Correlation properties and respiratory frequency of ECG-derived heart rate variability during multiple race-pace running intervals in female and male long-distance runners.

Aim was to evaluate alterations of the non-linear short-term scaling exponent alpha1 of detrended fluctuation analysis (DFAa1) of heart rate (HR) variability (HRV) as a sensitive marker for assessing global physiological demands during multiple running intervals. As a secondary analysis, agreement of ECG-derived respiratory frequency (EDR) compared to respiratory frequency (RF) derived from the metabolic cart was evaluated with the same chest belt device. Fifteen trained female and male long-distance runners completed four running bouts over 5 min on a treadmill at marathon pace. During the last 3 min of each bout gas exchange data and a single-channel ECG for the determination of HR, DFAa1 of HRV, EDR and RF were analyzed. Additionally, blood lactate concentration (BLC) was determined and rating of perceived exertion (RPE) was requested. DFAa1, oxygen consumption, BLC, and RPE showed stable behaviors comparing the running intervals. Only HR (p < 0.001, d = 0.17) and RF (p = 0.012, d = 0.20) indicated slight increases with small effect sizes. In addition, results point towards inter-individual differences in all internal load metrics. The comparison of EDR with RF during running revealed high correlations (r = 0.80, p < 0.001, ICC_3,1 = 0.87) and low mean differences (1.8 ± 4.4 breaths/min), but rather large limits of agreement with 10.4 to -6.8 breaths/min. Results show the necessity of EDR methodology improvement before being used in a wide range of individuals and sports applications. Relationship of DFAa1 to other internal load metrics, including RF, in quasi-steady-state conditions bears the potential for further evaluation of exercise prescription and may enlighten decoupling mechanisms during prolonged exercise bouts.