- Email: [email protected]
Missed opportunities for reperfusion therapy for ST-segment elevation myocardial infarction: Results of the Emergency Department Quality in Myocardial Infarction (EDQMI) study
Missed opportunities for reperfusion therapy for ST-segment elevation myocardial infarction: Results of the Emergency Department Quality in Myocardial Infarction (EDQMI) study Albert J. Tricomi, MD,a David J. Magid, MD, MPH,b,c John S. Rumsfeld, MD, PhD,b,c,d David R. Vinson, MD,e Ella E. Lyons, MS,b Laurie Crounse, MPH,b P. Michael Ho, MD, PhD,b,c,d Pamela N. Peterson, MD, MSPH,b,c,f and Frederick A. Masoudi, MD, MSPHb,c,f, for the Emergency Department Quality in Myocardial Infarction (EDQMI) Study Investigators Rochester, NY; Denver, CO; and Roseville, CA
Background Although it is known that reperfusion therapy for ST-elevation myocardial infarction (STEMI) is underused, the reasons for the failure to provide this potentially life-saving treatment are not well described. Methods In a cohort of 2215 consecutive patients presenting with acute myocardial infarction to 5 emergency departments in Colorado and California between 2000 and 2002, patients with ischemic symptoms and ST-segment elevation on electrocardiogram without documented guideline-based contraindications to therapy were identified as eligible reperfusion candidates. Multivariable logistic models were constructed to identify factors associated with the failure to receive reperfusion. The emergency department records of patients not receiving reperfusion were reviewed to categorize the reasons therapy was not provided. Results Of 460 eligible patients, 102 (22%) did not receive reperfusion therapy. Patient factors associated with failure to receive reperfusion therapy included older age, peripheral vascular disease, and absence of chest pain; patients seen by both resident and attending physicians were more likely to receive treatment than those seen by an attending alone. In cases where reperfusion was not provided, ST-segment elevation was not identified in 34% (n = 35), left bundle-branch block was not considered as an indication in 13% (n = 13), there was documentation of a reason for withholding therapy not supported by guidelines in 34% (n = 35), and there was no documentation of reasons for withholding reperfusion in 19% (n = 19). Conclusions Initiatives to improve electrocardiogram interpretation and evidence-based patient selection may reduce gaps in the delivery of reperfusion therapy to eligible candidates and thereby potentially improve STEMI outcomes. (Am Heart J 2008;155:471-7.) Several studies have shown that a significant proportion of eligible patients with ST-elevation myocardial infarction (STEMI) do not receive potentially life-saving reperfusion therapy. For instance, 24% of eligible patients with STEMI in the NRMI and 30% in the GRACE did not receive reperfusion therapy.1,2 Prior studies have identi-
From the aUniversity of Rochester School of Medicine, Rochester, New York, NY, bClinical Research Unit, Colorado Permanente Medical Group, Denver, CO, cUniversity of Colorado at Denver and Health Sciences Center, Denver, CO, dDenver VA Medical Center, Denver, CO, eKaiser Permanente Medical Center, Roseville, CA, and fDenver Health Medical Center, Denver, CO. Submitted May 18, 2007; accepted October 23, 2007. Reprint requests: Frederick A Masoudi, MD, MSPH Cardiology, MC0960, Denver Health Medical Center, 777 Bannock St. Denver, CO. E-mail: [email protected] 0002-8703/$ - see front matter © 2008, Published by Mosby, Inc. doi:10.1016/j.ahj.2007.10.023
fied specific patient factors associated with a lower likelihood of receiving reperfusion, including advanced age, female sex, nonwhite race, and noncardiac comorbidities.1,3,4 Although recognizing the patient factors associated with failure to receive reperfusion therapy is important, doing so does not directly support changes in the delivery of care to increase appropriate treatment. To date, little investigation has been dedicated to identifying the underlying reasons why reperfusion therapy is not administered to some eligible patients. An understanding of the specific reasons for the failure to provide appropriate care is a more informative to the development programs to optimize care delivery for acute myocardial infarction (AMI). The goal of this study was to better understand the reasons why eligible patients with STEMI did not receive fibrinolysis or primary percutaneous coronary
American Heart Journal March 2008
472 Tricomi et al
Figure 1
The study population. PCI, Percutaneous coronary intervention; EDQMI, Emergency Department Quality in Myocardial Infarction; LBBB, left bundle branch block; STEMI, ST-segment elevetion mycardial infarction.
intervention (PCI) in a multicenter cohort of patients presenting to the emergency department (ED). By reviewing the ED medical record and electrocardiograms (ECGs), our objective was to identify and categorize the causes for the failure to provide this important therapy. The results of this study are intended to inform initiatives to provide the highest quality of AMI care.
Methods Study design and setting The EDQMI project was a 2-year retrospective cohort study focusing on the quality of ED care for AMI.5 Consecutive patients presenting to 5 community EDs in Colorado and California from July 1, 2000, through June 30, 2002 were screened. Subjects were included if they (1) presented directly to the ED, (2) had elevated cardiac markers diagnostic for AMI in the ED or within 24 hours of ED arrival, and (3) had a discharge diagnosis of AMI according to the International Classification of Diseases, Ninth Revision, Clinical Modification (code 410.X). Elevated cardiac markers were defined as (1) elevations of total creatine kinase or creatine kinase–MB at least 2 times the upper limit of the reference range or (2) elevations in troponin that exceeded the reference and indeterminate range. Those patients who were directly admitted to the inpatient setting, transferred after initial ED care at another institution, or whose diagnostic cardiac marker elevations occurred N24 hours after ED arrival were excluded. Of the 5 hospitals participating in the study, 3 had cardiac catheterizations laboratories and 2 provided primary PCI. Institutional review board approval was obtained at all 5 study locations.
Study sample Patients were considered eligible reperfusion candidates if they presented with possible ischemic symptoms beginning within 12 hours of ED presentation; had an ECG obtained in the ED demonstrating ST-segment elevation of at least 1 mm in ≥2 contiguous leads or left bundle-branch block (LBBB) that was new or not known to be old; and had no contraindications to reperfusion therapy. Contraindications to reperfusion therapy included those previously described in the American Heart Association/American College of Cardiology guidelines.6 To identify this eligible cohort, we excluded patients who did not have ST-segment elevation or LBBB on an initial or subsequent ED ECG (n = 1619); those who presented N12 hours after symptoms onset (n = 51); and those who had documented contraindications to reperfusion therapy or other documented reasons (eg, patient refusal, n = 5) why reperfusion therapy was considered but not given (n = 85). The remaining patients (n = 460) were considered eligible candidates for reperfusion (Figure 1).
Data sources Data on the ED care processes were obtained through review of ED records. Four physician investigators conducted chart reviews using an electronic data abstraction tool. Data were collected on patient demographics, clinical history including contraindications to reperfusion, ED provider's ECG interpretation, and ED therapies. All abstractors received training on the content and coding of each data element, data handling and data transmission procedures, and protocols to handle possible questions or problems during the study. During the training period, abstractors also coded 10 practice charts. Although
American Heart Journal Volume 155, Number 3
Tricomi et al 473
Table I. Baseline characteristics of reperfusion-eligible patients Received reperfusion (n = 358)
Did not receive reperfusion (n = 102)
n
%
N
%
P⁎
Age † b65 65-80 N80
63.7 181 142 35
12.6 50.6 39.7 9.8
69.5 37 32 33
15.8 36.3 31.4 32.4
b.001 b.001
Female Male
109 249
30.4 69.6
38 64
37.3 62.7
.2
Caucasian Non Caucasian Other/unknown
236 68 54
65.9 19.0 15.1
66 19 17
64.7 18.6 16.7
b.001
Diabetes Hypertension Hypercholesterolemia Current smoker
70 160 181 111
19.6 44.7 50.6 31.0
24 51 40 25
23.5 50.0 39.2 24.5
.4 .3 .04 .2
44 31 20 12 21
12.3 8.7 5.6 3.4 5.9
15 6 8 11 9
14.7 5.9 7.8 10.8 8.8
.5 .4 .4 .002 .3
7 29 18 12
2.0 8.1 5.0 3.4
8 9 10 12
7.8 8.8 9.8 11.8
.003 .8 .08 b.001
17 341
4.7 95.3
23 79
22.5 77.5
b.001
297 61
83.0 17.0
96 6
94.1 5.9
b.001
Demographics Age (y)
Sex
Race
Cardiac risk factors
Cardiac history Previous MI PCI CABG CHF Angina Medical history CRI COPD Stroke PVD Chest pain at presentation No Yes Seen by resident in addition to attending physician in ED No Yes
CABG, Coronary artery bypass grafting; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CRI, chronic renal insufficiency; PVD, peripheral vascular disease. ⁎P value is for the comparison between patients receiving reperfusion and those not receiving reperfusion. †Presented as mean and SD.
the abstractors were aware that the objective of the EDQMI project was to assess the quality of ED AMI care, they were not aware at the time of the chart review of the goals of this specific analysis. Multiple processes were instituted to enhance the accuracy and reliability of the data abstraction process. A data manager monitored day-to-day data collection activities, answered coding or eligibility questions, and maintained a log of all coding decisions made subsequent to the start of the project. After the abstraction was complete and the data were entered, the data manager verified case eligibility, confirmed the validity of the responses, and attempted to locate any missing data. To assess interrater reliability, all abstractors coded a random sample of 30 charts, and the results of these abstractions were compared to a “gold-standard” abstraction. For the key outcome variable of the provision of reperfusion therapy, there was perfect interrater reliability for all abstractors (κ = 1.0).
In addition to the ED chart reviews, all ECGs obtained in the ED were reviewed by 1 of 3 board-certified cardiologists. These interpretations included an assessment for ST-segment elevation of at least 0.1 mV in at least 2 contiguous leads and LBBB. In cases where LBBB was present, the presenting ECG was compared with a prior tracing whenever possible to determine if the finding was new. When the prior tracing was not available, information from the chart abstraction was used to determine whether the LBBB was new or presumably new.
Categorization of failures to provide reperfusion Patients were considered to receive reperfusion therapy if they received a fibrinolytic agent or underwent coronary angiography with the intent of providing PCI regardless of the success of the reperfusion therapy. The ED charts and ECGs of patients who were initially classified as eligible but did not receive reperfusion, including those whose initial chart review
474 Tricomi et al
identified a contraindication to therapy, were reviewed again by a separate panel of 4 investigators to confirm the patient's eligibility for reperfusion and to assess for specific treatment contraindications. The confirmed cases of eligible reperfusion candidates not receiving reperfusion (n = 102) were assessed to determine the underlying reasons for the failure to provide treatment. All failures to provide reperfusion could be assigned to 1 of 4 categories: (1) failure to identify ST-elevation on the ED ECG, (2) failure to identify new or presumed new LBBB as an inclusion criterion for therapy, (3) documentation of a reason for withholding reperfusion that was not concordant with guideline recommendations, or (4) failure to document any rationale for withholding reperfusion.
Data analysis Among eligible candidates for reperfusion therapy, we compared characteristics of patients who received and did not receive reperfusion therapy with t tests for continuous variables and χ2 tests for categorical variables. We used multivariable hierarchical regression models to identify patient factors independently associated with the failure to provide reperfusion, adjusting for patient demographic and clinical characteristics and accounting for the clustering of patients within hospitals. Relatively small numbers of patients per physician precluded an analysis that accounted for clustering of patients within physician. The outcome variable for these models was reperfusion therapy (receipt of PCI or fibrinolysis). Regression models were developed using a stepwise approach in which covariates were assessed based on their contribution to the overall model. In deciding whether to include a covariate in the model, we considered whether the estimate of the coefficients of some of the primary predictor variables (eg, age, chest pain) meaningfully changed when the covariate was removed from the model as well as whether including the covariate improved precision. At the conclusion of the model building process, we assessed the global fit of the model using the Hosmer-Lemeshow test and the local fit using the Akaike's information criteria and the deviance. Including calendar time as a candidate variable allowed for the assessment of potential secular trends in the administration of reperfusion therapy and temporal variation in case mix. Odds ratios and 95% CIs were calculated for all covariates in the models. Statistical analyses were performed using SAS version 8 (SAS Inc, Cary, NC).
Results Characteristics of eligible reperfusion patients Of the 2215 patients enrolled in EDQMI, 596 patients presented with ST-segment elevation or new LBBB. After excluding the patients who presented with N12 hours of ischemic symptoms or who had guidelineconcordant contraindications to reperfusion (n = 136), 460 eligible reperfusion candidates were identified for inclusion in this study (Figure 1). The mean age of eligible candidates was 64.9 years, 32% were female, and 65% were Caucasian. There were high rates of traditional cardiac risk factors, and 13% of the study population had a history of previous MI, 8% had previously undergone PCI, and 6% had a history of prior coronary artery bypass grafting (Table I).
American Heart Journal March 2008
Rates of reperfusion therapy and factors associated with reperfusion Of the 460 eligible reperfusion candidates, 358 (78%) received reperfusion therapy. Of the eligible candidates, 114 (32%) underwent urgent coronary angiography and subsequent primary PCI if the culprit lesion was amenable to treatment and 244 (68%) received fibrinolysis. The remaining 102 (22%) eligible patients did not receive reperfusion therapy (Figure 1). Patients who did not receive reperfusion were on average older and were more likely to have a history of important comorbidities, including heart failure, renal insufficiency, and peripheral vascular disease (Table I). Patients who did not receive reperfusion were also less likely to present with typical chest discomfort and were less likely to be seen by a resident in conjunction with an attending physician. In multivariable analysis of eligible candidates, younger patients were more likely to receive reperfusion than older patients (OR 4.10, 95% CI 1.83-9.23 for age b65 years compared with age N80 years; OR 3.43, 95% CI 1.63-7.23 for age 65-80 years compared with age N80 years). Patients with chest pain as a presenting symptom were more likely to receive reperfusion than those who presented with atypical symptoms (OR 4.99, 95% CI 2.22-11.21). Patients with peripheral vascular disease were less likely to receive reperfusion (OR 0.23, 95% CI 0.08-0.71). Patients seen by a resident in conjunction with an attending physician were more likely to receive reperfusion compared with those seen by an attending physician alone (OR 3.11, 95% CI 1.17-8.29). Failures to treat eligible candidates Among the 102 eligible reperfusion candidates who did not receive therapy, ST-segment elevation was not correctly identified in 35 (34%). Of these 35 cases, 30 patients had an initial ED ECG that showed ST elevation and the other 5 patients presented initially without ST-segment elevation but developed this finding on a subsequent ECG obtained in the ED. In 13 cases (13%), the ED ECG showed an LBBB, but there was no mention in the chart as to whether the LBBB was new or old, nor was there documentation to suggest that reperfusion therapy was considered (Figure 1). In another 19 cases (19%), the ECG was correctly interpreted; however, reperfusion therapy was withheld for a reason that was not concordant with the guideline recommendations.6 A common reason for withholding reperfusion was the presence of intermittent symptoms over the several days before presentation (n = 6). Although these patients often had been experiencing anginal episodes for several days, the current episode of pain began b12 hours before their presentation to the ED, supporting their eligibility for reperfusion therapy. A second common reason for not providing reperfusion occurred in patients whose symptoms or ST elevation on ECG improved but did not resolve with
American Heart Journal Volume 155, Number 3
Table II. Selected reasons cited for not providing reperfusion therapy • Currently asymptomatic • Patient had 4 d of intermittent angina and with worsening persistent pain at presentation • Patient's old age without further qualification • Warfarin recently stopped, but INR was not elevated • Concern for abdominal aortic aneurysm, but CT negative for AAA • Subsequent ECG with less ST elevation • Pain resolved with other medications • Patient had been having pain for 10 h INR, International normalized ratio; CT, computed tomography; AAA, abdominal aortic aneurysm.
standard medical therapy (n = 5). Examples of the non– guideline-based reasons that were documented in patient charts for not providing reperfusion therapy to eligible patients are provided in Table II. In the remaining 35 cases (34%), ST elevation or new LBBB was identified by ED physician, but there was no documentation of any contraindication to reperfusion the medical record.
Discussion The primary objective of this study of patients with STEMI presenting to the ED was to assess the underlying reasons why eligible reperfusion candidates did not receive reperfusion. We found that more than 1 in 5 reperfusion-eligible patients with MI did not receive reperfusion therapy. Individual chart review revealed that the failure to identify significant ST-segment elevation, the lack of recognition of new or presumably new LBBB as an indication for reperfusion, and failure to identify guideline-based indications for reperfusion all played important roles in the failure to deliver reperfusion therapy. Despite substantial clinical trial evidence and support in practice guidelines for the use of reperfusion in selected patients, many who could benefit from reperfusion are deprived of this potentially life-saving therapy.6-8 Our study of ED patients shows rates of use of reperfusion therapy among eligible patients similar to prior studies, and identifies similar specific patient factors associated with the underuse of reperfusion—including age and comorbidities such as heart failure, renal insufficiency, and peripheral vascular disease.1,4,9 Unlike these prior investigations, the present study included a review of each patient's ED ECGs and a detailed case adjudication to confirm eligibility for treatment. Thus, this study is able to go beyond the identification of important gaps in the delivery of reperfusion therapy to patients in community settings, using detailed chart review to assess patient eligibility and identify specific reasons why reperfusion therapy was withheld. Although identifying patients at risk for undertreatment is important, the process of describing the reasons for the failure to provide therapy is markedly more informative to the development of quality improvement
Tricomi et al 475
initiatives. The current study identifies discrete and potentially remediable factors that pose important barriers to the provision of indicated treatment, including the failure to recognize ST-segment elevation, lack of knowledge of the importance of LBBB, and inadequate knowledge of contraindications to reperfusion. Among patients in this study who did not receive appropriate reperfusion therapy, one third had ECGs with ST-segment elevation that was not identified at either presentation or on subsequent tracings. A prior study from EDQMI found that the failure to identify high-risk ECG findings was associated with lower rates of treatment with evidence-based therapy, including reperfusion, among eligible candidates.5 This study confirms the importance of the ECG interpretation and suggests that a third of all failures to provide reperfusion therapy are attributable to the failure to identify ST-segment elevation. Although a prior study found that the failure to identify ST-segment elevation among all patients presenting with chest pain had modest consequences,10 this study of patients with documented AMI indicates that missing ST-segment elevation plays a central role in the failure to provide potentially lifesaving therapy. Although hospital mortality is higher for MI patients with LBBB than those without,11 and providing reperfusion therapy to patients with ischemic symptoms and LBBB that is not known to be old is recommended by guidelines,6 the present study corroborates previous data showing that such patients are undertreated with reperfusion.1 Despite prior research reporting that physicians have a lower threshold for hospitalizing patients with LBBB when presenting with chest pain,12 our findings indicate that this does not necessarily translate into higher rates of reperfusion therapy for eligible candidates. Thus, this study reemphasizes the need for providers who treat patients with MI in the immediate setting to understand the importance of LBBB as criterion for eligibility for immediate reperfusion therapy. Although misinterpretation of ST elevation and reluctance to aggressively intervene in patients with LBBB appear to be important causes of failure to provide reperfusion, we also observed that reperfusion therapy is either inappropriately dismissed or not considered at all for some eligible patients. In 19% of the missed opportunities for reperfusion therapy, the provider documented a perceived contraindication to reperfusion not supported by guidelines as rationale for withholding therapy. This finding suggests that improving the understanding of the evidence-based selection of appropriate candidates for treatment may improve the delivery of reperfusion therapy. In the remaining 34% of cases where eligible patients did not receive reperfusion, there was no reason for withholding reperfusion documented in the medical
476 Tricomi et al
record. Because of the difficulty in conclusively proving the absence of contraindications through chart abstraction, we cannot entirely exclude the possibility that true contraindications to reperfusion therapy were present but not documented in some cases. However, the consideration of reperfusion therapy was not documented in any of these cases, supporting the conclusion that these do not merely reflect the failure to document contraindications among ineligible patients. This study identifies possible mechanisms to enhance the delivery of reperfusion to eligible candidates with STEMI. Increased attention to the interpretation of the ECG as well as the ECG criteria for reperfusion during postgraduate training and continuing medical education courses may improve the identification of eligible patients. In addition, based on the findings of this and previous studies, specific emphasis on the importance of LBBB and the contraindications for reperfusion based on existing practice guidelines are warranted. Based on the literature of systems change, however, it is unlikely that education alone will be adequate to enhance care. The findings of this study also suggest that fundamental changes in the structure of care delivery may enhance the delivery of reperfusion therapy. Such systems might include secondary ECG interpretations by another provider. The finding that care including both a resident and an attending physician was associated with higher rates of reperfusion supports the hypothesis that secondary ECG interpretations may improve care. Although a secondary ECG interpretation by another provider may improve the identification of eligible patients, this strategy—requiring the evaluation of a large number of ECGs to identify the smaller group of reperfusion eligible patients—would necessitate significant resources and could potentially contribute to important delays in reperfusion therapy. Automated ECG interpretation could serve this role; however, although automated ECG interpretation has been shown in randomized trials to improve the accuracy of patient triage for reperfusion,13 such systems have not been widely implemented. Incorporating checklists of the appropriate selection criteria for reperfusion in critical pathways for patients presenting with chest pain may be useful. Finally, simulation-based training represents an emerging approach that may improve the quality of care for STEMI patients. This strategy has been most extensively used in anesthesia training, but more recent data suggest that it may also be a useful tool in emergency medicine and critical care education.14 In interpreting this study, certain factors warrant consideration. Because the study implemented retrospective medical record review, we cannot rule out the possibility of biases due to the presence of undocumented guideline-based contraindications to therapy or missing data. Undocumented guideline-based contraindications would result in an overestimation of the
American Heart Journal March 2008
number of eligible treatment candidates as well as the proportion of untreated patients. Nevertheless, among those patients in this study for whom inadequate documentation was identified, more than one third had documentation of a contraindication that was not concordant with guideline recommendations. Thus, many of these cases likely represent meaningful opportunities to improve care. Missing ECGs that were available to the treating provider but not documented in the record and not available for adjudication may also have increased the number of patients categorized as eligible. Although failures of documentation are arguably important, they may not have an impact on patient care comparable with the failure to provide appropriate care. These data were collected retrospectively and did not include detailed interviews of providers. Thus, it was not possible to assess all important aspects of care (eg, involvement of a cardiologist) or to better understand the thinking of clinicians who cared for patients who did not receive therapy. Studies involving detailed interviews would be useful complements to this study in better understanding the failure to provide reperfusion to eligible patients. Second, the panel of investigators performing medical record and chart review were aware that each patient in the study sample had been diagnosed with an MI. However, during ECG review, they were blinded to the results of the chart review and the original ECG interpretation. Third, the impact of the failure to provide reperfusion therapy on patient outcomes was not assessed. However, numerous studies have demonstrated higher mortality among eligible candidates who are not treated both in the context of trials and in community-based samples.1,15 Finally, as our study focused on the MI care provided in the ED, we were unable to assess the care of the patient in the hospital after leaving the ED. However, given the critical importance of time in the provision of reperfusion in patients with STEMI, the lack of any evidence of providing reperfusion to eligible patients in the ED record reflects meaningful gaps in care.
Conclusions We found that more than 20% of ED patients with AMI who are eligible candidates for reperfusion therapy do not receive this important evidenced-based intervention. In addition to identifying those patient groups less likely to receive appropriate reperfusion therapy, this study identifies the specific reasons for the failure to administer reperfusion therapy and suggests that systematic changes in the approach to patients presenting with AMI are likely necessary to ensure the identification of all appropriate treatment candidates. Such efforts will likely enhance outcomes for the population of patients presenting with AMI, where timely reperfusion can translate into substantial survival and functional benefits.
American Heart Journal Volume 155, Number 3
Tricomi et al 477
References 1. Barron HV, Bowlby LJ, Breen T, et al. Use of reperfusion therapy for acute myocardial infarction in the United States: data from the National Registry of Myocardial Infarction 2. Circulation 1998;97: 1150-6. 2. Fox KA, Goodman SG, Klein W, et al. Management of acute coronary syndromes. Variations in practice and outcome; findings from the Global Registry of Acute Coronary Events (GRACE). Eur Heart J 2002;23:1177-89. 3. Oka RK, Fortmann SP, Varady AN. Differences in treatment of acute myocardial infarction by sex, age, and other factors (the Stanford Five-City Project). Am J Cardiol 1996;78:861-5. 4. Eagle KA, Goodman SG, Avezum A, et al. Practice variation and missed opportunities for reperfusion in ST-segmentelevation myocardial infarction: findings from the Global Registry of Acute Coronary Events (GRACE). Lancet 2002;359: 373-7. 5. Masoudi FA, Magid DJ, Vinson DR, et al. Implications of the failure to identify high-risk electrocardiogram findings for the quality of care of patients with acute myocardial infarction: results of the Emergency Department Quality in Myocardial Infarction (EDQMI) study. Circulation 2006;114:1565-71. 6. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction—executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2004;110:588-636. 7. Fibrinolytic Therapy Trialists' (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results
8.
9.
10.
11.
12.
13.
14.
15.
from all randomised trials of more than 1000 patients. Lancet 1994;343:311-22. Grines CL, Browne KF, Marco J, et al. A comparison of immediate angioplasty with thrombolytic therapy for acute myocardial infarction. The Primary Angioplasty in Myocardial Infarction Study Group. N Engl J Med 1993;328:673-9. Krumholz HM, Murillo JE, Chen J, et al. Thrombolytic therapy for eligible elderly patients with acute myocardial infarction. JAMA 1997;277:1683-8. Brady WJ, Perron A, Ullman E. Errors in emergency physician interpretation of ST-segment elevation in emergency department chest pain patients. Acad Emerg Med 2000;7:1256-60. Go AS, Barron HV, Rundle AC, et al. Bundle-branch block and in-hospital mortality in acute myocardial infarction. National Registry of Myocardial Infarction 2 Investigators. Ann Intern Med 1998;129: 690-7. Pope JH, Ruthazer R, Kontos MC, et al. The impact of electrocardiographic left ventricular hypertrophy and bundle branch block on the triage and outcome of ED patients with a suspected acute coronary syndrome: a multicenter study. Am J Emerg Med 2004;22:156-63. Selker HP, Beshansky JR, Griffith JL. Use of the electrocardiographbased thrombolytic predictive instrument to assist thrombolytic and reperfusion therapy for acute myocardial infarction. A multicenter, randomized, controlled, clinical effectiveness trial. Ann Intern Med 2002;137:87-95. McLaughlin SA, Doezema D, Sklar DP. Human simulation in emergency medicine training: a model curriculum. Acad Emerg Med 2002;9:1310-8. McGovern PG, Pankow JS, Shahar E, et al. Recent trends in acute coronary heart disease—mortality, morbidity, medical care, and risk factors. N Engl J Med 1996;334:884-90.
Background Although it is known that reperfusion therapy for ST-elevation myocardial infarction (STEMI) is underused, the reasons for the failure to provide this potentially life-saving treatment are not well described. Methods In a cohort of 2215 consecutive patients presenting with acute myocardial infarction to 5 emergency departments in Colorado and California between 2000 and 2002, patients with ischemic symptoms and ST-segment elevation on electrocardiogram without documented guideline-based contraindications to therapy were identified as eligible reperfusion candidates. Multivariable logistic models were constructed to identify factors associated with the failure to receive reperfusion. The emergency department records of patients not receiving reperfusion were reviewed to categorize the reasons therapy was not provided. Results Of 460 eligible patients, 102 (22%) did not receive reperfusion therapy. Patient factors associated with failure to receive reperfusion therapy included older age, peripheral vascular disease, and absence of chest pain; patients seen by both resident and attending physicians were more likely to receive treatment than those seen by an attending alone. In cases where reperfusion was not provided, ST-segment elevation was not identified in 34% (n = 35), left bundle-branch block was not considered as an indication in 13% (n = 13), there was documentation of a reason for withholding therapy not supported by guidelines in 34% (n = 35), and there was no documentation of reasons for withholding reperfusion in 19% (n = 19). Conclusions Initiatives to improve electrocardiogram interpretation and evidence-based patient selection may reduce gaps in the delivery of reperfusion therapy to eligible candidates and thereby potentially improve STEMI outcomes. (Am Heart J 2008;155:471-7.) Several studies have shown that a significant proportion of eligible patients with ST-elevation myocardial infarction (STEMI) do not receive potentially life-saving reperfusion therapy. For instance, 24% of eligible patients with STEMI in the NRMI and 30% in the GRACE did not receive reperfusion therapy.1,2 Prior studies have identi-
From the aUniversity of Rochester School of Medicine, Rochester, New York, NY, bClinical Research Unit, Colorado Permanente Medical Group, Denver, CO, cUniversity of Colorado at Denver and Health Sciences Center, Denver, CO, dDenver VA Medical Center, Denver, CO, eKaiser Permanente Medical Center, Roseville, CA, and fDenver Health Medical Center, Denver, CO. Submitted May 18, 2007; accepted October 23, 2007. Reprint requests: Frederick A Masoudi, MD, MSPH Cardiology, MC0960, Denver Health Medical Center, 777 Bannock St. Denver, CO. E-mail: [email protected] 0002-8703/$ - see front matter © 2008, Published by Mosby, Inc. doi:10.1016/j.ahj.2007.10.023
fied specific patient factors associated with a lower likelihood of receiving reperfusion, including advanced age, female sex, nonwhite race, and noncardiac comorbidities.1,3,4 Although recognizing the patient factors associated with failure to receive reperfusion therapy is important, doing so does not directly support changes in the delivery of care to increase appropriate treatment. To date, little investigation has been dedicated to identifying the underlying reasons why reperfusion therapy is not administered to some eligible patients. An understanding of the specific reasons for the failure to provide appropriate care is a more informative to the development programs to optimize care delivery for acute myocardial infarction (AMI). The goal of this study was to better understand the reasons why eligible patients with STEMI did not receive fibrinolysis or primary percutaneous coronary
American Heart Journal March 2008
472 Tricomi et al
Figure 1
The study population. PCI, Percutaneous coronary intervention; EDQMI, Emergency Department Quality in Myocardial Infarction; LBBB, left bundle branch block; STEMI, ST-segment elevetion mycardial infarction.
intervention (PCI) in a multicenter cohort of patients presenting to the emergency department (ED). By reviewing the ED medical record and electrocardiograms (ECGs), our objective was to identify and categorize the causes for the failure to provide this important therapy. The results of this study are intended to inform initiatives to provide the highest quality of AMI care.
Methods Study design and setting The EDQMI project was a 2-year retrospective cohort study focusing on the quality of ED care for AMI.5 Consecutive patients presenting to 5 community EDs in Colorado and California from July 1, 2000, through June 30, 2002 were screened. Subjects were included if they (1) presented directly to the ED, (2) had elevated cardiac markers diagnostic for AMI in the ED or within 24 hours of ED arrival, and (3) had a discharge diagnosis of AMI according to the International Classification of Diseases, Ninth Revision, Clinical Modification (code 410.X). Elevated cardiac markers were defined as (1) elevations of total creatine kinase or creatine kinase–MB at least 2 times the upper limit of the reference range or (2) elevations in troponin that exceeded the reference and indeterminate range. Those patients who were directly admitted to the inpatient setting, transferred after initial ED care at another institution, or whose diagnostic cardiac marker elevations occurred N24 hours after ED arrival were excluded. Of the 5 hospitals participating in the study, 3 had cardiac catheterizations laboratories and 2 provided primary PCI. Institutional review board approval was obtained at all 5 study locations.
Study sample Patients were considered eligible reperfusion candidates if they presented with possible ischemic symptoms beginning within 12 hours of ED presentation; had an ECG obtained in the ED demonstrating ST-segment elevation of at least 1 mm in ≥2 contiguous leads or left bundle-branch block (LBBB) that was new or not known to be old; and had no contraindications to reperfusion therapy. Contraindications to reperfusion therapy included those previously described in the American Heart Association/American College of Cardiology guidelines.6 To identify this eligible cohort, we excluded patients who did not have ST-segment elevation or LBBB on an initial or subsequent ED ECG (n = 1619); those who presented N12 hours after symptoms onset (n = 51); and those who had documented contraindications to reperfusion therapy or other documented reasons (eg, patient refusal, n = 5) why reperfusion therapy was considered but not given (n = 85). The remaining patients (n = 460) were considered eligible candidates for reperfusion (Figure 1).
Data sources Data on the ED care processes were obtained through review of ED records. Four physician investigators conducted chart reviews using an electronic data abstraction tool. Data were collected on patient demographics, clinical history including contraindications to reperfusion, ED provider's ECG interpretation, and ED therapies. All abstractors received training on the content and coding of each data element, data handling and data transmission procedures, and protocols to handle possible questions or problems during the study. During the training period, abstractors also coded 10 practice charts. Although
American Heart Journal Volume 155, Number 3
Tricomi et al 473
Table I. Baseline characteristics of reperfusion-eligible patients Received reperfusion (n = 358)
Did not receive reperfusion (n = 102)
n
%
N
%
P⁎
Age † b65 65-80 N80
63.7 181 142 35
12.6 50.6 39.7 9.8
69.5 37 32 33
15.8 36.3 31.4 32.4
b.001 b.001
Female Male
109 249
30.4 69.6
38 64
37.3 62.7
.2
Caucasian Non Caucasian Other/unknown
236 68 54
65.9 19.0 15.1
66 19 17
64.7 18.6 16.7
b.001
Diabetes Hypertension Hypercholesterolemia Current smoker
70 160 181 111
19.6 44.7 50.6 31.0
24 51 40 25
23.5 50.0 39.2 24.5
.4 .3 .04 .2
44 31 20 12 21
12.3 8.7 5.6 3.4 5.9
15 6 8 11 9
14.7 5.9 7.8 10.8 8.8
.5 .4 .4 .002 .3
7 29 18 12
2.0 8.1 5.0 3.4
8 9 10 12
7.8 8.8 9.8 11.8
.003 .8 .08 b.001
17 341
4.7 95.3
23 79
22.5 77.5
b.001
297 61
83.0 17.0
96 6
94.1 5.9
b.001
Demographics Age (y)
Sex
Race
Cardiac risk factors
Cardiac history Previous MI PCI CABG CHF Angina Medical history CRI COPD Stroke PVD Chest pain at presentation No Yes Seen by resident in addition to attending physician in ED No Yes
CABG, Coronary artery bypass grafting; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CRI, chronic renal insufficiency; PVD, peripheral vascular disease. ⁎P value is for the comparison between patients receiving reperfusion and those not receiving reperfusion. †Presented as mean and SD.
the abstractors were aware that the objective of the EDQMI project was to assess the quality of ED AMI care, they were not aware at the time of the chart review of the goals of this specific analysis. Multiple processes were instituted to enhance the accuracy and reliability of the data abstraction process. A data manager monitored day-to-day data collection activities, answered coding or eligibility questions, and maintained a log of all coding decisions made subsequent to the start of the project. After the abstraction was complete and the data were entered, the data manager verified case eligibility, confirmed the validity of the responses, and attempted to locate any missing data. To assess interrater reliability, all abstractors coded a random sample of 30 charts, and the results of these abstractions were compared to a “gold-standard” abstraction. For the key outcome variable of the provision of reperfusion therapy, there was perfect interrater reliability for all abstractors (κ = 1.0).
In addition to the ED chart reviews, all ECGs obtained in the ED were reviewed by 1 of 3 board-certified cardiologists. These interpretations included an assessment for ST-segment elevation of at least 0.1 mV in at least 2 contiguous leads and LBBB. In cases where LBBB was present, the presenting ECG was compared with a prior tracing whenever possible to determine if the finding was new. When the prior tracing was not available, information from the chart abstraction was used to determine whether the LBBB was new or presumably new.
Categorization of failures to provide reperfusion Patients were considered to receive reperfusion therapy if they received a fibrinolytic agent or underwent coronary angiography with the intent of providing PCI regardless of the success of the reperfusion therapy. The ED charts and ECGs of patients who were initially classified as eligible but did not receive reperfusion, including those whose initial chart review
474 Tricomi et al
identified a contraindication to therapy, were reviewed again by a separate panel of 4 investigators to confirm the patient's eligibility for reperfusion and to assess for specific treatment contraindications. The confirmed cases of eligible reperfusion candidates not receiving reperfusion (n = 102) were assessed to determine the underlying reasons for the failure to provide treatment. All failures to provide reperfusion could be assigned to 1 of 4 categories: (1) failure to identify ST-elevation on the ED ECG, (2) failure to identify new or presumed new LBBB as an inclusion criterion for therapy, (3) documentation of a reason for withholding reperfusion that was not concordant with guideline recommendations, or (4) failure to document any rationale for withholding reperfusion.
Data analysis Among eligible candidates for reperfusion therapy, we compared characteristics of patients who received and did not receive reperfusion therapy with t tests for continuous variables and χ2 tests for categorical variables. We used multivariable hierarchical regression models to identify patient factors independently associated with the failure to provide reperfusion, adjusting for patient demographic and clinical characteristics and accounting for the clustering of patients within hospitals. Relatively small numbers of patients per physician precluded an analysis that accounted for clustering of patients within physician. The outcome variable for these models was reperfusion therapy (receipt of PCI or fibrinolysis). Regression models were developed using a stepwise approach in which covariates were assessed based on their contribution to the overall model. In deciding whether to include a covariate in the model, we considered whether the estimate of the coefficients of some of the primary predictor variables (eg, age, chest pain) meaningfully changed when the covariate was removed from the model as well as whether including the covariate improved precision. At the conclusion of the model building process, we assessed the global fit of the model using the Hosmer-Lemeshow test and the local fit using the Akaike's information criteria and the deviance. Including calendar time as a candidate variable allowed for the assessment of potential secular trends in the administration of reperfusion therapy and temporal variation in case mix. Odds ratios and 95% CIs were calculated for all covariates in the models. Statistical analyses were performed using SAS version 8 (SAS Inc, Cary, NC).
Results Characteristics of eligible reperfusion patients Of the 2215 patients enrolled in EDQMI, 596 patients presented with ST-segment elevation or new LBBB. After excluding the patients who presented with N12 hours of ischemic symptoms or who had guidelineconcordant contraindications to reperfusion (n = 136), 460 eligible reperfusion candidates were identified for inclusion in this study (Figure 1). The mean age of eligible candidates was 64.9 years, 32% were female, and 65% were Caucasian. There were high rates of traditional cardiac risk factors, and 13% of the study population had a history of previous MI, 8% had previously undergone PCI, and 6% had a history of prior coronary artery bypass grafting (Table I).
American Heart Journal March 2008
Rates of reperfusion therapy and factors associated with reperfusion Of the 460 eligible reperfusion candidates, 358 (78%) received reperfusion therapy. Of the eligible candidates, 114 (32%) underwent urgent coronary angiography and subsequent primary PCI if the culprit lesion was amenable to treatment and 244 (68%) received fibrinolysis. The remaining 102 (22%) eligible patients did not receive reperfusion therapy (Figure 1). Patients who did not receive reperfusion were on average older and were more likely to have a history of important comorbidities, including heart failure, renal insufficiency, and peripheral vascular disease (Table I). Patients who did not receive reperfusion were also less likely to present with typical chest discomfort and were less likely to be seen by a resident in conjunction with an attending physician. In multivariable analysis of eligible candidates, younger patients were more likely to receive reperfusion than older patients (OR 4.10, 95% CI 1.83-9.23 for age b65 years compared with age N80 years; OR 3.43, 95% CI 1.63-7.23 for age 65-80 years compared with age N80 years). Patients with chest pain as a presenting symptom were more likely to receive reperfusion than those who presented with atypical symptoms (OR 4.99, 95% CI 2.22-11.21). Patients with peripheral vascular disease were less likely to receive reperfusion (OR 0.23, 95% CI 0.08-0.71). Patients seen by a resident in conjunction with an attending physician were more likely to receive reperfusion compared with those seen by an attending physician alone (OR 3.11, 95% CI 1.17-8.29). Failures to treat eligible candidates Among the 102 eligible reperfusion candidates who did not receive therapy, ST-segment elevation was not correctly identified in 35 (34%). Of these 35 cases, 30 patients had an initial ED ECG that showed ST elevation and the other 5 patients presented initially without ST-segment elevation but developed this finding on a subsequent ECG obtained in the ED. In 13 cases (13%), the ED ECG showed an LBBB, but there was no mention in the chart as to whether the LBBB was new or old, nor was there documentation to suggest that reperfusion therapy was considered (Figure 1). In another 19 cases (19%), the ECG was correctly interpreted; however, reperfusion therapy was withheld for a reason that was not concordant with the guideline recommendations.6 A common reason for withholding reperfusion was the presence of intermittent symptoms over the several days before presentation (n = 6). Although these patients often had been experiencing anginal episodes for several days, the current episode of pain began b12 hours before their presentation to the ED, supporting their eligibility for reperfusion therapy. A second common reason for not providing reperfusion occurred in patients whose symptoms or ST elevation on ECG improved but did not resolve with
American Heart Journal Volume 155, Number 3
Table II. Selected reasons cited for not providing reperfusion therapy • Currently asymptomatic • Patient had 4 d of intermittent angina and with worsening persistent pain at presentation • Patient's old age without further qualification • Warfarin recently stopped, but INR was not elevated • Concern for abdominal aortic aneurysm, but CT negative for AAA • Subsequent ECG with less ST elevation • Pain resolved with other medications • Patient had been having pain for 10 h INR, International normalized ratio; CT, computed tomography; AAA, abdominal aortic aneurysm.
standard medical therapy (n = 5). Examples of the non– guideline-based reasons that were documented in patient charts for not providing reperfusion therapy to eligible patients are provided in Table II. In the remaining 35 cases (34%), ST elevation or new LBBB was identified by ED physician, but there was no documentation of any contraindication to reperfusion the medical record.
Discussion The primary objective of this study of patients with STEMI presenting to the ED was to assess the underlying reasons why eligible reperfusion candidates did not receive reperfusion. We found that more than 1 in 5 reperfusion-eligible patients with MI did not receive reperfusion therapy. Individual chart review revealed that the failure to identify significant ST-segment elevation, the lack of recognition of new or presumably new LBBB as an indication for reperfusion, and failure to identify guideline-based indications for reperfusion all played important roles in the failure to deliver reperfusion therapy. Despite substantial clinical trial evidence and support in practice guidelines for the use of reperfusion in selected patients, many who could benefit from reperfusion are deprived of this potentially life-saving therapy.6-8 Our study of ED patients shows rates of use of reperfusion therapy among eligible patients similar to prior studies, and identifies similar specific patient factors associated with the underuse of reperfusion—including age and comorbidities such as heart failure, renal insufficiency, and peripheral vascular disease.1,4,9 Unlike these prior investigations, the present study included a review of each patient's ED ECGs and a detailed case adjudication to confirm eligibility for treatment. Thus, this study is able to go beyond the identification of important gaps in the delivery of reperfusion therapy to patients in community settings, using detailed chart review to assess patient eligibility and identify specific reasons why reperfusion therapy was withheld. Although identifying patients at risk for undertreatment is important, the process of describing the reasons for the failure to provide therapy is markedly more informative to the development of quality improvement
Tricomi et al 475
initiatives. The current study identifies discrete and potentially remediable factors that pose important barriers to the provision of indicated treatment, including the failure to recognize ST-segment elevation, lack of knowledge of the importance of LBBB, and inadequate knowledge of contraindications to reperfusion. Among patients in this study who did not receive appropriate reperfusion therapy, one third had ECGs with ST-segment elevation that was not identified at either presentation or on subsequent tracings. A prior study from EDQMI found that the failure to identify high-risk ECG findings was associated with lower rates of treatment with evidence-based therapy, including reperfusion, among eligible candidates.5 This study confirms the importance of the ECG interpretation and suggests that a third of all failures to provide reperfusion therapy are attributable to the failure to identify ST-segment elevation. Although a prior study found that the failure to identify ST-segment elevation among all patients presenting with chest pain had modest consequences,10 this study of patients with documented AMI indicates that missing ST-segment elevation plays a central role in the failure to provide potentially lifesaving therapy. Although hospital mortality is higher for MI patients with LBBB than those without,11 and providing reperfusion therapy to patients with ischemic symptoms and LBBB that is not known to be old is recommended by guidelines,6 the present study corroborates previous data showing that such patients are undertreated with reperfusion.1 Despite prior research reporting that physicians have a lower threshold for hospitalizing patients with LBBB when presenting with chest pain,12 our findings indicate that this does not necessarily translate into higher rates of reperfusion therapy for eligible candidates. Thus, this study reemphasizes the need for providers who treat patients with MI in the immediate setting to understand the importance of LBBB as criterion for eligibility for immediate reperfusion therapy. Although misinterpretation of ST elevation and reluctance to aggressively intervene in patients with LBBB appear to be important causes of failure to provide reperfusion, we also observed that reperfusion therapy is either inappropriately dismissed or not considered at all for some eligible patients. In 19% of the missed opportunities for reperfusion therapy, the provider documented a perceived contraindication to reperfusion not supported by guidelines as rationale for withholding therapy. This finding suggests that improving the understanding of the evidence-based selection of appropriate candidates for treatment may improve the delivery of reperfusion therapy. In the remaining 34% of cases where eligible patients did not receive reperfusion, there was no reason for withholding reperfusion documented in the medical
476 Tricomi et al
record. Because of the difficulty in conclusively proving the absence of contraindications through chart abstraction, we cannot entirely exclude the possibility that true contraindications to reperfusion therapy were present but not documented in some cases. However, the consideration of reperfusion therapy was not documented in any of these cases, supporting the conclusion that these do not merely reflect the failure to document contraindications among ineligible patients. This study identifies possible mechanisms to enhance the delivery of reperfusion to eligible candidates with STEMI. Increased attention to the interpretation of the ECG as well as the ECG criteria for reperfusion during postgraduate training and continuing medical education courses may improve the identification of eligible patients. In addition, based on the findings of this and previous studies, specific emphasis on the importance of LBBB and the contraindications for reperfusion based on existing practice guidelines are warranted. Based on the literature of systems change, however, it is unlikely that education alone will be adequate to enhance care. The findings of this study also suggest that fundamental changes in the structure of care delivery may enhance the delivery of reperfusion therapy. Such systems might include secondary ECG interpretations by another provider. The finding that care including both a resident and an attending physician was associated with higher rates of reperfusion supports the hypothesis that secondary ECG interpretations may improve care. Although a secondary ECG interpretation by another provider may improve the identification of eligible patients, this strategy—requiring the evaluation of a large number of ECGs to identify the smaller group of reperfusion eligible patients—would necessitate significant resources and could potentially contribute to important delays in reperfusion therapy. Automated ECG interpretation could serve this role; however, although automated ECG interpretation has been shown in randomized trials to improve the accuracy of patient triage for reperfusion,13 such systems have not been widely implemented. Incorporating checklists of the appropriate selection criteria for reperfusion in critical pathways for patients presenting with chest pain may be useful. Finally, simulation-based training represents an emerging approach that may improve the quality of care for STEMI patients. This strategy has been most extensively used in anesthesia training, but more recent data suggest that it may also be a useful tool in emergency medicine and critical care education.14 In interpreting this study, certain factors warrant consideration. Because the study implemented retrospective medical record review, we cannot rule out the possibility of biases due to the presence of undocumented guideline-based contraindications to therapy or missing data. Undocumented guideline-based contraindications would result in an overestimation of the
American Heart Journal March 2008
number of eligible treatment candidates as well as the proportion of untreated patients. Nevertheless, among those patients in this study for whom inadequate documentation was identified, more than one third had documentation of a contraindication that was not concordant with guideline recommendations. Thus, many of these cases likely represent meaningful opportunities to improve care. Missing ECGs that were available to the treating provider but not documented in the record and not available for adjudication may also have increased the number of patients categorized as eligible. Although failures of documentation are arguably important, they may not have an impact on patient care comparable with the failure to provide appropriate care. These data were collected retrospectively and did not include detailed interviews of providers. Thus, it was not possible to assess all important aspects of care (eg, involvement of a cardiologist) or to better understand the thinking of clinicians who cared for patients who did not receive therapy. Studies involving detailed interviews would be useful complements to this study in better understanding the failure to provide reperfusion to eligible patients. Second, the panel of investigators performing medical record and chart review were aware that each patient in the study sample had been diagnosed with an MI. However, during ECG review, they were blinded to the results of the chart review and the original ECG interpretation. Third, the impact of the failure to provide reperfusion therapy on patient outcomes was not assessed. However, numerous studies have demonstrated higher mortality among eligible candidates who are not treated both in the context of trials and in community-based samples.1,15 Finally, as our study focused on the MI care provided in the ED, we were unable to assess the care of the patient in the hospital after leaving the ED. However, given the critical importance of time in the provision of reperfusion in patients with STEMI, the lack of any evidence of providing reperfusion to eligible patients in the ED record reflects meaningful gaps in care.
Conclusions We found that more than 20% of ED patients with AMI who are eligible candidates for reperfusion therapy do not receive this important evidenced-based intervention. In addition to identifying those patient groups less likely to receive appropriate reperfusion therapy, this study identifies the specific reasons for the failure to administer reperfusion therapy and suggests that systematic changes in the approach to patients presenting with AMI are likely necessary to ensure the identification of all appropriate treatment candidates. Such efforts will likely enhance outcomes for the population of patients presenting with AMI, where timely reperfusion can translate into substantial survival and functional benefits.
American Heart Journal Volume 155, Number 3
Tricomi et al 477
References 1. Barron HV, Bowlby LJ, Breen T, et al. Use of reperfusion therapy for acute myocardial infarction in the United States: data from the National Registry of Myocardial Infarction 2. Circulation 1998;97: 1150-6. 2. Fox KA, Goodman SG, Klein W, et al. Management of acute coronary syndromes. Variations in practice and outcome; findings from the Global Registry of Acute Coronary Events (GRACE). Eur Heart J 2002;23:1177-89. 3. Oka RK, Fortmann SP, Varady AN. Differences in treatment of acute myocardial infarction by sex, age, and other factors (the Stanford Five-City Project). Am J Cardiol 1996;78:861-5. 4. Eagle KA, Goodman SG, Avezum A, et al. Practice variation and missed opportunities for reperfusion in ST-segmentelevation myocardial infarction: findings from the Global Registry of Acute Coronary Events (GRACE). Lancet 2002;359: 373-7. 5. Masoudi FA, Magid DJ, Vinson DR, et al. Implications of the failure to identify high-risk electrocardiogram findings for the quality of care of patients with acute myocardial infarction: results of the Emergency Department Quality in Myocardial Infarction (EDQMI) study. Circulation 2006;114:1565-71. 6. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction—executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2004;110:588-636. 7. Fibrinolytic Therapy Trialists' (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results
8.
9.
10.
11.
12.
13.
14.
15.
from all randomised trials of more than 1000 patients. Lancet 1994;343:311-22. Grines CL, Browne KF, Marco J, et al. A comparison of immediate angioplasty with thrombolytic therapy for acute myocardial infarction. The Primary Angioplasty in Myocardial Infarction Study Group. N Engl J Med 1993;328:673-9. Krumholz HM, Murillo JE, Chen J, et al. Thrombolytic therapy for eligible elderly patients with acute myocardial infarction. JAMA 1997;277:1683-8. Brady WJ, Perron A, Ullman E. Errors in emergency physician interpretation of ST-segment elevation in emergency department chest pain patients. Acad Emerg Med 2000;7:1256-60. Go AS, Barron HV, Rundle AC, et al. Bundle-branch block and in-hospital mortality in acute myocardial infarction. National Registry of Myocardial Infarction 2 Investigators. Ann Intern Med 1998;129: 690-7. Pope JH, Ruthazer R, Kontos MC, et al. The impact of electrocardiographic left ventricular hypertrophy and bundle branch block on the triage and outcome of ED patients with a suspected acute coronary syndrome: a multicenter study. Am J Emerg Med 2004;22:156-63. Selker HP, Beshansky JR, Griffith JL. Use of the electrocardiographbased thrombolytic predictive instrument to assist thrombolytic and reperfusion therapy for acute myocardial infarction. A multicenter, randomized, controlled, clinical effectiveness trial. Ann Intern Med 2002;137:87-95. McLaughlin SA, Doezema D, Sklar DP. Human simulation in emergency medicine training: a model curriculum. Acad Emerg Med 2002;9:1310-8. McGovern PG, Pankow JS, Shahar E, et al. Recent trends in acute coronary heart disease—mortality, morbidity, medical care, and risk factors. N Engl J Med 1996;334:884-90.