• Electrocardiography
• myocardial ischemia
• stress echocardiography

Several methodologies have been developed for the detection of myocardial ischemia. However, only electrocardiography and echocardiography, both at rest and during stress, are non-invasive, inexpensive and readily available in almost every center. While ECG remains a pillar in the diagnosis of ischemia, it has been hypothesized that the filters routinely applied to the raw ECG signal to reduce noise, might also cut out clinically relevant signal and result in a lower sensitivity for myocardial ischemia. In fact, most filers remove low frequencies that contain information on the ventricular repolarizations (ST segment and T wave).
Here we sought to investigate whether analysis of the raw unfiltered ECG signal might improve our ability to diagnose transient myocardial ischemia.
In order to test this, we recorded unfiltered ECGs during percutaneous coronary angioplasty in patients with single coronary artery disease undergoing clinically indicated revascularization. The exact time and duration of intracoronary balloon inflation were recorded. Next, we analyzed the unfiltered raw ECG signal during 30-second long transient coronary occlusions (TCO). The raw ECG signal appears as a fundamental frequency follow by arithmetic harmonics. A significant increase in the signal intensity compared to baseline was found in the fundamental frequency or in one of the first 3 harmonics during every TCO. The increase signal intensity corresponded also to morphological ECG modification that most times met the criteria for the diagnosis of transient myocardial ischemia. We observed a significant correlation between harmonic signal intensity and point J voltage at 60 ms (J60) both at baseline and during TCO.
Once we identified a marker of myocardial ischemia in the harmonic signal during TCO, we tried to apply this knowledge to the detection of inducible myocardial ischemia during a stress test. We recorded raw unfiltered ECG traces in patients undergoing stress echocardiography for clinical reasons. Similar to what observed during coronary angiography, ischemic episodes during stress echocardiography presented with a significant increase in signal intensity compared to baseline. The signal intensity significantly correlated with the J60. This is the proof of principle that harmonic ECG signal can be used to identify myocardial ischemia. Further tests will be necessary to evaluate the sensitivity of this type of analysis compared to morphological ECG.