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
Background— The impact of misinterpretation of the ECG in patients with acute myocardial infarction (AMI) in the emergency department (ED) setting is not well known. Our goal was to assess the prevalence of the failure to identify high-risk ECG findings in ED patients with AMI and to determine whether this failure is associated with lower-quality care.
Methods and Results— In a retrospective cohort study of consecutive patients presenting to 5 EDs in California and Colorado from July 1, 2000, through June 30, 2002, with confirmed AMI (n=1684), we determined the frequency of the failure by the treating provider to identify significant ST-segment depressions, ST-segment elevations, or T-wave inversions on the presenting ECG. In multivariable models, we assessed the relationship between missed high-risk ECG findings and evidence-based therapy in the ED after adjustment for patient characteristics and site of care. High-risk ECG findings were not documented in 201 patients (12%). The failure to identify high-risk findings was independently associated with a higher odds of not receiving treatment among ideal candidates for aspirin (odds ratio [OR], 2.13; 95% confidence interval [CI], 1.51 to 2.94), β-blockers (OR, 1.85; 95% CI, 1.14 to 3.03), and reperfusion therapy (OR, 7.69; 95% CI, 3.57 to 16.67). Among patients with missed high-risk ECG findings, in-hospital mortality was 7.9% compared with 4.9% among those without missed findings (P=0.1).
Conclusions— The failure to identify high-risk ECG findings in patients with AMI results in lower-quality care in the ED. Systematic processes to improve ECG interpretation may have important implications for patient treatment and outcomes.
Received February 27, 2006; revision received June 30, 2006; accepted August 4, 2006.
The identification of acute myocardial infarction (AMI) in patients presenting to the emergency department (ED) with symptoms of ischemia is critical to delivering appropriate medical care. Current guidelines stress the importance of identifying patients who have an increased likelihood of AMI and are thus likely to benefit from the prompt delivery of evidence-based medical therapy.1–3 The failure to identify AMI and to treat these high-risk patients may have important negative implications for patient outcomes.
Clinical Perspective p 1571
The interpretation of the ECG in the ED is central to the assessment of patients with possible cardiac ischemia.1–3 High-risk findings on the ECG inform the diagnosis and treatment of patients with suspected AMI. Several investigators have studied ECG interpretation by ED care providers,4–8 but the impact of these discrepancies on triage and treatment decisions has been debated.4,5 Although some studies suggest that ECG misinterpretation may result in inappropriate patient triage7,9 or missed opportunities to provide acute reperfusion therapy,10,11 little is known about the relationship of ECG interpretation to the quality of care for AMI delivered in the ED.
Accordingly, we assessed the frequency of failure to identify high-risk ECG findings in the ED setting in a multicenter cohort of patients presenting with AMI and the degree to which missed ECG findings were associated with the failure to provide therapy with aspirin, β-blockers, or reperfusion therapy in ideal treatment candidates. The results of this study may have important implications for the process of ECG interpretation in the ED, the training of ED providers in ECG interpretation, and the development and testing of ED quality improvement interventions.
Patient Population and Setting
The Emergency Department Quality in Myocardial Infarction (EDQMI) project was a 2-year retrospective cohort study focusing on the quality of care of AMI in the ED. Consecutive patients presenting to 5 EDs in California and Colorado from July 1, 2000, through June 30, 2002, were screened. Subjects were included if they presented directly to the ED, had an elevated cardiac marker diagnostic of AMI in the ED or within 24 hours of ED arrival, and had a discharge diagnosis of AMI according to the International Classification of Diseases, 9th revision, clinical modification (code 410.X). An elevated cardiac marker was defined as either an elevation of total creatine kinase or creatine kinase-MB at least 2 times the upper limit of the normal range or an elevation in troponin that exceeded the normal and indeterminate ranges. Patients were excluded if they were directly admitted to the inpatient setting, if they were transferred after initial ED care at another institution, or if the diagnostic cardiac marker elevations occurred >24 hours after ED arrival. The total EDQMI cohort consisted of 2215 patients. For the purposes of this study, patients with left bundle-branch block were excluded because ST-T changes on the ECG for ischemia or infarction lack sensitivity in the presence of left bundle-branch block (n=101).12,13 Among patients without left bundle-branch block, those whose records did not include a documented ECG interpretation by the ED physician (n=342) or whose ECGs were not available for adjudication (n=88) were not included in the analysis. The analytic cohort for the present study comprised 1684 patients.
The 5 hospitals included in this study were members of not-for-profit health maintenance organizations and ranged in size from 115 to 456 beds. Three of the 5 had cardiac catheterization laboratories, and 2 provided percutaneous coronary interventions on site. Institutional Review Board approval was obtained for the conduct of the study at all participating sites.
Data were collected from the ED record on patient demographics, clinical history, including contraindications to reperfusion, ED provider’s ECG interpretation, and ED therapies. Chart reviews were conducted by 4 physician investigators using an electronic data abstraction tool. All abstractors were trained with regard to the content and coding of each data element, data handling and data transmission procedures, and protocols to address questions during the study. In the training period, abstractors coded 10 practice charts. The abstractors were not aware of the specific hypotheses evaluated in this analysis.
Multiple processes were instituted to enhance the accuracy and reliability of abstraction. A data manager monitored data collection activities, answered coding or eligibility questions, and maintained a log of all coding decisions throughout the project. After the abstraction was complete, the data manager verified case eligibility and confirmed the validity of responses. To assess interrater reliability, a random sample of 30 charts was coded by all abstractors, and the results of these abstractions were compared with a reference standard abstraction. For all variables used in this study, there was moderate to excellent interrater reliability for all abstractors (κ=0.5 to 1.0).
The interpretation of the ECG by the ED provider was abstracted from the medical record. These interpretations were categorized as indicating possible or definite ischemia/injury if there was mention of ST-segment elevation, ST-segment depression, or T-wave inversion, unless it was explicitly noted that the findings were “nonspecific.” Other qualitative terms such as “acute MI,” “acute injury,” or “ischemia” also were considered to fulfill the criteria for possible or definite ischemia or injury.
Hard copies of each patient’s ECG tracing(s) obtained in the ED, and prior tracings if available, were reviewed by 1 of 4 cardiologists who were blinded to the patient’s clinical record and the ED provider’s interpretation. Using the American College of Cardiology/American Heart Association (ACC/AHA) guidelines to define high-risk ECG findings,1,2 these interpreters considered any of the following in ≥2 contiguous leads as indicative of possible or definite ischemia: ST-segment elevation of at least 0.1 mV, ST-segment depression of at least 0.05 mV, or T-wave inversion of at least 0.2 mV. If prior ECGs were available for review, index findings that had been present on an earlier ECG were considered nonspecific.
All cases in which the cardiologist identified possible or definite ischemia but the ED provider did not (n=209) were subsequently reviewed by a panel of 4 physicians to adjudicate the presence or absence of discordance. This adjudication process identified 8 cases (3.8%) in which the original cardiologist’s interpretation was overruled. The remaining 201 cases were confirmed as having a missed high-risk ECG finding during the ED encounter.
Definitions of Ideal Candidates for Evidence-Based Therapy
Patients were considered ideal candidates for a therapy if they were eligible for, and had no contraindication to, that therapy based on current ACC/AHA guidelines.1,2 Patients were eligible for aspirin therapy if they had a positive cardiac marker in the ED or had an adjudicated ED ECG demonstrating new ST-segment elevation. Patients also were eligible for aspirin if they presented to the ED with discomfort or pressure in the chest, arm, neck, or jaw and had an adjudicated ED ECG demonstrating new ST-segment depression or T-wave inversion or had an admission diagnosis of unstable angina or “rule-out” myocardial infarction. Eligible patients were considered ideal treatment candidates in the absence of contraindications to aspirin therapy, which included active bleeding on or before ED arrival, aspirin allergy or hypersensitivity, an international normalized ratio ≥ 3.5, or other reasons documented by the ED provider for not prescribing aspirin. Patients were counted as having received aspirin therapy if they received aspirin or other antiplatelet agent (eg, clopidogrel) in the ED or if there was documentation that they had taken one of these agents in the 24 hours before arriving at the ED.
Patients were eligible for β-blocker therapy if they had a confirmed AMI while in the ED defined as either a positive cardiac marker in the ED or an adjudicated ED ECG demonstrating new ST-segment elevation. Eligible patients were considered ideal treatment candidates in the absence of contraindications to β-blockers, which included heart rate <60 bpm, systolic blood pressure <100 mm Hg, second- or third-degree heart block, heart failure or pulmonary edema documented on physical examination or chest x-ray, a history of asthma, known allergy to β-blockers, or any other reason documented by the ED provider for not prescribing a β-blocker.
Patients were eligible for reperfusion therapy (either fibrinolysis or percutaneous coronary intervention) if they presented to the ED within 12 hours of AMI symptom onset and had an adjudicated ED ECG demonstrating ST-segment elevation of at least 0.1 mV in ≥2 contiguous leads. Eligible candidates were considered ideal for treatment in the absence of contraindications to reperfusion therapy, which included cardiac arrest with cardiopulmonary resuscitation or other traumatic resuscitation; noncompressible vascular puncture; aortic dissection; other active bleeding processes; refractory hypertension (blood pressure >180 mm Hg systolic or >100 mm Hg diastolic); acute stroke or a history of hemorrhagic stroke, ischemic stroke, transient ischemic attack, central nervous system neoplasm, or other central nervous system lesion; recent major trauma; history of surgery or head injury within 2 months; known bleeding disorder; history of gastrointestinal bleeding or other internal bleeding within the last month; pregnancy or within 1 week postpartum; chronic liver disease, infective endocarditis, active peptic ulcer, or oral anticoagulant therapy; or any other reason documented by the ED provider for not providing reperfusion therapy. Patients who were transferred for percutaneous coronary intervention were considered to have received reperfusion therapy.
Characteristics of patients with missed high-risk ECG findings were compared with other study subjects using t tests for continuous variables and χ2 tests for categorical variables. To identify factors independently associated with the failure to identify high-risk ECG findings, multivariable regression models were constructed with missed high-risk ECG as the dependent variable, adjusting for patient demographic and clinical characteristics (as listed in the Table) and for study site.
Subsequently, multivariable logistic models were constructed to assess the relationship between missed high-risk ECG findings and the use of aspirin, β-blockers, and reperfusion therapy. In these models, treatment with the evidence-based intervention (aspirin, β-blockers, or reperfusion therapy) was the dependent variable, and only ideal candidates for a treatment as defined above were included. After entering a variable indicating whether or not a high-risk ECG finding was missed, the models were expanded to include patient demographic and clinical characteristics and adjusted for site. All regression models were developed by applying backward selection to the demographic and clinical variables listed in the Table (P<0.10 for entry, P<0.05 to remain in model). Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for all covariates in the models. In the treatment models, ORs >1 indicate a lower likelihood of evidence-based treatment among ideal candidates.
Although the study was not powered to assess mortality differences, unadjusted in-hospital mortality rates were compared by use of χ2 tests. All statistical analyses were performed with SAS version 8.0 (SAS, Inc, Cary, NC). The authors had full access to the data and take full responsibility for their integrity. All authors have read and agree to the manuscript as written.
Of the 1684 patients in the study population, almost 40% were >75 years, and the majority were male and white (Table). Cardiovascular and other comorbidities were common. Most patients were hemodynamically stable, with 3.2% of patients presenting with hypotension and 6.5% in Killip class III or IV. Almost three quarters of patients presented within 6 hours of symptom onset.
Prevalence and Correlates of Missed High-Risk ECG Abnormalities
A total of 201 patients (12%) had a high-risk ECG abnormality that was not detected by the ED provider. Rates of missed ECG findings by hospital ranged from 5.6% to 15.1%. Patients with missed ECG abnormalities were older, more commonly had a history of heart failure or other cardiac history, and less frequently had chest pain as a presenting symptom (Table). Missed high-risk findings were less common among patients with ST-segment elevation (39 of 488, 8%) than for those with T-wave inversion (68 of 485, 14%) or ST-segment depression (138 of 762, 18%). Compared with patients whose charts contained documentation of ST-segment elevation, those without documentation on average had fewer leads with at least 1-mm ST-segment elevation (3.2±1.2 versus 2.9±1.0; P=0.4), had lesser degrees of ST elevation in the most affected lead (2.53±1.9 versus 2.1±1.0 mm; P=0.5), and less frequently had reciprocal ST-segment depression (76% versus 66%; P=0.2). None of these differences were statistically significant.
In multivariable models adjusting for patient characteristics and site, patients with a history of heart failure (OR, 1.78; 95% CI, 1.19 to 2.66; P=0.005) and other cardiovascular history (OR, 1.44; 95% CI, 1.04 to 2.00; P=0.03) were significantly more likely to have a missed high-risk ECG abnormality, whereas patients presenting with chest pain (OR, 0.46; 95% CI, 0.32 to 0.64; P<0.001) were significantly less likely to have a missed high-risk abnormality. Other patient and admission variables were not independently correlated after multivariable adjustment. Among the subgroup of patients with ST-segment elevation, patients with chest pain at presentation were significantly less likely to have this high-risk ECG finding missed (OR, 0.12; 95% CI, 0.05 to 0.25).
Missed High-Risk ECG Findings, Therapy, and Mortality
Overall, 1631 patients (96.8%) were ideal candidates for aspirin, 757 (45.0%) for β-blockers, and 403 (23.9%) for acute reperfusion therapy. Among ideal candidates, 81.4% were treated with aspirin, 60.9% were treated with β-blockers, and 79.2% received reperfusion therapy. In absolute terms, ideal candidates with missed high-risk ECG abnormalities were 21% less likely to receive aspirin, 20% less likely to receive β-blockers, and 48% less likely to receive reperfusion therapy (P<0.001 for all treatments; Figure). After adjustment for demographics, medical history, and admission characteristics, a missed high-risk ECG abnormality was associated with significantly higher odds of not receiving treatment among ideal candidates for aspirin (OR, 2.13; 95% CI, 1.51 to 2.94), β-blockers (OR, 1.85; 95% CI, 1.14 to 3.03), and reperfusion therapy (OR, 7.69; 95% CI, 3.57 to 16.67). Among patients with missed high-risk ECG findings, in-hospital mortality was 7.9% compared with 4.9% among those without missed findings (P=0.1).
The objectives of this study were to assess the frequency with which high-risk ECG abnormalities are missed in AMI patients presenting to the ED and to evaluate the relationship between missed ECG findings and quality of care. We found that 1 in 8 AMI patients had high-risk ECG findings that were not identified and that, among these patients, the delivery of evidence-based therapies in the ED, including aspirin, β-blockers, and acute reperfusion, was markedly lower. There was also a trend toward higher in-hospital mortality rates in patients with missed high-risk ECG findings. These findings suggest that the failure to identify ECG abnormalities is a critical shortfall in the process of caring for patients with AMI with important implications for treatment and potential adverse consequences for patient outcomes.
Our study suggests that the misinterpretation of the ECG in the ED setting is relatively common and has a consistent and significant relationship with lower rates of use of clinically effective interventions that have been shown to improve health outcomes. Randomized clinical trials clearly support the prompt use of medical therapy in eligible patients with AMI. In clinical trials of AMI, treatment has been associated with absolute reductions in short-term mortality ranging from 2% to 5% with aspirin treatment, 0.5% to 1% for β-blockers, and at least 3% for acute reperfusion.1,2,14–18 Based on these trials, current guidelines consider these interventions in the acute setting as class I recommendations, or those for which there is evidence and/or general agreement that they are beneficial, useful, and effective.1,2 Although our study was designed to assess differences in quality of care and was not adequately powered to ascertain differences in mortality, it is reasonable to believe that the poorer-quality care provided to those patients with missed high-risk ECG features would result in worse outcomes for the population with AMI.
The absence of chest pain on presentation was an important correlate of the failure to identify high-risk ECG findings, suggesting that the absence of typical historical findings may inordinately reduce clinicians’ level of suspicion for AMI. Prior studies have demonstrated that patients with AMI, particularly the elderly, frequently present with atypical symptoms and may not have classic angina.19–21 The findings of this study suggest that greater scrutiny should be dedicated to interpreting the ECG in the ED regardless of whether the patient presents with typical ischemic symptoms, particularly because the failure to identify high-risk findings is related to the failure to provide evidence-based therapy among patients likely to benefit from treatment.
Prior studies of ED care have assessed discrepancies in ECG interpretation, but the conclusions about the implications of these discrepancies have been mixed. In a single-center study of >700 patients with an abnormal ED ECG, the investigators found a discordance rate of nearly 60% but identified only 2 cases in which the discrepancy would alter patient care.5 In another study, ED interpretations were significantly discordant with cardiologist interpretations in almost 20%, but none of the missed abnormalities was considered potentially life-threatening.4 However, studies that consider all ECGs obtained in the ED regardless of the indication may understate the importance of ECG misinterpretation in care. Investigations limited to patients presenting with suspected AMI suggest that rates of ECG misinterpretation in this population are higher than in studies of general ED populations.7,9 Furthermore, misinterpretations in populations with chest pain have been associated with errors in triage, delays in reperfusion, or failures to provide reperfusion.7,9–11 In our study, providers failed to identify high-risk ECG findings in nearly 1 in 8 patients, and these failures had important implications for the quality of care.
Existing survey data suggest a perception that ECG interpretation skills by ED physicians are not an important target for improvement. ED training program directors generally consider the ECG training delivered to residents to be adequate,22 and most do not endorse the development of formal examinations to assess competency in ECG interpretation.23 However, the present study suggests otherwise, providing strong evidence that additional training in ECG interpretation and the assessment of competency may be a critical component of the education of physicians who care for patients presenting with AMI.
To this point, sustainable systematic solutions for increasing the accuracy of ECG interpretation in the ED that are acceptable to providers have been elusive. ECG-based instruments such as the thrombolytic predictive instrument24 or the acute cardiac ischemia time-insensitive predictive instrument25 have been studied in randomized trials and result in more prompt and appropriate therapy. Unfortunately, however, these instruments have been difficult to integrate consistently into clinical care. Studies of neural networks also show promise for enhancing ECG interpretation, but these systems have been experimental and have not been incorporated into clinical practice.26 The present study suggests, however, that innovative strategies to improve ECG interpretation in the ED that are acceptable for widespread implementation could result in substantial improvements in care for patients with AMI.
To our knowledge, the present study is the first to assess the relationship between the failure to identify high-risk ECG findings and the quality of ED care delivered to patients with AMI. Certain issues should be considered in the interpretation of the results, however. First, the study was conducted in 5 EDs, potentially limiting the generalizability of the results to other institutions. Furthermore, the study did not have adequate power to exclude clinically important differences in mortality that may have resulted from the failure of ECG interpretation. We found a strong relationship, however, between ECG interpretation and the provision of medical therapy that is known to improve health outcomes. Finally, we were not able to determine whether the gaps in care in the ED persisted during hospitalization. To the extent, however, that transitions in care may be a point where errors in care are transmitted and potentially amplified, the patterns identified in this study are likely important.
The failure to identify high-risk ECG abnormalities occurred in 1 in 8 patients in this study of individuals presenting to the ED with AMI and was strongly associated with the failure to provide evidence-based medical care. This study highlights the importance of systems changes to enhance the accuracy of ECG interpretation. Such measures are likely to result in improved ED care for patients with AMI, a critical step in improving health outcomes.
Sources of Funding
The EDQMI Study was funded by the Garfield Memorial Foundation. The funding source played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.
Dr Masoudi has served on speakers’ bureaus for Pfizer and AstraZeneca. The other authors report no conflicts.
Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M, Hochman JS, Krumholz HM, Kushner FG, Lamas GA, Mullany CJ, Ornato JP, Pearle DL, Sloan MA, Smith SC Jr, Alpert JS, Anderson JL, Faxon DP, Fuster V, Gibbons RJ, Gregoratos G, Halperin JL, Hiratzka LF, Hunt SA, Jacobs AK. 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 (Writing Committee to Revise the 1999 Guidelines for the Management of Patients With Acute Myocardial Infarction). Circulation. 2004; 110: 588–636.
Braunwald E, Antman EM, Beasley JW, Califf RM, Cheitlin MD, Hochman JS, Jones RH, Kereiakes D, Kupersmith J, Levin TN, Pepine CJ, Schaeffer JW, Smith EE III, Steward DE, Theroux P, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Fuster V, Gregoratos G, Hiratzka LF, Jacobs AK, Smith SC Jr. ACC/AHA guideline update for the management of patients with unstable angina and non-ST-SEGMENT ELEVATION MYOCARDIAL INFARCTION–2002: summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). Circulation. 2002; 106: 1893–1900.
Sgarbossa EB, Pinski SL, Barbagelata A, Underwood DA, Gates KB, Topol EJ, Califf RM, Wagner GS. Electrocardiographic diagnosis of evolving acute myocardial infarction in the presence of left bundle-branch block: GUSTO-1 (Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries) Investigators. N Engl J Med. 1996; 334: 481–487.
Collaborative overview of randomised trials of antiplatelet therapy, I: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ. 1994; 308: 81–106.
Long-term effects of intravenous thrombolysis in acute myocardial infarction: final report of the GISSI study. Lancet. 1987; 330: 871–874.
Goldberg R, Goff D, Cooper L, Luepker R, Zapka J, Bittner V, Osganian S, Lessard D, Cornell C, Meshack A, Mann C, Gilliland J, Feldman H. Age and sex differences in presentation of symptoms among patients with acute coronary disease: the REACT Trial. Coron Artery Dis. 2000; 11: 399–407.
Selker HP, Beshansky JR, Griffith JL. Use of the electrocardiograph-based 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.
Selker HP, Beshansky JR, Griffith JL, Aufderheide TP, Ballin DS, Bernard SA, Crespo SG, Feldman JA, Fish SS, Gibler WB, Kiez DA, McNutt RA, Moulton AW, Ornato JP, Podrid PJ, Pope JH, Salem DN, Sayre MR, Woolard RH. Use of the acute cardiac ischemia time-insensitive predictive instrument (ACI-TIPI) to assist with triage of patients with chest pain or other symptoms suggestive of acute cardiac ischemia: a multicenter, controlled clinical trial. Ann Intern Med. 1998; 129: 845–855.
Heden B, Ohlin H, Rittner R, Edenbrandt L. Acute myocardial infarction detected in the 12-lead ECG by artificial neural networks. Circulation. 1997; 96: 1798–1802.
Clinical guidelines consider ECG interpretation a critical component of the identification and risk stratification of patients presenting with suspected acute myocardial infarction. The clinical implications of the interpretation of the ECG in the emergency department (ED) setting are not well known, however, and survey data suggest that ECG interpretation skills in the ED are not perceived as an important target for improvement. We studied 1684 patients with confirmed acute myocardial infarction presenting to the ED, assessing the prevalence of the failure to identify high-risk ECG findings (significant ST-segment depressions, ST-segment elevations, or T-wave inversions) and the relationship between this failure and evidence-based ED care for eligible patients. High-risk findings were not identified in 12%, and these patients had a significantly higher likelihood of not receiving treatment with aspirin, β-blockers, and acute reperfusion. These findings suggest that correctly interpreting the ECG in the ED is a critical factor in the delivery of evidence-based care to patients with acute myocardial infarction, which may translate into improved health outcomes. Unfortunately, existing systems intended to facilitate ECG interpretation have not been widely implemented, highlighting the ongoing need for acceptable systematic approaches to improve this skill among ED providers.