Low-frequency requirements for recording ischemic ST-segment abnormalities in coronary artery disease.

The objective of this investigation was to determine whether extended low-frequency response is required to record ischemic ST-segment abnormalities in humans. Bipolar electrocardiograms (ECGs) were recorded in 5 men with coronary artery disease using a high-fidelity instrumentation amplifier and FM tape recorder before, during and after erect bicycle exercise. In all patients, ischemic ST-segment abnormalities developed during exercise; 3 patients had angina and 2 remained asymptomatic throughout the test. Using a fast-Fourier transform (FFT) and a variable digitizing rate into a 1,024-point input array, FFT spectra were computed with low-frequency content extending to either 0.20, 0.98 or 1.95 Hz for both a rest and exercise ECG. From these spectra, ECGs were resynthesized using the inverse FFT and compared with the original records. Visual inspection of the original and resynthesized ECGs revealed no obvious differences when low-frequency content extended to 0.20, 0.98 or 1.95 Hz. Numerical comparisons were made by calculating the coefficient of determination (R2) between the original and resynthesized ECGs. The R2 (mean +/- standard deviation) for these comparisons was 0.998 +/- 0.001. It is concluded that the amplitude-response characteristics of electrocardiographic recording equipment do not require extended low-frequency range (such as that found in FM systems) to accurately reproduce ischemic ST-segment abnormalities in humans.