The exploration of heart rate variability (HRV) for clinical purposes in medical sciences as well as for follow-up and monitoring the effects of training in exercise sciences has gained increased popularity in the recent years.

The technique is simple, non invasive, and allows an exploration of Autonomic Nervous System (ANS) activity based on the recording of successive heart beats. Usually measured overnight, over 24h or 5 minutes, the HRV indices can be calculated using time-domain, frequency-domain (Fast Fourier Transform), time-frequency domain (Wavelets) or geometrical as well as Poincaré analysis. These obtained indices reflect global autonomic activity, sympathetic and parasympathetic activity.

HRV indices have proved to be valuable markers of the evolution of the disease or pathological states. Moreover, they have also been associated to good or bad fitness status or performance achievement, which led to the wide use of HRV indices to monitor the effects of training on fitness, fatigue and performance prediction. Indeed, both at a group and at an individual level, HRV indices evolved negatively with heavy cumulated training loads. Inversely, their positive evolution during tapering periods were associated with better performance achievement: the higher the ANS activity, the better the performance. ANS monitoring is thus useful on the short term, on a daily basis, as well as on the long term, on months and years.

Besides, the application of HRV indices in training monitoring opened the exploration of ANS activity in the context of overtraining evolution and thus, prevention. The constant decrease in sympathetic and/or parasympathetic as well as global ANS activity with important cumulated training loads, associated with a lack of reincrease during recovery periods, led to the hypothesis of a possible evolution of overreaching to overtraining through a successive step down of sympathetic and parasympathetic activity. This lower activity of the ANS seems coherent with the numerous clinical syndromes associated with overtraining and could thus be a determinant background of the overtraining state.