Title: Does This Patient With Chest Pain Have Acute Coronary Syndrome?
Abstract: Take-Home MessageThe accuracy of individual risk factors, symptoms, and signs in isolation to diagnose acute coronary syndrome is poor. The history, ECG, age, risk factors, and troponin score (HEART) and the Thrombolysis in Myocardial Infarction (TIMI) score have strong diagnostic value and should be used as part of the clinical assessment of these patients.MethodsData SourcesElectronic English-language searches included MEDLINE and EMBASE from January 1, 1995, to July 31, 2015. No search for unpublished studies or other gray literature was performed.Study SelectionProspective studies selected examined diagnostic test accuracy among patients admitted to the emergency department (ED) with a clinical picture suggesting acute coronary syndrome. Excluded studies were those that tested decision aids, those without an adequate acute coronary syndrome endpoint, accelerated diagnostic protocols requiring serial troponin levels and ECG, high-sensitivity troponin, and preselected populations according to risk.Data Extraction and SynthesisThe rational clinical examination level of evidence and the Quality Assessment of Diagnostic Accuracy Studies criteria were evaluated; low-quality studies were excluded (ie, rational clinical examination quality levels 3 to 5). Heterogeneity was assessed for the likelihood ratios (LRs). Publication bias was assessed and none was detected. For meta-analysis of risk scores, studies were included only if they provided data for each risk score stratum. The accuracy of individual risk factors, symptoms, and signs in isolation to diagnose acute coronary syndrome is poor. The history, ECG, age, risk factors, and troponin score (HEART) and the Thrombolysis in Myocardial Infarction (TIMI) score have strong diagnostic value and should be used as part of the clinical assessment of these patients. Electronic English-language searches included MEDLINE and EMBASE from January 1, 1995, to July 31, 2015. No search for unpublished studies or other gray literature was performed. Prospective studies selected examined diagnostic test accuracy among patients admitted to the emergency department (ED) with a clinical picture suggesting acute coronary syndrome. Excluded studies were those that tested decision aids, those without an adequate acute coronary syndrome endpoint, accelerated diagnostic protocols requiring serial troponin levels and ECG, high-sensitivity troponin, and preselected populations according to risk. The rational clinical examination level of evidence and the Quality Assessment of Diagnostic Accuracy Studies criteria were evaluated; low-quality studies were excluded (ie, rational clinical examination quality levels 3 to 5). Heterogeneity was assessed for the likelihood ratios (LRs). Publication bias was assessed and none was detected. For meta-analysis of risk scores, studies were included only if they provided data for each risk score stratum. Table 1Performance of select aspects of clinical diagnosis tools for acute coronary syndrome.Clinical Decision ToolsThresholdLR (95% CI)∗Summary LR from studies that report original data at each threshold without combining across clinical decision rule thresholds.I2Risk of BiasHigh riskHEART score7–1013 (7.0–24)89LowTIMI score5–76.8 (5.2–8.9)56LowHFA/CSANZ ruleHigh risk2.8 (2.6–3.0)0LowLow riskHEART score0–30.20 (0.13–0.30)78LowTIMI score0–10.31 (0.23–0.43)96LowHFA/CSANZ ruleLow to intermediate risk0.24 (0.19–0.31)10LowCI, Confidence interval; HFA/CSANZ, Heart Foundation of Australia/Cardiac Society of Australia and New Zealand.∗ Summary LR from studies that report original data at each threshold without combining across clinical decision rule thresholds. Open table in a new tab Table 2Performance of select cardiac risk factors, chest pain characteristics, physical examination elements, and ECG findings in diagnosing acute coronary syndrome.∗Ischemic ECG defined as any T-wave inversion, ST depression, or Q waves. Fifty-eight articles from the 2,992 that were screened met inclusion criteria. The overall acute coronary syndrome event rate ranged from 5% to 42% (median 14%; interquartile range 10% to 20%). Substantial heterogeneity was identified for some of the decision tools (I2>50%). Furthermore, many studies were confounded by incorporation and verification biases. Moreover, data pooling was limited because of the wide variation of variables tested among included studies with the intention to reduce bias from any single study (minimum 560 patients per variable tested).TestNo. of StudiesNo. of PatientsRisk of BiasLR+ (95% CI)LR– (95% CI)Cardiac risk factorsPeripheral arterial disease36,034Low2.7 (1.5–4.8)0.96 (0.94–0.98)Chest pain characteristicsPain similar to previous ischemia12,718Low2.2 (2.0–2.6)0.67 (0.60–0.74)Change in pattern during previous 24 h12,718Low2.0 (1.6–2.5)0.84 (0.79–0.90)Physical examination elementsPain reproduced by palpation1839Low0.28 (0.14–0.54)1.2 (1.0–1.2)Performance of ECG (all interpreted by emergency physicians)Ischemic ECG716,559Low3.6 (1.6–5.7)0.74 (0.68–0.81)∗ Ischemic ECG defined as any T-wave inversion, ST depression, or Q waves. Fifty-eight articles from the 2,992 that were screened met inclusion criteria. The overall acute coronary syndrome event rate ranged from 5% to 42% (median 14%; interquartile range 10% to 20%). Substantial heterogeneity was identified for some of the decision tools (I2>50%). Furthermore, many studies were confounded by incorporation and verification biases. Moreover, data pooling was limited because of the wide variation of variables tested among included studies with the intention to reduce bias from any single study (minimum 560 patients per variable tested). Open table in a new tab CI, Confidence interval; HFA/CSANZ, Heart Foundation of Australia/Cardiac Society of Australia and New Zealand. Coronary heart disease caused 1 of every 7 deaths in the United States in 2011.1Mozaffarian D. Benjamin E.J. Go A.S. et al.Heart disease and stroke statistics—2015 update: a report from the American Heart Association.Circulation. 2015; 131: e29-e322Crossref PubMed Scopus (5711) Google Scholar Furthermore, in 2010, there were more than 1.1 million hospitalizations for acute coronary syndrome in the United States.1Mozaffarian D. Benjamin E.J. Go A.S. et al.Heart disease and stroke statistics—2015 update: a report from the American Heart Association.Circulation. 2015; 131: e29-e322Crossref PubMed Scopus (5711) Google Scholar However, a minority of patients with acute coronary syndrome have ST-segment elevation myocardial infarction (STEMI); the remainder of these patients have non-STEMI or unstable angina.2Amsterdam E.A. Kirk J.D. Bluemke D.A. et al.Testing of low-risk patients presenting to the emergency department with chest pain: a scientific statement from the American Heart Association.Circulation. 2010; 122: 1756-1776Crossref PubMed Scopus (470) Google Scholar Because the ECG can exclude STEMI, the key distinction that must be made in the majority of patients is differentiating between acute coronary syndrome and noncardiac chest pain. The focus of this systematic review was the initial examination of the undifferentiated patient with potential acute coronary syndrome. This review demonstrated that in isolation, the accuracy of individual risk factors, symptoms, and signs in diagnosing acute coronary syndrome was generally poor. The absence of risk factors did not necessarily exclude acute coronary syndrome, although a history of peripheral arterial disease (positive LR [LR+]=2.7) did support an acute coronary syndrome diagnosis. Response to nitroglycerin or pleuritic features had no discriminative value for the diagnosis of acute coronary syndrome. However, pain reproduction with palpation decreases the likelihood of acute coronary syndrome (LR+=0.28). The items with the largest positive LRs for predicting acute coronary syndrome in chest pain patients were ST depression (LR+=5.3), ischemic ECG (LR+=3.6), TIMI score of 5 to 7 (LR+=6.8), and HEART score of 7 to 10 (LR+=13). In addition, a HEART score of 0 to 3 (LR=0.20) and a TIMI score of 0 to 1 (LR=0.31) decreased the likelihood of acute coronary syndrome.3Antman E.M. Cohen M. Bernink P.J. et al.The TIMI risk score for unstable angina/non-ST elevation myocardial infarction: a method for prognostication and therapeutic decision making.JAMA. 2000; 284: 835-842Crossref PubMed Scopus (2563) Google Scholar, 4Six A.J. Backus B.E. Kelder J.C. Chest pain in the emergency room: value of the HEART score.Neth Heart J. 2008; 16: 191-196Crossref PubMed Scopus (350) Google Scholar, 5Acute Coronary Syndrome Guidelines Working GroupGuidelines for the management of acute coronary syndromes 2006.Med J Aust. 2006; 184: S9-S29PubMed Google Scholar Studies using high-sensitivity troponins were excluded because they are not available in the United States and have not been adequately evaluated as components of the most commonly used risk scores. This systematic review indicates that clinical features alone are not discriminatory for ED chest pain patients with suspected acute coronary syndrome; initial or serial troponin levels and ECG in the context of symptoms suggestive of an underlying ischemic cause should be part of the initial assessment. Risk scores such as TIMI and HEART appear to be useful tools to help discriminate acute coronary syndrome from noncardiac chest pain. These findings are consistent with the 2014 non-ST-elevation–acute coronary syndrome guidelines for managing non-STEMI patients with possible acute coronary syndrome.6Amsterdam E.A. Wenger N.K. Brindis R.G. et al.2014 AHA/ACC guidelines for the management of patients with non-ST-elevation acute coronary syndromes: executive summary.Circulation. 2014; 130: 2354-2398Crossref PubMed Scopus (723) Google Scholar