Study objective: Accurate diagnosis in emergency department patients with possible myocardial ischemia is problematic. Two-dimensional echocardiography has a high sensitivity for identifying patients with myocardial infarction (MI); however, few studies have investigated its diagnostic ability when used acutely in ED patients with possible myocardial ischemia. Therefore we investigated the ability of ED echocardiography for predicting cardiac events in patients with possible myocardial ischemia.
Methods: Echocardiography was performed within 4 hours of ED presentation in 260 patients with possible myocardial ischemia, and was considered positive if there were segmental wall motion abnormalities or the ejection fraction was less than 40%. ECGs were considered abnormal if there was an ST-segment elevation or depression of greater than or equal to 1 mm, or ischemic T-wave inversion. Cardiac events included MI and revascularization.
Results: Of the 260 patients studied, 45 had cardiac events (23 MI, 19 percutaneous transluminal angioplasty, 3 coronary bypass surgery). The sensitivity of echocardiography for predicting cardiac events was 91% (95% confidence interval 79% to 97%]), which was significantly higher than the ECG (40% [95% CI 27% to 55%]: P < .0001), although specificity was lower (75% [95% CI 69% to 81%] versus 94% [95% CI 90% to 97%]; P < .001). Addition of the echocardiography results to baseline clinical variables and the ECG added significant incremental diagnostic value (P < .001). With use of multivariate analysis, only male gender (P < .03, odds ratio [OR] 2.4 [1.1 to 5.3]), and a positive echocardiographic finding (P < .0001, OR 24 [9 to 65]) predicted cardiac events. Excluding patients with abnormal ECGs (N = 30) did not affect sensitivity (85%) or specificity (74%) of echocardiography.
Conclusion: Echocardiography performed in ED patients with possible myocardial ischemia identifies those who will have cardiac events, is more sensitive than the ECG, and has significant incremental value when added to baseline clinical variables and the ECG.