CIRCADIAN AND POWER SPECTRAL CHANGES OF RR AND QT INTERVALS DURING TREATMENT OF PATIENTS WITH ANGINA-PECTORIS WITH NADOLOL PROVIDING EVIDENCE FOR DIFFERENTIAL AUTONOMIC MODULATION OF HEART-RATE AND VENTRICULARREPOLARIZATION

This study evaluates the effects of autonomic manipulation by chronic beta blockade with nadolol on the circadian and power spectral changes of heart rate and QT interval. It was hypothesized that differential innervation of the atria and ventricles by sympathetic and parasympath etic fibers may produce differential effects an heart rate and QT inte rval variabilities. Halter recordings of 12 male patients (age 63 +/- 7 years) with stable angina were analyzed before and after 3 weeks of nadolol therapy. The QT intervals were individually normalized by an e xponential for mula to study the circadian variation of QTc. Power spe ctra of RR and uncorrected QT intervals were obtained by fast-Fourier analysis from 256 consecutive sinus beats during the day at maximal he art rate and during the night at minimal heart rate. Frequency-specifi c variability was determined from areas under the spectral plots. Both heart rate and QTc exhibited significant circadian patterns (p < 0.01 ) in opposite phase with each other. Mean heart rate was significantly reduced with nadolol (81 +/- 12 vs 67 +/- 12 beats/min, p < 0.001), w ith greater reduction during daytime. The mean QTc was unexpectedly re duced after nadolol treatment, with borderline significance (p = 0.06) . The RR variability in the frequency range of 0.05 to 0.25 Hz was sig nificantly increased with nadolol at 3:00 A.M. (p < 0.01) but not at 1 :00 P.M. The BT variability in the same frequency range was not signif icantly increased with nadolol. The power spectra of RR and QT interva ls were dissimilar except at the lower frequencies around 0.05 Hz. Con sistent with this observation, the QT versus RR plots of 256 beat-by-b eat values showed poor correlation in all subjects. It is concluded th at: (1) the RR power spectra from Holter recordings predominantly refl ect vagal activity in the 0.05 to 0.25 Hz range; (2) the QT interval v ariability is rela tively insensitive to vagal activity; (3) QTc reduc tion with nadolol may reflect reduced dispersion of ventricular repola rization in the present patient group; and (4) the relation between RR and QT is frequency-dependent, and is effectively masked at frequenci es > 0.05 Hz.