Quantitative analysis of the high-frequency components of the terminal portion of the body surface QRS in normal subjects and in patients with ventricular tachycardia.

Quantitative analysis of the high-frequency components of the terminal portion of the surface QRS was performed in 42 normal subjects (group 1, ages 18-67 years, mean +/- SEM 34.7 +/- 2.2 years) and in 12 patients with symptomatic, sustained ventricular tachycardia (VT) (group 2, ages 48-76 years, mean 59 +/- 2.3 years). Signal averaging and high-pass, bidirectional digital filtering were used for analysis. The total duration of the QRS, the duration of the low-amplitude signals (less than 40 microV) in the terminal portion of the QRS and the amplitude of the signals in the last 40 and 50 msec of the QRS were measured at filter settings of 25 and 40 Hz. Reproducibility of the measurements was tested in 15 normal subjects by comparing results obtained from two consecutive recordings. Significant differences were found between normal subjects and VT patients for all four indexes at both 25- and 40-Hz filters. Specific values for each of the indexes were identified at the 40-Hz filtering, which could separate normal subjects from VT patients (20 microV for the amplitude of last 40 msec; 30 microV for the amplitude of last 50 msec; 120 msec for the total duration; and 39 msec for the low-amplitude signal of the filtered QRS). Using these values for the four indexes, respectively, 90%, 98%, 100% and 90% of the normal subjects and 83%, 83%, 58% and 83% for the VT group were correctly classified. The results show that the high-frequency analysis of the signal-averaged body surface QRS is a reliable, reproducible, noninvasive method for distinguishing patients with VT from normal subjects.