Dynamics of low-frequency R-R interval oscillations preceding spontaneous ventricular tachycardia

Abstract

Background Increased sympathetic activity is believed to be an important trigger of sustained ventricular tachyarrhythmias (VT) and is believed to be responsible for the increased heart rate that we and others have reported before the onset of spontaneous VT. However, in the patients reported herein, heart rate variability (HRV) indexes that reflect sympathetic activity unexpectedly declined, whereas heart rate increased. To explain this apparent paradoxic behavior, we tested the hypothesis that baseline levels of HRV determine its reaction to short-term autonomic perturbations before the onset of VT. Methods and Results Holter electrocardiograms from 47 patients (ejection fraction 36% ± 15%) with recorded VT were analyzed. Frequency domain HRV indexes (low-frequency power [LFP] 0.04 to 0.15 Hz, high-frequency power [HFP] 0.15 to 0.4 Hz, and total power [TP] 0.01 to 0.4 Hz) were studied in 5-minute intervals and over a period of 24 hours. Patients were divided into those with a decrease in LFP in the 2-hour period before VT (group A, n = 32) and those with an increase or no change (group B, n = 15). The data were logarithmically transformed. Heart rate increased 15 minutes before the onset of VT compared with the 24-hour mean in both groups (group A: 80.3 ± 15.4 to 86.1 ± 20.0 beats/min, P = .005; group B: 80.6 ± 13.5 to 86.7 ± 14.0 beats/min, P = .017). Group A had higher TP, LFP, and LFP/HFP 2 hours before VT, and these variables decreased 15 minutes before the onset of VT (TP from 7.31 ± 1.28 to 6.88 ± 1.35, LFP from 6.09 ± 1.28 to 5.38 ± 1.33, LFP/HFP from 1.33 ± 0.89 to 0.96 ± 0.80, P < .001 for all 3 variables). HFP also decreased 15 minutes before VT compared with 2 hours (from 4.78 ± 1.05 to 4.49 ± 1.24, P = .028). In group B, which had lower baseline TP, LFP, and LFP/HFP at 2 hours before VT, these variables increased 15 minutes before the event (TP from 6.41 ± 1.41 to 6.86 ± 1.42, P = .004; LFP from 4.59 ± 1.51 to 4.95 ± 0.62, P < .001; LFP/HFP from 0.22 ± 1.22 to 0.52 ± 1.38, P = .10), whereas HFP did not change significantly (4.40 ± 0.94 and 4.53 ± 1.01, P = .50). Conclusions An increase in heart rate and a drop in the low-frequency oscillations of R-R intervals before the onset of VT occurred in patients with higher baseline level of oscillatory activity. These changes suggest a dissociation between the average and rhythmic modulation of R-R intervals. A decline of the low-frequency oscillations in the setting of increasing heart rate could reflect an abnormal response to increased sympathetic activity in most of the patients from the studied group. The different behaviors of the HRV indexes before the onset of VT in the 2 groups suggest that change in the dynamics of R-R intervals, rather than the direction of change, facilitates arrhythmogenesis. (Am Heart J 2000;139:126-33.)