Incorporating Fragmented QRS on Surface Electrocardiogram to Exercise Stress Test

Dear Editor,

We read with interest the study by Tusun et al. recently published in the journal.1 Enhancing the sensitivity/specificity of exercise stress test (EST) is much desired as it is the major drawback of this method. However, there are a few issues that deserve attention. Researchers had a low threshold for coronary angiography (CAG) in this study that is reflected in the high proportion of patients (46.3%) without significant coronary artery disease (CAD). As described in the methods, patients with acute coronary syndromes including unstable angina were excluded, so the study population consisted of patients with suspicious stable ischemic heart disease. All patients with positive EST that was defined as horizontal or downsloping exercise‐induced ST segment depression of more than 1.0 mm at 0.08 seconds after the J‐point were referred to CAG. It is not clear whether the indication of CAG was based on cardiologist's discretion or on the study protocol. Guidelines do not recommend CAG for patients who are found to be at low risk in noninvasive risk stratification including those with treadmill score ≥5 while achieving maximal exercise level.2 Reporting Duke Treadmill Score of patients would have been of value in interpreting the result.

Fragmented QRS has been an indicator of left ventricular hypertrophy in hypertensive patients with normal coronary arteries.3 An unexpected finding in this study is that age, hypertension, and diabetes that are known risk factors for CAD were not independent predictor of CAD in multiple logistic regression analysis, instead fragmentation of QRS was found to be the predictor of CAD. One would assume that the low population study could have contributed to this unanticipated finding. Moreover, diabetes, hypertension, and age are well‐recognized risk factors of CAD. On the other hand, fragmented QRS is probably merely a reflection of myocardial scar and certainly not a risk factor for CAD and mixing them may not be the best way to perform multivariate analysis.

Another unexpected finding is the high number of patients with fragmentation (49.5% of total and 85.1% of those with significant CAD). Fragmented QRS has been reported in 19.7% of 10,904 subjects, 2543 of whom had evidence of cardiac disease.4 Das et al. report a frequency of 27.35% in patients for stress test.5 Fragmentation has been reported in 53% of patients 2 months after first Q‐wave myocardial infarction,6 and in 51% of patients with myocardial infarction.7 Since it is thought to be a marker of myocardial scarring, the high frequency of fragmentation in a group of patients without a history of myocardial infarction needs to be explained.