HIGH-FREQUENCY ANALYSIS OF THE SIGNAL-AVERAGED ECG - CORRELATION WITH LEFT-VENTRICULAR MASS IN RABBITS

Standard electrocardiographic (ECG) criteria have exhibited poor correlation with left ventricular mass and poor sensitivity for left ventricular hypertrophy at acceptable levels of specificity. To assess the ability of the high-frequency filtered signal-averaged ECG to improve ECG correlation with left ventricular mass, signal-averaged orthogonal lead recordings in 29 normal rabbits and seven rabbits with left ventricular hypertrophy due to chronic aortic regurgitation were compared with left ventricular mass corrected for body weight. Voltage of the vector QRS complex was integrated over the total duration of the QRS after separate filtering with standard frequency (0-100 Hz) low-pass and high-frequency (44 Hz) high-pass filters. Measurement of individual X, Y, and Z lead R and S wave voltage was performed on averaged, standard frequency filtered complexes, and the maximal spatial vector magnitude was determined from the standard frequency filtered vectors. Voltage of the 44 Hz high-pass filtered vector QRS complex integrated over the total duration of the QRS (high-frequency vector integral) correlated closely with indexed left ventricular mass (r = 0.84, p < 0.0001), significantly better than the correlation of standard frequency vector integral or maximal spatial vector magnitude voltages (r = 0.35 and r = 0.61, each p < 0.01 vs high-frequency vector integral) and the correlation of orthogonal lead X R wave or lead Y S wave voltages (r = 0.55 and r = 0.37, respectively, each p < 0.01 vs high-frequency vector integral). Receiver operating characteristic curve analysis demonstrated that overall performance of the high-frequency vector integral for the detection of left ventricular hypertrophy was significantly better than performance of the maximal spatial vector magnitude, the best standard frequency criteria (area 0.95 vs 0.84, p < 0.05). These findings suggest that use of the time-voltage integral of the high-frequency signal-averaged vector QRS in rabbits can improve the correlation of ECG findings with left ventricular mass and can increase the accuracy of ECG methods for the detection of left ventricular hypertrophy.