Status of ECG Q-Waves
Abstract & Commentary
By Michael H. Crawford, MD, Professor of Medicine, Division of Cardiology, University of New Mexico, Albuquerque. Dr. Abrams serves on the speaker's bureau for Merck, Pfizer, and Parke-Davis.
Synopsis: The new Q-wave criteria may be too nonspecific, resulting in an inappropriately high number of false-positive results.
Source: Jensen JK, et al. Redefinition of the Q Wave—Is There a Clinical Problem? Am J Cardiol. 2006:97:974-976.
In 2000, the joint european society of cardiology (ESC) and American College of Cardiology (ACC) task force redefined ECG Q-waves indicative of myocardial infarction (MI). Thus, Jensen and colleagues from Denmark evaluated the diagnostic value of the old criteria vs the new criteria as compared to radionuclide myocardial perfusion scanning. From patients referred for coronary artery disease, 79 met the World Health Organization criteria for MI after reviewing their medical record (pain, biomarkers, ECG). Also, 77 control patients were selected who had not had an MI clinically or by perfusion scanning. Abnormal Q-waves were classically defined as ≥ 40 ms or ≥ 25% of the R-wave in 2 contiguous [limb] leads. The new definition is any Q-wave ≥ 30 ms in V1-V3 or any Q-wave ≥ 1 mm in 2 contiguous limb leads or V4-V6. As would be expected, there were several differences between the baseline characteristics of those with and without prior MI. Specifically, the percent with no significant coronary lesions was 17% vs 68%, P < .001. By design none of the controls had a perfusion defect.
Results: As expected, the sensitivity of the new criteria for MI was 71% vs 33% for the old criteria, but specificity decreased to 60% new vs 97% old. Thus, the positive predictive value of a Q-wave by the new criteria was only 64% vs 93% by the old. Negative predictive value was a little better by the new criteria 67% vs 59%. Jensen et al concluded that the new Q-wave criteria are nonspecific and will result in frequent false positive diagnoses.
This joint decision by the ESC and ACC is laudable for uniting world opinion and for clarifying the variable applications of the older Minnesota code. However, it may leave us with more false positive diagnoses to evaluate at some increased cost to the health care system. In this study, false positives went from 7% to 36%. This seems to be part of some recent movement toward more sensitive tests that has been spearheaded by radiologists and emergency department physicians. These specialties prefer high sensitivity and very low false negative rates to reduce missed diagnoses and potential liability. False positives are not their problem, as these patients are passed on to other physicians to sort out. Hence, the unbridled enthusiasm for troponin and BNP. Such tests leave those of us who take care of patients after the presumptive diagnosis is made with a lot of extra nonproductive work, anxious patients, and clogged systems. Also, if cardiac catheterization is done, the patient is subjected to unnecessary risks.
Some might argue what's wrong with very sensitive tests for prior MI since this is an important diagnosis? Why not use the most sensitive tests? Missing an MI is worse than a false-positive diagnosis. The clinical scenario assessed in this study was chest pain patients. Currently, < 15% of patients presenting to the hospital with chest pain have acute coronary syndromes; so more specific criteria seem appropriate. Also, modern non-cardiac surgery carries a very low risk of acute cardiovascular events, and extensive diagnostic testing has not been shown to reduce these low event rates. Anything that increases the perceived need for more tests is not going to help the situation. In addition, it is naïve to think that the snapshot of any one test is all there is to diagnosis. Many other factors come into play, and the progress of time often clarifies diagnostic dilemmas. Long-term patient management physicians have always been comfortable with test sensitivities and specificities in the 80% range. We don't always need a 99% sensitivity and a 60% specificity. Perhaps the new Q-wave criteria would make sense if you are evaluating an airline pilot, but in everyday practice I see trouble.
There are some limitations to this study. It is small, retrospective, and biased toward chest pain patients likely to have coronary artery disease. However, I cannot imagine less false-positives in a more unselected population. Also, myocardial perfusion imaging as the MI gold standard is problematic since there are false-positives and negatives with imaging. Clearly, more data are needed, but I don't think the basic outcome will change. The new criteria are more sensitive by design and, consequently, will be less specific. Use them at your own risk.