Точність визначення тривалості кардіоінтервалів при застосуванні програмно-апаратного комплексу «Ритм» в умовах виконання фізичних навантажень

Abstract

The important task for modern physiology is remote monitoring of the functions of physiological systems of the human organism during the competitive and training activity. It is well known that analysis of heart rate variability is one of the effective methods to evaluate the physiological changes which occur in the response to physical loads. In order to perform the correct analysis of heart rate variability by newly designed devices, it is necessary to confirm the sufficient level of accuracy in the registration of RR intervals. The purpose of our research was to analyze the accuracy of RR time series measurements by software-hardware complex “Rytm” and validity of subsequently calculated heart rate variability indexes in conditions of exercise performance. Material and methods. The study involved 20 healthy male adults 20-21 years old. Recording of cardio intervals was performed simultaneously with «Polar RS800», and software-hardware complex “Rytm”. The subjects performed a step test in a rate of 20 steps per minute, platform height – 40 cm, duration – 2 minutes. Results and discussion. Heart rate variability indexes were calculated by Kubios HRV 2.1 software. The totally 4707 pairs of RR intervals were analyzed. The average bias between the RR interval, registered by software-hardware complex «Rytm» and «Polar RS800», was only 0.06 s. We revealed the narrow Bland–Altman limits of agreement (3.72 − -3.83 ms) and the highest value of the intraclass correlation coefficient (1.000) between the data of these two devices. The Bland–Altman plot confirmed good agreement between the devices in the measurements of RR intervals. At the same time, the significant difference (p = 0.002) of the two data sets was found according to paired Wilcoxon test. As the final goal of the registration of RR time series is calculation of individual heart rate variability indexes, we intended to test the presence of substantial differences in the heart rate variability indexes, derived from the data from two devices − «Polar RS800» and software-hardware complex «Rytm». We compared the results of time-domain (HR, STD RR, RMSSD, pNN50), frequency-domain (VLF, LF, HF, LF / HF) and nonlinear (RR tri index, SD1, SD2) analysis of heart rate variability. It was found that only for the LF/HF ratio a statistically significant difference was present. Conclusion. The results suggest the good agreement between most of the heart rate variability indexes based on data of software-hardware complex «Rytm» and well approved heart rate monitoring systems («Polar RS800»)