlipidemia, and diabetes; were more likely to have a prior history of revascularization and congestive heart failure; and were more likely to be treated with secondary prevention medications at home compared with those without prior MI (Table I). Medication use within 24 hours of arrival and inhospital procedure use were similar by prior-MI status in STEMI patients. For NSTEMI, prior-MI patients were shown to have higher use of thienopyridine and statins, lower use of glycoprotein IIb-IIIa inhibitors, and lower use of angiography and revascularization procedures (Table II). The use of discharge medications and interventions was mostly similar by prior-MI status for both STEMI and NSTEMI patients. (Table III). In-hospital outcomes The frequency of in-hospital mortality was higher in prior-MI patients, both for STEMI and NSTEMI patients. However, after adjustment, the excess mortality risk was not apparent for prior-MI status for STEMI (adjusted OR = 1.06, 95% CI 0.97-1.15) but was marginally higher for prior MI in NSTEMI (adjusted OR = 1.10, 95% CI 1.041.15) (Table IV). The adjusted risk of in-hospital major bleeding was also similar among patients with (versus without) prior MI for STEMI but was marginally lower for NSTEMI (Table IV).
American Heart Journal Volume 0, Number 0
Figure
Patients with prior MI.Proportion of patients with prior MI among STEMI versus NSTEMI by quarter (2007 Q1-2012 Q1).
Discussion Among patients with an index acute MI, a history of prior MI was more prevalent in those whose index acute MI was NSTEMI compared with STEMI. Patients with a history of prior MI were older and had more comorbidities compared with patients without prior MI and were more commonly treated with secondary prevention medications at home prior to admission. Subsequently, those with prior MI have a higher risk of unadjusted inhospital mortality; however, after adjustment, there was a modestly excessive risk of in-hospital mortality only for NSTEMI patients, with no increased risk with prior MI for STEMI patients. Data from previous ACTION Registry-GWTG analyses have demonstrated that approximately 20% of MI patients have had a prior MI, but these results were not differentiated by MI classification (STEMI vs NSTEMI). 11 Another study showed that NSTEMI patients have a higher rate of prior MI compared with STEMI patients (42.6% vs 23.1%, respectively). 3 Similarly, we also found that prior MI is more likely in NSTEMI versus STEMI patients (29% vs 19%, respectively); so our findings are consistent with previous studies. 12 Nevertheless, we found that prior MI was associated with older age, a greater prevalence of cardiovascular risk factors, more frequent prior or current heart failure, and lower creatinine clearance (CrCl) values. Yet although patients with a prior MI were more likely to be on secondary prevention medications at home, only about 70% were on aspirin and b65% were on a statin. Therefore, patients with a prior MI have a greater burden of comorbidities when they present with an index acute MI; furthermore, they appear to be suboptimally treated with secondary prevention medications in the chronic setting. 13,14 The findings from our study revealed that differences in mortality may be partially explained by differences in baseline characteristics and home medications between those with versus without a prior MI, as the higher
Shen et al 3
unadjusted risks of in-hospital mortality associated with prior MI disappeared in STEMI and were attenuated in NSTEMI with risk adjustment. Prior MI has been considered a high-risk feature in several recent acute coronary syndrome clinical trials that enrolled unstable angina/NSTEMI patients and used prior MI as one of the enrichment criteria to enhance long-term cardiovascular event rates 15–17 because prior MI has proven to be independently associated with worse outcomes through 6 months post-NSTEMI in both the Thrombolysis in Myocardial Infarction and Global Registry of Acute Cardiac Events risk models. 4,5,18,19 Although our study was limited to the in-hospital period, we have shown that there is only a modest increase in the risk of in-hospital mortality with this key baseline characteristic—a finding that highlights the differential characteristics associated with the risk of early versus later mortality following acute MI presentation. Historically, adherence to guideline-recommended therapies after MI was significantly lower in NSTEMI patients compared with STEMI patients. 20,21 Despite recent improvements in the guideline adherence gap for NSTEMI versus STEMI patients, 22 we found that, in a contemporary population of both NSTEMI and STEMI patients, guideline adherence was relatively similar except for a lower use of angiography, revascularization, discharge thienopyridines, and cardiac rehabilitation referral for NSTEMI versus STEMI patients. Our study findings provide further insight into these treatment disparities by delineating that the presence of a prior MI did not appear to impact the quality of care for STEMI patients but was associated with a lower use of angiography and revascularization for NSTEMI patients (compared with NSTEMI patients without a prior MI). These findings are likely explained by the much higher incidence of prior percutaneous coronary intervention (PCI) or CABG and relatively lower CrCl values seen with prior MI in NSTEMI patients. These features likely impact the risk-benefit calculations that guide decision making for an early invasive management strategy that concurrently influences the use of antiplatelet therapies and referral for cardiac rehabilitation in NSTEMI patients.
Study limitations Our study has several limitations. First, despite the fact that we have 307,646 consecutive MI patients from 446 hospitals, the participating centers were voluntarily participating in the ACTION Registry-GWTG; therefore, these finding may not apply to patients and hospitals that differ significantly from those participating in the ACTION Registry-GWTG. Second, the detection of prior MI was abstracted from medical records, which are dependent upon the documentation accuracy of each hospital. Therefore, we could not ascertain when the prior MI happened or whether the index MI events
American Heart Journal Month Year
4 Shen et al
Table I. Baseline clinical and home medications *
Variables Demographics Age (y) BMI (kg/m2) Female sex (%) Race (%) White African American Asian Hispanic Other Medical history (%) Hypertension Diabetes mellitus Peripheral arterial disease Current/recent smoking Hyperlipidemia Prior PCI Prior CABG Prior heart failure Prior stroke Currently on dialysis Presentation features Initial heart rate (beat/min) Initial systolic BP (mm Hg) Cardiogenic shock (%) Signs of heart failure (%) ST depression or transient ST elevation (%) Initial laboratory results Baseline troponin ratio, ×ULN Hemoglobin (g/dL) Serum creatinine (mg/dL) CrCl (mL/min) ACTION mortality risk score Home medications (%) Aspirin Thienopyridine † β-Blocker Statin
STEMI
NSTEMI
n = 124,535
n = 194,617
No prior MI
Prior MI
No prior MI
Prior MI
n = 100,811
n = 23,724
n = 137,867
n = 56,750
81%
19%
71%
29%
60 (51-70) 28 (25-32) 31
62 (53-72) 28 (25-32) 26
66 (55-77) 28 (25-33) 40
69 (59-79) 28 (25-33) 34
84 8 2 4 1
84 9 1 4 1
82 11 1 4 1
83 11 1 4 1
59 22 4 43 46 8 4 3 4 1
82 32 11 46 78 70 20 14 9 2
73 32 9 31 57 13 11 11 8 3
89 46 20 30 81 59 38 31 14 5
78 (66-93) 140 (120-161) 7 9
78 (66-93) 137 (116-158) 7 13
83 (70-98) 147 (127-168) 2 17 25
84 (71-100) 144 (123-165) 2 26 25
1.2 (0.3-14.4) 14 (13-16) 1 (0.9-1.2) 88 (63-114) 32 (27-38)
0.9 (0.2-7.0) 14 (13-15) 1 (0.9-1.3) 83 (58-112) 33 (28-40)
2.5 (0.6-13) 14 (12-15) 1 (0.9-1.3) 78 (51-109) 28 (22-35)
1.8 (0.4-9.2) 13 (12-15) 1.1 (0.9-1.5) 68 (44-99) 31 (24-38)
26 5 20 24
64 29 58 57
40 11 34 36
71 37 69 65
BMI, Body mass index; BP, blood pressure; ULN, times the upper limit of normal. * All values reported as percentage or median (interquartile range) unless otherwise specified. † Clopidogrel or prasugrel.
reported in the ACTION Registry-GWTG were spontaneous MI or stent thrombosis MI events (for patients who had undergone prior PCI). Furthermore, we could not determine the location of the infarct-related artery or the position of the culprit lesion within that artery; so we cannot comment on the potential differences in the pathophysiology of the index MI events in prior-MI versus no-prior-MI patients. Similarly, the records did not
capture other medical features that might influence treatment decisions and outcomes such as infectious or malignant disease. Finally, only in-hospital outcomes are collected in the ACTION Registry-GWTG; therefore, the long-term prognostic implications of prior MI for STEMI versus NSTEMI patients could not be assessed, and the impact of prior MI is more likely to be manifest with longer follow-up.
American Heart Journal Volume 0, Number 0
Shen et al 5
Table II. In-hospital treatments
Variables
Table IV. In-hospital mortality and major bleeding by prior-MI status Unadjusted outcomes
STEMI
NSTEMI
n = 124,535
n = 194,617
No prior MI
Prior MI
No prior MI
Prior MI
n= 100,811
n= 23,724
n= 137,867
n= 56,750
99 89 64 73 93 96
99 88 62 73 93 95
97 56 33 60 88 92
97 61 27 66 89 91
96 86 7
95 86 6
83 50 13
75 44 8
Acute medication (%) Aspirin Thienopyridine* GP IIb/IIIa inhibitor Statin β-Blocker Anticoagulants Invasive procedure (%) Angiography PCI CABG
GP IIb/IIIa, Glycoprotein IIb/IIIa. * Clopidogrel or prasugrel.
Table III. Discharge medications and interventions STEMI
NSTEMI
n = 124,535
n = 194,617
No prior No prior MI Prior MI MI Prior MI
Variables
n= 100,811
n= 23,724
99 91 95 97 84
99 91 95 98 82
97 72 89 95 75
98 77 91 97 72
98
98
97
97
Discharge medication (%) Aspirin Thienopyridine* Statin β-Blocker Cardiac rehabilitation referral (%) Smoking cessation counseling (%)
n= n= 137,867 56,750
* Either clopidogrel or prasugrel.
Conclusions The prevalence of prior MI is higher among NSTEMI versus STEMI patients and is associated with a greater burden of co-morbidities on presentation. Nonetheless, we found only a modest independent incremental risk of in-hospital mortality with prior MI in NSTEMI patients but not in STEMI patients.
STEMI Mortality Major bleeding NSTEMI Mortality Major bleeding
Unadjusted OR*
Adjusted OR*
No prior MI
Prior MI
(95% CI)
(95% CI)
5.2 10.3
5.9 10.5
1.15 (1.08-1.22) 1.01 (0.97-1.06)
1.06 (0.97-1.15) 0.97 (0.92-1.03)
3.4 8.2
4.3 9.5
1.28 (1.22-1.34) 1.18 (1.14-1.23)
1.10 (1.04-1.15) 0.96 (0.92-0.99)
* OR of the outcomes between prior MI and no prior MI (reference group).
References 1. Anavekar NS, McMurray JJV, Velazquez EJ, et al. Relation between renal dysfunction and cardiovascular outcomes after myocardial infarction. N Engl J Med 2004;351:1285-95. 2. Roe MT, Chen AY, Cannon CP, et al. Temporal changes in the use of drug-eluting stents for patients with non–ST-segment elevation myocardial infarction undergoing percutaneous coronary intervention from 2006 to 2008: results from the Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA guidelines (CRUSADE) and Acute Coronary Treatment and Intervention Outcomes Network—Get With The Guidelines (ACTionGWTG) registries. Circ Cardiovasc Qual Outcomes 2009;2:414-20. 3. Motivala AA, Tamhane U, Ramanath VS, et al. A prior myocardial infarction: how does it affect management and outcomes in recurrent acute coronary syndromes? Clin Cardiol 2008;31:590-6. 4. Eagle KA, Lim MJ, Dabbous OH, et al. A validated prediction model for all forms of acute coronary syndrome: estimating the risk of 6-month postdischarge death in an international registry. JAMA 2004;291:2727-33. 5. Morrow DA, Antman EM, Charlesworth A, et al. TIMI risk score for ST-elevation myocardial infarction: a convenient, bedside, clinical score for risk assessment at presentation: an intravenous nPA for treatment of infarcting myocardium early II trial substudy. Circulation 2000;102:2031-7. 6. Peterson ED, Roe MT, Rumsfeld JS, et al. A call to ACTION (Acute Coronary Treatment and Intervention Outcomes Network): a national effort to promote timely clinical feedback and support continuous quality improvement for acute myocardial infarction. Circ Cardiovasc Qual Outcomes 2009;2:491-9. 7. NCDR® ACTION Registry®-GWTG™ v2.2 coder’s data dictionary. National Cardiovascular Data Registry web site, https://www.ncdr. com/WebNCDR/docs/public-data-collection-documents/ action_v2_codersdictionary_2-2.pdf?sfvrsn=2. [Updated February 3, 2011. Accessed March 6, 2014]. 8. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121-30. 9. Chin CT, Chen AY, Wang TY, et al. Risk adjustment for in-hospital mortality of contemporary patients with acute myocardial infarction: the Acute Coronary Treatment and Intervention Outcomes Network (ACTION) Registry—Get With The Guidelines (GWTG) acute myocardial infarction mortality model and risk score. Am Heart J 2011;161:113-122.e2.
6 Shen et al
10. Mathews R, Peterson ED, Chen AY, et al. In-hospital major bleeding during ST-elevation and non–ST-elevation myocardial infarction care: derivation and validation of a model from the ACTION Registry®-GWTG™. Am J Cardiol 2011;107:1136-43. 11. Lopes RD, Peterson ED, Chen AY, et al. Antithrombotic strategy in non–ST-segment elevation myocardial infarction patients undergoing percutaneous coronary intervention: insights from the ACTION (Acute Coronary Treatment and Intervention Outcomes Network) Registry. JACC Cardiovasc Interv 2010;3:669-77. 12. Haffner SM, Lehto S, Rönnemaa T, et al. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction. N Engl J Med 1998;339:229-34. 13. Alexander KP, Chen AY, Newby LK, et al. Sex differences in major bleeding with glycoprotein IIb/IIIa inhibitors: results from the CRUSADE (Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA guidelines) initiative. Circulation 2006;114:1380-7. 14. Bhatt DL, Roe MT, Peterson ED, et al. Utilization of early invasive management strategies for high-risk patients with non–ST-segment elevation acute coronary syndromes: results from the CRUSADE Quality Improvement Initiative. JAMA 2004;292:2096-104. 15. Alexander JH, Lopes RD, James S, et al. Apixaban with antiplatelet therapy after acute coronary syndrome. N Engl J Med 2011;365: 699-708. 16. Roe MT, Armstrong PW, Fox KA, et al. Prasugrel versus clopidogrel for acute coronary syndromes without revascularization. N Engl J Med 2012;367:1297-309. 17. Tricoci P, Huang Z, Held C, et al. Thrombin-receptor antagonist vorapaxar in acute coronary syndromes. N Engl J Med 2012;366:20-33.
American Heart Journal Month Year
18. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients with Unstable Angina/Non ST-elevation Myocardial Infarction): developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons: endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. Circulation 2007;116:e148-304. 19. Hamm CW, Bassand JP, Agewall S, et al. ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: the Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J 2011;32:2999-3054. 20. Montalescot G, Dallongeville J, Van Belle E, et al. STEMI and NSTEMI: are they so different? 1 year outcomes in acute myocardial infarction as defined by the ESC/ACC definition (the OPERA registry). Eur Heart J 2007;28:1409-17. 21. Roe MT, Parsons LS, Pollack Jr CV, et al. Quality of care by classification of myocardial infarction: treatment patterns for ST-segment elevation vs non–ST-segment elevation myocardial infarction. Arch Int Med 2005;165:1630-6. 22. Somma KA, Bhatt DL, Fonarow GC, et al. Guideline adherence after ST-segment elevation versus non-ST segment elevation myocardial infarction. Circ Cardiovasc Qual Outcomes 2012;5: 654-61.
Background Prior myocardial infarction (MI) is a known risk factor for long-term mortality among acute MI patients; but its prevalence and implications for the short-term outcomes of patients with a new, acute MI remain uncertain. Methods We studied a total of 319,152 consecutively enrolled ST-segment elevation MI (STEMI) and non-STEMI (NSTEMI) patients in the National Cardiovascular Data Registry Acute Coronary Treatment and Intervention Outcomes Network Registry—Get With The Guidelines (01/2007-03/2012). Baseline characteristics, home and in-hospital treatments, mortality rates, and major bleeding were compared separately for STEMI and NSTEMI by prior MI status, with adjustment for mortality and major bleeding. Results Prior MI was documented in 19% of STEMI (n = 124,535) and 29% of NSTEMI (n = 194,617) patients, who were older, were more likely to have comorbidities or prior revascularization, and were more commonly taking secondary prevention medications at home. Guideline-recommended treatments in-hospital and at discharge did not differ in prior-MI STEMI patients, but invasive management was lower for prior-MI NSTEMI patients. The frequency of in-hospital mortality was higher for prior-MI STEMI (5.9% vs 5.2%) and NSTEMI patients (4.3% vs 3.4%). After adjustment, the excess mortality risk associated with prior MI was no longer present for STEMI (odds ratio = 1.06, 95% CI 0.97-1.15), with only modest excess risk for NSTEMI (odds ratio = 1.10, 95% CI 1.04-1.15). The risk of in-hospital major bleeding was marginally lower for prior-MI NSTEMI. Conclusion More than 20% of patients with acute MI treated in contemporary practice have a history of a prior MI; despite differences in the baseline risk profile, there was little difference in the adjusted risk of in-hospital mortality by prior-MI status. (Am Heart J 2014;0:1-6.)
Despite the use of secondary prevention therapies, the rate of acute myocardial infarction (MI) recurrence among patients with a prior history of MI remains high. Prior studies have shown that N20% of patients presenting with an MI have had a prior MI, 1,2 and these patients tend to have worse long-term outcomes compared with patients without prior MI. 3 As a result, prior MI has been identified as a known risk factor for an increased risk of adverse outcomes and is consequently included as a variable in the Thrombolysis in Myocardial Infarction and Global Registry of Acute Cardiac Events risk scores. 4,5 Nonetheless, the contemporary prevalence of prior MI in
From the aDuke Clinical Research Institute, Durham, NC, bShanghai Renji Hospital, Department of Cardiology, Shanghai, China, cVeterans Affairs Boston Healthcare System, Brigham and Women's Hospital, and Harvard Medical School, Boston, MA, and d Veterans Affairs Eastern Colorado Healthcare System, and University of Colorado, Denver, CO. Submitted September 11, 2013; accepted March 29, 2014. Reprint requests: Matthew T. Roe, MD, MHS, Duke Clinical Research Institute, 2400 Pratt St, Durham, NC 27705. E-mail: [email protected] http://dx.doi.org/10.1016/j.ahj.2014.03.009 0002-8703 © 2014, Mosby, Inc. All rights reserved.
ST-segment elevation MI (STEMI) versus non-STEMI (NSTEMI) and its association with treatments and shortterm outcomes among patients presenting with a new MI event have not been evaluated. Using a large contemporary dataset, we sought to (1) determine the differential prevalence of prior MI by STEMI versus NSTEMI, (2) assess differences in baseline characteristics and treatment patterns by prior-MI status for STEMI versus NSTEMI, and (3) delineate the impact of prior MI on major in-hospital outcomes in STEMI versus NSTEMI. Using clinical data from National Cardiovascular Data Registry (NCDR) Acute Coronary Treatment and Intervention Outcomes Network Registry—Get With The Guidelines (ACTION Registry-GWTG), we examine the association between prior MI and in-hospital outcomes.
Methods Data collection and patient population The ACTION Registry-GWTG is a voluntary registry and quality improvement program that began on January 1, 2007. Details of this registry’s effort have been previously defined. 6 Briefly, patients were eligible for the ACTION Registry-GWTG if they presented within 24 hours of the onset of an ischemic
American Heart Journal Month Year
2 Shen et al
syndrome and if the final primary diagnosis was acute MI (either STEMI or NSTEMI). Data were abstracted by a trained data collector at each hospital. Variables collected included prehospital data, medical history, treatments administered, as well as associated major contraindications to evidence-based therapies and in-hospital outcomes. Prior MI was defined in the data definition dictionary as patients who have had at least one documented previous MI at any time between birth and arrival at the first facility for the index MI event. STEMI, NSTEMI, inhospital mortality, and prior MI were defined by the NCDR ACTION Registry-GWTG v2.2 Data Dictionary Full Specifications. 7 Before data were submitted, all participating sites received local institutional review board approval. The population for the current study was derived from 370,476 patients enrolled in the ACTION Registry-GWTG using the data collection form used from January 1, 2007, to March 31, 2012, at 605 hospitals. Of these patients, we excluded the following: patients from sites without coronary artery bypass graft (CABG) surgery capabilities to fairly compare the use of invasive procedures among the analysis categories (n = 41,360), transferout patients who did not have clinical information collected after transfer because of privacy rules (n = 6,159), patients with cardiac arrest on presentation (this variable did not start to be collected until 2011, so we could not ascertain the frequency of this variable in the earlier time frame of our analysis time period) (n = 3,387), and patients for whom information on previous MI was missing (n = 418). The final population included 319,152 patients (STEMI = 124,535 and NSTEMI = 194,617) from 446 hospitals.
Statistical analysis Patient demographics, cardiac risk factors, and presentation features were described for patients with prior MI and patients without prior MI. Continuous variables were expressed as means (standard deviation) and medians (interquartile range) and compared with the Wilcoxon rank test. Categorical variables were presented as frequencies and percentages and compared with the χ 2 test. The primary outcomes of interest were in-hospital mortality and in-hospital major bleeding. To evaluate the association of prior MI with in-hospital outcomes (mortality and major bleeding), logistic generalized estimating equations 8 method with exchangeable working correlation matrix was used to account for within-hospital clustering, as patients within hospitals tend to have more similar responses relative to patients at other hospitals. This method produces estimates similar to those from ordinary logistic regression, but variances are adjusted for the correlation of outcomes within a hospital. Variables from the ACTION Registry-GWTG in-hospital mortality model 9 (c-index = 0.84) that include patient demographics, medical history, laboratory values, and home medication use were used for in-hospital mortality adjustment. Additionally, variables from the ACTION Registry-GWTG major bleeding model (c-index = 0.71) 10 that include patient demographics, medical history, laboratory values, and home medication use were used for in-hospital major bleeding adjustment. In-hospital major bleeding (defined as an absolute hemoglobin decrease) and home medication use were used for in-hospital major bleeding adjustment. In-hospital major bleeding was defined as an absolute hemoglobin decrease of ≥4 g/dL, intracranial hemorrhage, any red blood cell transfusion with baseline hemoglobin ≥9 g/dL, or red blood cell transfusion
if hemoglobin b9 g/dL and a suspected bleeding event. Adjusted associations for outcomes were displayed as odds ratios (ORs) and 95% CIs. A 2-sided P b .05 was established as the level of statistical significance for all tests. All analyses were performed using SAS software (version 9.2; SAS Institute, Cary, NC). This research was supported by the American College of Cardiology Foundation’s NCDR. The views expressed in this article represent those of the authors and do not necessarily represent the official views of the NCDR or its associated professional societies identified at www.ncdr.com. The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper, and its final contents.
Results Frequency of prior MI The prevalence of prior MI was greater for NSTEMI patients versus STEMI patients (29% vs 19% over the entire time period). Over the 5-year analysis time period, the temporal frequency of prior MI was similar for STEMI patients (trend P value = .32) but increased for NSTEMI patients (27.3% in quarter 1, 2007, to 30.2% in Quarter 1, 2012; trend P value b .0001) (Figure). Baseline characteristics, treatments, and procedures Both for STEMI and NSTEMI patients, prior-MI patients were more likely to be older, be male, and have comorbidities such as hypertension, hyperlipidemia, and diabetes; were more likely to have a prior history of revascularization and congestive heart failure; and were more likely to be treated with secondary prevention medications at home compared with those without prior MI (Table I). Medication use within 24 hours of arrival and inhospital procedure use were similar by prior-MI status in STEMI patients. For NSTEMI, prior-MI patients were shown to have higher use of thienopyridine and statins, lower use of glycoprotein IIb-IIIa inhibitors, and lower use of angiography and revascularization procedures (Table II). The use of discharge medications and interventions was mostly similar by prior-MI status for both STEMI and NSTEMI patients. (Table III). In-hospital outcomes The frequency of in-hospital mortality was higher in prior-MI patients, both for STEMI and NSTEMI patients. However, after adjustment, the excess mortality risk was not apparent for prior-MI status for STEMI (adjusted OR = 1.06, 95% CI 0.97-1.15) but was marginally higher for prior MI in NSTEMI (adjusted OR = 1.10, 95% CI 1.041.15) (Table IV). The adjusted risk of in-hospital major bleeding was also similar among patients with (versus without) prior MI for STEMI but was marginally lower for NSTEMI (Table IV).
American Heart Journal Volume 0, Number 0
Figure
Patients with prior MI.Proportion of patients with prior MI among STEMI versus NSTEMI by quarter (2007 Q1-2012 Q1).
Discussion Among patients with an index acute MI, a history of prior MI was more prevalent in those whose index acute MI was NSTEMI compared with STEMI. Patients with a history of prior MI were older and had more comorbidities compared with patients without prior MI and were more commonly treated with secondary prevention medications at home prior to admission. Subsequently, those with prior MI have a higher risk of unadjusted inhospital mortality; however, after adjustment, there was a modestly excessive risk of in-hospital mortality only for NSTEMI patients, with no increased risk with prior MI for STEMI patients. Data from previous ACTION Registry-GWTG analyses have demonstrated that approximately 20% of MI patients have had a prior MI, but these results were not differentiated by MI classification (STEMI vs NSTEMI). 11 Another study showed that NSTEMI patients have a higher rate of prior MI compared with STEMI patients (42.6% vs 23.1%, respectively). 3 Similarly, we also found that prior MI is more likely in NSTEMI versus STEMI patients (29% vs 19%, respectively); so our findings are consistent with previous studies. 12 Nevertheless, we found that prior MI was associated with older age, a greater prevalence of cardiovascular risk factors, more frequent prior or current heart failure, and lower creatinine clearance (CrCl) values. Yet although patients with a prior MI were more likely to be on secondary prevention medications at home, only about 70% were on aspirin and b65% were on a statin. Therefore, patients with a prior MI have a greater burden of comorbidities when they present with an index acute MI; furthermore, they appear to be suboptimally treated with secondary prevention medications in the chronic setting. 13,14 The findings from our study revealed that differences in mortality may be partially explained by differences in baseline characteristics and home medications between those with versus without a prior MI, as the higher
Shen et al 3
unadjusted risks of in-hospital mortality associated with prior MI disappeared in STEMI and were attenuated in NSTEMI with risk adjustment. Prior MI has been considered a high-risk feature in several recent acute coronary syndrome clinical trials that enrolled unstable angina/NSTEMI patients and used prior MI as one of the enrichment criteria to enhance long-term cardiovascular event rates 15–17 because prior MI has proven to be independently associated with worse outcomes through 6 months post-NSTEMI in both the Thrombolysis in Myocardial Infarction and Global Registry of Acute Cardiac Events risk models. 4,5,18,19 Although our study was limited to the in-hospital period, we have shown that there is only a modest increase in the risk of in-hospital mortality with this key baseline characteristic—a finding that highlights the differential characteristics associated with the risk of early versus later mortality following acute MI presentation. Historically, adherence to guideline-recommended therapies after MI was significantly lower in NSTEMI patients compared with STEMI patients. 20,21 Despite recent improvements in the guideline adherence gap for NSTEMI versus STEMI patients, 22 we found that, in a contemporary population of both NSTEMI and STEMI patients, guideline adherence was relatively similar except for a lower use of angiography, revascularization, discharge thienopyridines, and cardiac rehabilitation referral for NSTEMI versus STEMI patients. Our study findings provide further insight into these treatment disparities by delineating that the presence of a prior MI did not appear to impact the quality of care for STEMI patients but was associated with a lower use of angiography and revascularization for NSTEMI patients (compared with NSTEMI patients without a prior MI). These findings are likely explained by the much higher incidence of prior percutaneous coronary intervention (PCI) or CABG and relatively lower CrCl values seen with prior MI in NSTEMI patients. These features likely impact the risk-benefit calculations that guide decision making for an early invasive management strategy that concurrently influences the use of antiplatelet therapies and referral for cardiac rehabilitation in NSTEMI patients.
Study limitations Our study has several limitations. First, despite the fact that we have 307,646 consecutive MI patients from 446 hospitals, the participating centers were voluntarily participating in the ACTION Registry-GWTG; therefore, these finding may not apply to patients and hospitals that differ significantly from those participating in the ACTION Registry-GWTG. Second, the detection of prior MI was abstracted from medical records, which are dependent upon the documentation accuracy of each hospital. Therefore, we could not ascertain when the prior MI happened or whether the index MI events
American Heart Journal Month Year
4 Shen et al
Table I. Baseline clinical and home medications *
Variables Demographics Age (y) BMI (kg/m2) Female sex (%) Race (%) White African American Asian Hispanic Other Medical history (%) Hypertension Diabetes mellitus Peripheral arterial disease Current/recent smoking Hyperlipidemia Prior PCI Prior CABG Prior heart failure Prior stroke Currently on dialysis Presentation features Initial heart rate (beat/min) Initial systolic BP (mm Hg) Cardiogenic shock (%) Signs of heart failure (%) ST depression or transient ST elevation (%) Initial laboratory results Baseline troponin ratio, ×ULN Hemoglobin (g/dL) Serum creatinine (mg/dL) CrCl (mL/min) ACTION mortality risk score Home medications (%) Aspirin Thienopyridine † β-Blocker Statin
STEMI
NSTEMI
n = 124,535
n = 194,617
No prior MI
Prior MI
No prior MI
Prior MI
n = 100,811
n = 23,724
n = 137,867
n = 56,750
81%
19%
71%
29%
60 (51-70) 28 (25-32) 31
62 (53-72) 28 (25-32) 26
66 (55-77) 28 (25-33) 40
69 (59-79) 28 (25-33) 34
84 8 2 4 1
84 9 1 4 1
82 11 1 4 1
83 11 1 4 1
59 22 4 43 46 8 4 3 4 1
82 32 11 46 78 70 20 14 9 2
73 32 9 31 57 13 11 11 8 3
89 46 20 30 81 59 38 31 14 5
78 (66-93) 140 (120-161) 7 9
78 (66-93) 137 (116-158) 7 13
83 (70-98) 147 (127-168) 2 17 25
84 (71-100) 144 (123-165) 2 26 25
1.2 (0.3-14.4) 14 (13-16) 1 (0.9-1.2) 88 (63-114) 32 (27-38)
0.9 (0.2-7.0) 14 (13-15) 1 (0.9-1.3) 83 (58-112) 33 (28-40)
2.5 (0.6-13) 14 (12-15) 1 (0.9-1.3) 78 (51-109) 28 (22-35)
1.8 (0.4-9.2) 13 (12-15) 1.1 (0.9-1.5) 68 (44-99) 31 (24-38)
26 5 20 24
64 29 58 57
40 11 34 36
71 37 69 65
BMI, Body mass index; BP, blood pressure; ULN, times the upper limit of normal. * All values reported as percentage or median (interquartile range) unless otherwise specified. † Clopidogrel or prasugrel.
reported in the ACTION Registry-GWTG were spontaneous MI or stent thrombosis MI events (for patients who had undergone prior PCI). Furthermore, we could not determine the location of the infarct-related artery or the position of the culprit lesion within that artery; so we cannot comment on the potential differences in the pathophysiology of the index MI events in prior-MI versus no-prior-MI patients. Similarly, the records did not
capture other medical features that might influence treatment decisions and outcomes such as infectious or malignant disease. Finally, only in-hospital outcomes are collected in the ACTION Registry-GWTG; therefore, the long-term prognostic implications of prior MI for STEMI versus NSTEMI patients could not be assessed, and the impact of prior MI is more likely to be manifest with longer follow-up.
American Heart Journal Volume 0, Number 0
Shen et al 5
Table II. In-hospital treatments
Variables
Table IV. In-hospital mortality and major bleeding by prior-MI status Unadjusted outcomes
STEMI
NSTEMI
n = 124,535
n = 194,617
No prior MI
Prior MI
No prior MI
Prior MI
n= 100,811
n= 23,724
n= 137,867
n= 56,750
99 89 64 73 93 96
99 88 62 73 93 95
97 56 33 60 88 92
97 61 27 66 89 91
96 86 7
95 86 6
83 50 13
75 44 8
Acute medication (%) Aspirin Thienopyridine* GP IIb/IIIa inhibitor Statin β-Blocker Anticoagulants Invasive procedure (%) Angiography PCI CABG
GP IIb/IIIa, Glycoprotein IIb/IIIa. * Clopidogrel or prasugrel.
Table III. Discharge medications and interventions STEMI
NSTEMI
n = 124,535
n = 194,617
No prior No prior MI Prior MI MI Prior MI
Variables
n= 100,811
n= 23,724
99 91 95 97 84
99 91 95 98 82
97 72 89 95 75
98 77 91 97 72
98
98
97
97
Discharge medication (%) Aspirin Thienopyridine* Statin β-Blocker Cardiac rehabilitation referral (%) Smoking cessation counseling (%)
n= n= 137,867 56,750
* Either clopidogrel or prasugrel.
Conclusions The prevalence of prior MI is higher among NSTEMI versus STEMI patients and is associated with a greater burden of co-morbidities on presentation. Nonetheless, we found only a modest independent incremental risk of in-hospital mortality with prior MI in NSTEMI patients but not in STEMI patients.
STEMI Mortality Major bleeding NSTEMI Mortality Major bleeding
Unadjusted OR*
Adjusted OR*
No prior MI
Prior MI
(95% CI)
(95% CI)
5.2 10.3
5.9 10.5
1.15 (1.08-1.22) 1.01 (0.97-1.06)
1.06 (0.97-1.15) 0.97 (0.92-1.03)
3.4 8.2
4.3 9.5
1.28 (1.22-1.34) 1.18 (1.14-1.23)
1.10 (1.04-1.15) 0.96 (0.92-0.99)
* OR of the outcomes between prior MI and no prior MI (reference group).
References 1. Anavekar NS, McMurray JJV, Velazquez EJ, et al. Relation between renal dysfunction and cardiovascular outcomes after myocardial infarction. N Engl J Med 2004;351:1285-95. 2. Roe MT, Chen AY, Cannon CP, et al. Temporal changes in the use of drug-eluting stents for patients with non–ST-segment elevation myocardial infarction undergoing percutaneous coronary intervention from 2006 to 2008: results from the Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA guidelines (CRUSADE) and Acute Coronary Treatment and Intervention Outcomes Network—Get With The Guidelines (ACTionGWTG) registries. Circ Cardiovasc Qual Outcomes 2009;2:414-20. 3. Motivala AA, Tamhane U, Ramanath VS, et al. A prior myocardial infarction: how does it affect management and outcomes in recurrent acute coronary syndromes? Clin Cardiol 2008;31:590-6. 4. Eagle KA, Lim MJ, Dabbous OH, et al. A validated prediction model for all forms of acute coronary syndrome: estimating the risk of 6-month postdischarge death in an international registry. JAMA 2004;291:2727-33. 5. Morrow DA, Antman EM, Charlesworth A, et al. TIMI risk score for ST-elevation myocardial infarction: a convenient, bedside, clinical score for risk assessment at presentation: an intravenous nPA for treatment of infarcting myocardium early II trial substudy. Circulation 2000;102:2031-7. 6. Peterson ED, Roe MT, Rumsfeld JS, et al. A call to ACTION (Acute Coronary Treatment and Intervention Outcomes Network): a national effort to promote timely clinical feedback and support continuous quality improvement for acute myocardial infarction. Circ Cardiovasc Qual Outcomes 2009;2:491-9. 7. NCDR® ACTION Registry®-GWTG™ v2.2 coder’s data dictionary. National Cardiovascular Data Registry web site, https://www.ncdr. com/WebNCDR/docs/public-data-collection-documents/ action_v2_codersdictionary_2-2.pdf?sfvrsn=2. [Updated February 3, 2011. Accessed March 6, 2014]. 8. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121-30. 9. Chin CT, Chen AY, Wang TY, et al. Risk adjustment for in-hospital mortality of contemporary patients with acute myocardial infarction: the Acute Coronary Treatment and Intervention Outcomes Network (ACTION) Registry—Get With The Guidelines (GWTG) acute myocardial infarction mortality model and risk score. Am Heart J 2011;161:113-122.e2.
6 Shen et al
10. Mathews R, Peterson ED, Chen AY, et al. In-hospital major bleeding during ST-elevation and non–ST-elevation myocardial infarction care: derivation and validation of a model from the ACTION Registry®-GWTG™. Am J Cardiol 2011;107:1136-43. 11. Lopes RD, Peterson ED, Chen AY, et al. Antithrombotic strategy in non–ST-segment elevation myocardial infarction patients undergoing percutaneous coronary intervention: insights from the ACTION (Acute Coronary Treatment and Intervention Outcomes Network) Registry. JACC Cardiovasc Interv 2010;3:669-77. 12. Haffner SM, Lehto S, Rönnemaa T, et al. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction. N Engl J Med 1998;339:229-34. 13. Alexander KP, Chen AY, Newby LK, et al. Sex differences in major bleeding with glycoprotein IIb/IIIa inhibitors: results from the CRUSADE (Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA guidelines) initiative. Circulation 2006;114:1380-7. 14. Bhatt DL, Roe MT, Peterson ED, et al. Utilization of early invasive management strategies for high-risk patients with non–ST-segment elevation acute coronary syndromes: results from the CRUSADE Quality Improvement Initiative. JAMA 2004;292:2096-104. 15. Alexander JH, Lopes RD, James S, et al. Apixaban with antiplatelet therapy after acute coronary syndrome. N Engl J Med 2011;365: 699-708. 16. Roe MT, Armstrong PW, Fox KA, et al. Prasugrel versus clopidogrel for acute coronary syndromes without revascularization. N Engl J Med 2012;367:1297-309. 17. Tricoci P, Huang Z, Held C, et al. Thrombin-receptor antagonist vorapaxar in acute coronary syndromes. N Engl J Med 2012;366:20-33.
American Heart Journal Month Year
18. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients with Unstable Angina/Non ST-elevation Myocardial Infarction): developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons: endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. Circulation 2007;116:e148-304. 19. Hamm CW, Bassand JP, Agewall S, et al. ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: the Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J 2011;32:2999-3054. 20. Montalescot G, Dallongeville J, Van Belle E, et al. STEMI and NSTEMI: are they so different? 1 year outcomes in acute myocardial infarction as defined by the ESC/ACC definition (the OPERA registry). Eur Heart J 2007;28:1409-17. 21. Roe MT, Parsons LS, Pollack Jr CV, et al. Quality of care by classification of myocardial infarction: treatment patterns for ST-segment elevation vs non–ST-segment elevation myocardial infarction. Arch Int Med 2005;165:1630-6. 22. Somma KA, Bhatt DL, Fonarow GC, et al. Guideline adherence after ST-segment elevation versus non-ST segment elevation myocardial infarction. Circ Cardiovasc Qual Outcomes 2012;5: 654-61.
