Introduction Heart Rate Variability (HRV) at rest is often taken as an indicator of cardiac adaptability. This apparently random change in time between two systoles is indeed chaotically structured, as it is often the case when many regulating processes interact. At rest, parasympathetic, sympathetic or respiratory oscillations seem to be the most implicated. During exercise, HRV dramatically decreases due to the lack of influence of the autonomous nervous system, but is still chaotic. For long duration exercise, heart rate tends to increase, but HRV during stable exercise still remains unknown. In addition, there is no data on the potential effect of blood manipulation on HRV. After RHuEPO administration, a 10% increase in VO2max and maximal power has been reported (Audran et al., 1999). This increase in performance is associated with an increase in hematocrit, hemoglobin and blood viscosity, which are responsible of higher arterial tension and lower heart rate. This may leads to changes in HRV behaviour. The goal of the present study is to assess the evolution of HRV during a 60 min exercise at stable intensity, using both linear and nonlinear indexes and to evaluate the effect of rHuEPO administration on HRV during exercise. Methods 8 high-level cyclists took part into the study. After a maximal test (VO2max-1), they performed a first one-hour exercise at intensity corresponding at 65% of VO2max-1 (rect1). During the subsequent month, subjects were administered injections 3 times a week (either 50 UI/kg of rHuEPO or NaCl) and supplemented in Fe (200 mg/day). Then, all subjects had another maximal test (VO2max-2) and each of them performed a second one-hour exercise at 65% of VO2max-1 (rect2, same absolute intensity as rect1) and third one-hour exercise at 65% of VO2max-2 (rect3, same relative intensity as rect1). ECG was recorded at 1000 Hz using 3 leads high precision holter (Ela medicals). The automatically detected R peaks (public domain routines) were visually checked and each one-hour exercise was divided in 3 windows of 1000 beats to analyse HRV at the beginning, the middle and the end of the effort. The dependant variables were the time interval between 2 heartbeats (RR), the standard deviation of this time (SDRR, which measures the average amount of variability in the record) and the Largest Lyapunov Exponent (LLE, which measures the divergence: when it is positive, the system is chaotic). The statistical analysis was made on the dependent variables (RR, SDRR, LLE) using a 3 way ANOVA: EPO(2) * RECT(3) * WINDOW(3). Results Exercise duration effect: RR and SDRR significantly decreased between the window at the beginning, middle, and end of exercise (P < 0.05); but LLE did not change significantly (p > 0.05). rHuEPO effect: As previously published, the rHuEPO group exhibited an increase in VO2max , hematocrit and hemoglobin (respectively +7%, +9% and +10%) and also a greater lipid oxidation (personal data unpublished). The mean power of the rectangular test increased (rect1 = rect2 = 250.6 ± 6.7; rect3 = 273.9± 8.9, +9%; p < 0.05). One important result of the present analyses is the lack of significant influence of rHuEPO on the HRV dependant variables (p > 0.05 in all cases). RR interval, SDRR and LLE neither change significantly when comparing rect1 and rect3 (same relative intensity) nor when comparing rect1 and rect2 (same absolute intensity). No significant differences were found in the NaCl group. Discussion/Conclusion We found that during a 1 hour constant load exercise, RR interval and SDRR decrease with exercise duration (cardiac trend), but these small changes do not significantly modify the chaotic nature of HRV: the maximum divergence stay similar. This result indicate that the very nature of the regulations that determine HRV did not change during exercise, which in turn is a cue that adaptability of heart rate might be the result of the fractal (or multifractal) nature of the involved regulations. In spite of the numerous physiological changes that determine the observed increase in aerobic performance (e.g., increase in VO2max, hematocrit, blood viscosity, hemoglobin, lipids oxidation...), we found no significant effect of one month rHuEPO administration on HRV during exercise (and at rest). This result was rather surprising as the higher systemic vascular resistance and blood viscosity after rHuepo administration were expected to influence HRV. A potential explanation might be our choice to limit the increase in hematocrit below 50%, which is a physiological amount. Similarly, as the subjects were well-trained cyclists, their adaptability is high and might confuse slight changes in HRV. However, this lack of influence of rHuEPO on HRV might as well result of the combination of positive and negative influences that still remain to unveil.
© Copyright 2004 Book of Abstracts - 9th Annual Congress European College of Sport Science, July 3-6, 2004, Clermont-Ferrand, France. Alle Rechte vorbehalten.