The standard ECG correlates poorly with LV mass. The SAECG precisely measures myocardial energy and may allow more exact noninvasive assessment of LV mass. A commercially available system (Corazonix Predictor I) was tested for its ability to reproduce and measure known input energy in square wave and QRS waveforms, using frequency bandwidths different from those used for late potential analysis. A test group of 15 patients was studied to determine optimum filter type and bandwidth for comparison of SAECG energy measurement versus LV mass as determined by echocardiography (Penn conversion) and LV hypertrophy via standard ECG criteria. The best means of energy measurement were maximum and total RMS voltage and the integral of the area under the QRS curve. Optimum correlation with echocardiography was seen with a bidirectional band-pass filter of 5 to 250 Hz applied either to the vector sum of the three orthogonal leads or to the Z lead alone (r values 0.61 to 0.73), which was equal to or superior to standard ECG LV hypertrophy determinants. A second group of 20 patients was studied prospectively to confirm these findings, which yielded similar results.
Conclusion: (1) the SAECG when appropriately modified serves as a rapid noninvasive assessment of LV mass. (2) These modifications must examine the entire duration and energy spectrum of the surface ECG and not just the region of late potentials. (3) Very low frequencies (below 5 Hz) must be excluded to eliminate the energy present due to DC offset voltage.