High resolution electrocardiographic abnormalities as possible indexes of electric instability of the myocardium.

OBJECTIVE

To evaluate the signal-averaged high resolution electrocardiogram in the time domain and the frequency domain in various cardiac patients.

DESIGN

Patients in seven clinical diagnostic categories were compared with a healthy control group.

PATIENTS AND

METHODS

The control group consisted of 70 clinically healthy persons. The 525 cardiac patients were divided as follows: group 1, 29 patients with surgically corrected mitralvalvular disease; group 2, 42 postinfarction patients with-bypass grafting; group 3, 57 patients with ventricular tachycardia; group 4, 198 patients with medically treated angina; group 5, 63 patients with hypertension; group 6, 46 patients with paroxysmal atrial fibrillation; and group 7, 90 patients with acute pulmonary insufficiency. The Frank orthogonal leads were used to derive the signal-averaged electrocardiogram in the usual manner by averaging over 200 complexes. Time domain and frequency domain data were recorded for both the P wave and the QRS complex.

RESULTS

Ventricular late potentials were found most often in patients with postinfarction ventricular tachycardia (59%). The prevalence of ventricular late potentials in the mitral valvular disease group increased from 45% to 70% after surgery; in the bypass grafting group it also increased. Spectral temporal mapping revealed a substantial decrease of the total spectral power in the bypass grafting group, to 129 +/- 19 microV2/Hz compared with the preoperative period (205 +/- 16 microV2/Hz). The high frequency power decreased, so the ratio of low frequency to high frequency power increased. A significant correlation was found between the root-mean-square signal in the last 40 ms of the QRS complex and each of the amplitude parameter, the frequency domain localization and time domain localization of the power density peaks.

CONCLUSIONS

High resolution electrocardiography including time domain and frequency domain analyses may prove helpful in diagnosis and management.