Heart transplant rejection detected by signal averaged QRS analysis. Preliminary results.

Detailed QRS wave analysis from the limb leads of the surface high resolution electrocardiograms recorded in five cyclosporine-treated heart transplant recipients gave good correlation with the corresponding results of endomyocardial biopsy. That is when the result of the biopsies has identified the presence of rejection, a concomitant (p = 0.001) variation in some parameters of the QRS has been observed. The reproducibility of such parameters (established by means of the correlation coefficient r) gave r values ranging from 0.6 up to 0.93. The correlation regarded parameters analyzed both in time and frequency domain. Biopsy results were divided into two classes on the basis of the presence or absence of rejection. The most significant parameters obtained from high-frequency (25-300 Hz) ECGs within a few hours of each biopsy were: a) the total high-frequency voltage amplitude Vt of the QRS and the voltage amplitude of its initial Vi and middle thirds Vm; b) the peak voltage amplitude Vp of the QRS; c) the QRS duration L; d) the integrated voltage time product I of the QRS; e) three mean voltage amplitudes V5, V6, V7, of the power spectral density constructed on the basis of the 512-point fast Fourier transform applied on each recording. The above mentioned parameters appear to be useful in predicting the biopsy result in terms of the presence or absence of rejection.