RR interval time series from healthy subjects

Abstract

Heart rate variability (HRV) is mainly controlled by the autonomic nervous system (ANS) interacting with receptors on the sinoatrial node (SAN). HRV is influenced by many several factors such as chemical, hormonal and neural modulations, circadian changes, exercise, emotions, posture, and preload. The heart rate adaptation to changing factors is carried out by the activity of a variety of regulatory subsystems and results in a complex linear and non-linear time behaviour, which changes with age, gender and pathological conditions. Many studies explore an age range of 1-99 years. This study focused on the development of a database of long-term time series of healthy volunteers aged between 1 month and 55 years.

Background

Heart rate variability (HRV) is mainly controlled by the autonomic nervous system (ANS) interacting with receptors on the sinoatrial node (SAN). HRV is influenced by many several factors such as chemical, hormonal and neural modulations, circadian changes, exercise, emotions, posture, and preload. The heart rate adaptation to changing factors is carried out by the activity of a variety of regulatory subsystems and results in a complex linear and non-linear time behaviour, which changes with age, gender and pathological conditions.

The time series contained in this database were used to perform studies on the non-linear characteristics of the heart rate variability at the University of La Plata, Argentina [1-7]. Here we share the resulting database to contribute to the study of long-term characteristics of heart rate variability.

Methods

We analysed Holter recordings from healthy subjects who were recruited as volunteers after an exhaustive interview and clinical examination. Approval for this study was granted by the Ethics Committee of the National University of La Plata (UNLP) for data protection and privacy. In accordance with the Declaration of Helsinki, prior to participation in the study, all subjects were informed about the study and gave their written consent (in the case of children under 16, their parent's or legal guardian's consent was required). The study included participants without clinical symptoms of disease, who were not on medication and whose electrocardiograms (ECG) were normal according to the criteria summarized in Table 1

Holter monitoring was recorded for 24 h with digital three-channel DMS300 7 and DMS300 3A recorders, and Galix recorders, using 3M electrodes. The Galix recorders had a programmable read in sampling rate of 512 and 1024 Hz, and a write out sampling rate of 128 Hz. The DMS recorders had a sampling rate of 1024 Hz per channel for signal-averaged electrocardiography (SAEG) analysis, a read in sampling rate of 512 Hz, and a write out sampling rate of 128 Hz in the other cases. The signals were analysed with Galix software, and CardioScan 10.0, 11.0 software for DMS recorders. The error in the RR interval determination was of about 8 ms (twice the error in the determination of the R peak). The cardiac events in the records were automatically detected and classified by the Holter software, and then examined and corrected by two cardiologists. The normality criteria in Table 1 impose a maximum number of nonsinus beats on the signal, so the few nonsinus beats present were not removed from the records.

The records were then analysed beat by beat, to identify and correct as many R peaks as possible. In this way we reduced the amount and duration of artifacts in the signal. After this analysis, the HRV time series were obtained and further analysed. Time series with more than 8% artifacts were not included, nor those with artifacts longer than 20 s. These selection criteria were established to remove the artifact segments without significantly affecting the dynamic behavior of the time series, as evaluated in [1-3].

In the accepted series the segments containing artifacts were removed. The time series do not contain artifacts. Also stationarity was evaluated, and surrogate analysis was performed as in [2].

Data Description

The database contains the HRV time series of 147 individuals, 72 males and 67 females (gender data on the other 8 individuals are unavailable). Of the 147 individuals, 71 were under one year of age, and 10 were over 18 years of age. In the database each individual is identified by an ID number, and the HRV time series of this individual is an ID.txt file. The files are text files (.txt, ANSI format) and contain RR intervals in milliseconds. Additional patient information is stored in the file patient-info.csv.

Usage Notes

The artifacts in the time series of this database were removed, so that RR intervals from the two disjoint segments before and after the gap were simply concatenated, and the RR intervals may not be contiguous. This modification introduces noise into the series, and we kept it at low levels by limiting both the maximum number of artifacts to 8%, and their duration to 20s (below enough the correlation length of the series).

The time series included in this database were used in [1-7]. The criteria for the level of artifacts were established in [1-3], to prevent noise from affecting the dynamic behaviour of the time series. The time series in this database are presented as supporting information for [7]. In this article, we performed an analysis of the effect of age on the HRV, including a comparison with other studies [8-13].

These data are from healthy population, including paediatric subjects, and can be used as a control group for the study of diseases with progressive HRV alterations [6], congenital heart disease, etc. No special software is required to use the data. For the non-linear analysis performed in [1] we use the TISEAN package [14].

Conflicts of Interest

All contributors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

1. Andrés DS, Irurzun IM, Mitelman J, Mola EE. Increase in the embedding dimension in the heart rate variability associated with left ventricular abnormalities. Appl. Phys. Lett. 2006; 89, 144111, doi:10.1063/1.2349285.
2. Irurzun IM, Mola EE. Heart Rate Variability: A View from Chaos Theory. 1st. Edition. LAP Lambert Academic Publishing, 2015.
3. Andrés DS. Desarrollo de nuevas herramientas y algoritmos diagnósticos para el monitoreo cardíaco. PhD Thesis. Universidad Nacional de La Plata. Argentina, 2008.
4. Defeo MM. Influencia del modo de estimulación y de la tasa de captura sobre las propiedades complejas de la variabilidad de la frecuencia cardíaca en pacientes con marcapasos implantables. PhD Thesis. University of La Plata. Argentina, 2021.
5. Irurzun IM, Defeo MM, Garavaglia L, Thomas Mailland J, Mola EE. Scaling behavior in the heart rate variability characteristics with age, arXiv:1810.04269 [q-bio.QM]. Sep. 2018.
6. Irurzun IM, Defeo MM, De Battista MR, Mola EE. PS204 Losses of Long-Range Correlation in the Heart Rate Variability of Patients with Chagas Disease. Global Heart, 2016, 11 (2 Supp): e 47-48, doi: 10.1016/j.gheart.2016.03.166
7. Garavaglia L, Gulich D, Defeo MM, Thomas Mailland J, Irurzun IM. The Effect of Age on the Heart Rate Variability of Healthy Subjects. Accepted in PlosOne.
8. Bobkowski W, Stefaniak ME, Krauze T, Gendera K, Wykretowicz A, Piskorski J, Guzik P. Measures of heart rate variability in 24-h ECGs depend on age but not gender of healthy children. Front. Physiol 2017; 8:311, doi: 10.3389/fphys.2017.00311.
9. Pikkujamsa S. Heart Rate Variability and Baroreflex Sensitivity in Subjects without Heart Disease. Effects of age, sex and cardiovascular risk factors. PhD Thesis. Oulu University. Finland, 1999. Available from http://jultika.oulu.fi/files/isbn9514252276.pdf
10. Umetani K, Singer DH, McCraty R, Atkinson M. Twenty-four hour time domain heart rate variability and heart rate: relations to age and gender over nine decades. JACC 1998; 31(3): 593-601, doi: 10.1016/s0735-1097(97)00554-8.
11. Goldberger A, Amaral L, Glass L, Hausdorff J, Ivanov P.C, Mark R, Mietus JE, Moody GB, Peng CK, Stanley HE. PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation 2000; 101 (23): e215- e220, doi: 10.1161/01.cir.101.23.e215.
12. Normal Sinus Rhythm RR Interval Database. doi.org/10.13026/C2S881.
13. MIT-BIH Normal Sinus Rhythm Database. doi.org/10.13026/C2NK5R.
14. Hegger R, Kantz H, Schreiber T. Practical implementation of nonlinear time series methods: The TISEAN package, Chaos 1999; 9: 413.