Evidence for an intrinsic mechanism regulating heart rate variability in the transplanted and the intact heart during submaximal dynamic exercise?

STUDY OBJECTIVE: The aim was to assess the changes in sympatho-vagal balance which occur with exercise. DESIGN: The power spectrum of RR interval fluctuations (low frequency LF and high frequency components HF) was determined before, during, and after graded work load exercise on a cycle ergometer. The power spectrum of the respiratory signal, oxygen consumption, and respiratory volumes were also evaluated. In all subjects HF was considered to be an index of respiratory sinus arrhythmia. In normal subjects HF and LF were considered to be indices of relative vagal and sympathetic activity, respectively, whereas in heart transplant subjects HF was considered as a respiratory modulation of the intrinsic heart rate, and not dependent on autonomic tone. Heart rate variability was evaluated as RR interval variance. SUBJECTS: 15 normal subjects (six trained cyclists and nine healthy sedentary subjects) and six orthotopic heart transplant recipients took part in the study. MEASUREMENTS AND MAIN RESULTS: During the first part of exercise, heart rate increased, RR interval variance decreased, HF decreased, and the relative amount of LF increased both in sedentary and athletic subjects, suggesting a relative increase in sympathetic tone. However, when approaching peak exercise, while heart rate further increased and the variance slightly decreased, the relative proportion of LF decreased and HF proportionally increased. At peak exercise HF accounted for 99.9% of heart rate variability in athletic subjects and for 88.9% in sedentary subjects (p less than 0.001 v baseline and v LF in both groups). In heart transplant subjects both the variance and the HF increased from the beginning of exercise (p less than 0.05), and showed a direct correlation with ventilatory variables and an inverse correlation with heart rate (r = 0.794, p less than 0.001, multiple regression analysis). No measurable LF components could be obtained in these subjects. During recovery, while the heart rate decreased and the RR interval variance increased, there was a relative increase in LF and a relative decrease in HF in normal subjects (either sedentary or athletic). Similarly, in heart transplant subjects, there was a decrease in HF during recovery. Thus the increase in HF at peak exercise in normal subjects contrasts with all the other data which suggest a prevalence in sympathetic tone during the entire exercise and the early recovery period, but appears similar to the increase in HF observed in heart transplant subjects due to the effect of increased ventilation during exercise. CONCLUSIONS: These findings suggest that at peak exercise a non-autonomic mechanism, possibly intrinsic to the heart muscle, may determine heart rate fluctuations in synchrony with ventilation in the intact as well as in the denervated human heart.