Discriminating between two autonomic profiles related to posture in Olympic athletes

Autonomic assessment might be useful in training management. We planned to assess whether oscillatory metrics of RR variability (such as LFnu) would be more efficient than static indices from low order statistics (RR variance) at discriminating laying rest from stand posture, as an analog of a shift to sympathetic dominance.

We studied a large population of elite Olympic athletes: a total of 406 athletes (162 females and 244 males, of similar age 21.7 and 24.4 years) participating to the selection for the upcoming 2016 Olympic games. We employed various methods to extract autonomic indices from RR variability and employed a stepwise statistical approach combining factor and discriminant analysis.

We observed that that relative power of oscillatory components from spectral analysis of RR variability (such as LF or HF in nu) and indices from symbolic analysis (particularly 0V) clearly outperform RR variance in discriminating between two physiological conditions (laying rest and stand) related to posture and autonomic activation.

In world class Olympic athletes we have shown that a small subset of RR variability indices, related to sympathovagal balance, may be more appropriate than RR variance to assess excitatory sympathetic autonomic responsiveness of the SA node. These findings may have practical implications for the use of RR variability in guiding training and predicting success in competitions.