Influence of duration and hour of recording on spectral measurements of heart rate variability

• Published in: Journal of the autonomic nervous system 1998-08, Vol.73 (1), p.1-6
• Main Authors: Fauchier, Laurent, Babuty, Dominique, Autret, Marie Laurence, Poret, Philippe, Cosnay, Pierre, Fauchier, Jean Paul
• Format: Article
• Language: English
• Subjects: Adult; Analysis of Variance; Autonomic Nervous System; Autonomic Nervous System - physiology; Biological and medical sciences; Circadian Rhythm; Circadian Rhythm - physiology; Electrocardiography, Ambulatory; Female; Fundamental and applied biological sciences. Psychology; Heart; Heart Rate; Heart Rate - physiology; Heart rate variability; Humanities and Social Sciences; Humans; Male; Middle Aged; Spectral analysis; Vertebrates: cardiovascular system
• ISSN: 0165-1838; 1872-7476
• DOI: 10.1016/S0165-1838(98)00110-6

Frequency-domain analysis of heart rate variability provides information about influences of autonomic nervous system on the heart and can be measured on short-term or long-term electrocardiogram recording. We compared heart rate variability on 24 h electrocardiogram recording with measurements of heart rate variability over shorter periods of 1 h and 12 h in order to determine the influence of duration and hour of recording on the heart rate variability parameters in the frequency domain. Heart rate variability was analyzed in 24 subjects with normal heart, with measurement of total power (Tot P), ULF, VLF, LF, HF and LF/HF ratio. Reference measurements over 24 h were compared with measurements over 4 periods of 1 h (2 AM–3 AM, 8 AM–9 AM, 2 PM–3 PM, 8 PM–9 PM) and over 4 periods of 12 h (2 AM–2 PM, 8 AM–8 PM, 2 PM–2 AM, 8 PM–8 AM). The correlations with measurements over 24 h were lower for the measurements over 1 h than for the measurements over 12 h. ULF had the lowest mean of correlation coefficients for measurements over 1 h and 12 h, whereas the other frequency bands had high correlations with measurements over 24 h. When using Bland and Altman method, only measurements over 2 AM–2 PM and 2 PM–2 AM periods were a reliable estimation of heart rate variability over 24 h. Conclusion: Spectral measurements of heart rate variability over 12 h have high correlations with measurements on 24 h. ULF is the parameter the most affected by the duration of the recording. Spectral measurements over 2 AM–2 PM and 2 PM–2 AM periods are reliable appreciations of heart rate variability over 24 h.