RESUMO
Compared to other non-invasive methods, the conventional 12-lead electrocardiogram (ECG) has low sensitivity and specificity for identifying coronary artery disease (CAD). We compared the newly developed high-frequency QRS electrocardiogram (HFQRS ECG, 150-250 Hz) with adenosine sestamibi myocardial perfusion study (MPI)-the most sensitive non-invasive study in cardiology practice. Using advanced 12-lead computer-based ECG software recently developed at NASA, criteria for a positive 12-lead HFQRS-ECG test for obstructive CAD were developed using 300 signal-averaged beats from patients undergoing elective coronary angiograms for evaluation of chest pain. These criteria, which rely strictly upon the presence or the absence of morphologic 'reduced amplitude zones' (RAZs) and not upon the 'root mean squared' (RMS) voltage amplitudes of the HFQRS complexes, were then applied prospectively to 18 patients undergoing MPI. Active CAD was considered present when reversible ischemic defects were present on MPI. Of the 18 patients, 9 had reversible defects on MPI (positive scan), whereas the other 9 had no reversible perfusion defects (negative scan). Patients with a positive nuclear study went on to coronary angiography confirming CAD, except in one patient who had nonobstructive coronary disease (<50% stenosis). Eight of the 18 subjects therefore had active CAD, whereas 10 were judged not to have active CAD. The 12-lead HFQRS-ECG result was consistent with nuclear scan results in 14 of 18 patients. The HFQRS-ECG and nuclear results differed in: (1) one patient who had a low score positive MPI, negative HFQRS-ECG and normal coronary angiogram; (2) one patient who had a small reversible anterior wall perfusion defect, 60% LAD lesion on angiogram but a negative HF-QRS result; and (3) two individuals who had positive HFQRS-ECG results in the face of negative nuclear scans but who did not undergo angiography. 12-lead HFQRS ECG had excellent sensitivity (87.5% based on 7/8 true positives correctly identified) and specificity (no worse than 80%, >8/10 true negatives correctly identified) for identifying CAD. 12-lead HFQRS ECG is an easily performed, inexpensive and potentially widely available technique that utilizes the same leads and electrodes as the conventional 12-lead ECG. It had accuracy comparable to MPI in this study. Resting 12-lead HFQRS ECG appears to be a very promising non-invasive technique for identifying CAD and may represent a viable alternative to many of the more expensive and time-consuming techniques presently utilized for non-invasively identifying CAD.