Determining STEMI from ECGs Alone Challenging

Download this article's Factoid (PDF & PPT for Gold Subscribers)

 

The ability to diagnose ischemic ST-elevation in symptomatic patients using electrocardiogram (ECG) alone varies greatly among interventional cardiologists, according to a study published in the October 15, 2011, issue of the American Journal of Cardiology. The finding suggests that triggering a primary PCI protocol based on transmission of ECG results from the field without a clinical context is problematic, the authors say, especially in a population with a high prevalence of abnormal baseline ECGs.

To assess the accuracy of STEMI diagnosis, Yochai Birnbaum, MD, of Baylor College of Medicine (Houston, TX), and colleagues collected records of 240 consecutive patients suspected of having acute STEMI from a database of a large urban medical center in Texas. After confirming ST-elevation in 84 ECGs, the researchers asked 7 experienced interventional cardiologists to review the ECGs to determine if primary PCI protocols should be activated based on this information alone. For patients in whom STEMI was not diagnosed, readers were asked to select at least 1 out of 12 possible causes for why nonischemic ST-elevation was present.

Wide Variation in Diagnosis

True STEMI was confirmed in 48% of patients, while 52% had nonischemic ST-elevation. Of the latter, 30% had cardiac markers suggestive of NSTEMI.

The presenting symptoms, to which the ECG readers were blinded, included chest pain (74%), shortness of breath (12%), and weakness (6%).

With regard to clinical risk factors, 68% of patients had hypertension, 55% dyslipidemia, 36% diabetes, and 32% CAD.

The percentage of ECGs for which readers recommended primary PCI varied widely from 33% to 75%. The readers’ sensitivity in identifying STEMI ranged from 55% to 83% (average 71%) and specificity ranged from 32% to 86% (average 63%). The positive and negative predictive values of readers’ determinations ranged from 52% to 79% (average 66%) and 67% to 79% (average 71%), respectively.

When readers suspected that ST elevation was nonischemic in origin, the cause they chose varied, with left ventricular hypertrophy (6%-31%) and old MI/aneurysm (10%-26%) being the most selected reasons and spontaneous reperfusion (0-5%) the least frequent choice.

Implications for Physicians Greater than for Patients

With the emphasis on shortening door-to-balloon time, Dr. Birnbaum said that physicians can feel the consequences of making split-second decisions more so than patients.

“Physicians are under a lot of pressure to perform,” he told TCTMD in a telephone interview. “When the guidelines are transformed into rigid quality assurance rules and measures, physicians in the emergency department and elsewhere are afraid to miss [a STEMI]. They are under a lot of pressure to make decisions rapidly, and they [may be] overridden.”

On the other hand, to avoid false positives and staff burnout, David A. Cox, MD, of Lehigh Valley Hospital (Allentown, PA), told TCTMD in a telephone interview that a sound clinical decision needs to be made by the emergency room physician, the interventional and the noninvasive cardiologists as to whether or not the primary PCI process should be activated. “We don’t want to break the backs of our cath lab team,” he said, adding that sending nonischemic ST-elevation patients to the cath lab in the daytime is not as much of an issue as it is on weekends and at night.

“We just don’t have a way of identifying [nonischemic] patients very clearly,” he said. “There are these in-between cases and, unfortunately, even expert opinion by interventional cardiologists doesn’t help as much.”

Clinical Information Necessary

Dr. Cox said that “interventional cardiologists are spectacular at reading ECGs,” and the variation seen in this study was based on the lack of clinical information, not expertise. For that reason, unless a STEMI is obvious from the ECG, he believes that the patient should be taken to the emergency room for a clinical examination, even though timing can be crucial.

“We accept that there is a gray zone here,” he said, adding that if a patient has minimal ST-elevation but a truly occluded artery, “We accept that we lose time [by sending them to the emergency room], but their risk is minimal compared to those patients who have marked ST-elevation. That avoids a knee-jerk response of activating the cath lab for anybody with any ST-elevation.”

In a telephone interview with TCTMD, Sorin J. Brener, MD, of Weill Cornell Medical College (New York, NY), said that in reality, there are minimal risks in sending a patient to the cath lab who may not need it. 

“It’s expected that some of the patients we take to the cath lab don’t have a real STEMI, but we would still take them to the cath lab,” he said. “Every procedure carries a risk, but the risks are very small in most patients. Even though you might think that it was left ventricular hypertrophy, you’d still have to consider very strongly taking them to the cath lab because the alternative is letting somebody have a large anterior wall MI that’s not reperfused. While if you’re wrong, you just took 3 pictures and it took 15 minutes.”

Better Risk Stratification Needed

Dr. Birnbaum plans to follow up this paper with a larger-scale, international version of the study to see how a wider array of cardiologists analyze ECGs and if the findings are specific to Texas. “I have experience reading ECGs from all over the world, and the situation at least in Texas is different because we have a lot of patients with abnormal ECGs at baseline,” he said.

Because of the difficulty of accurately differentiating nonischemic ST elevation from STEMI, Dr. Cox said he would like to see future research on “better defining the risk of patients with nonischemic ST-elevation” to aid physicians in being better able to determine if primary PCI should be activated and whether or not the emergency department can be bypassed.

 

---

Source:
Tran V, Huang HD, Diez JG, et al. Differentiating ST-elevation myocardial infarction from nonischemic ST-elevation in patients with chest pain. Am J Cardiol. 2011;108:1096-1101.

 

 

Related Stories:

• Study Highlights Reasons for Delays in STEMI Transfer System
• Adherence to NSTEMI Guidelines Reduces Mortality