What is the adequate sampling interval of the ECG signal for heart rate variability analysis in the time domain?

Laszlo Hejjel, Elizabeth Roth

Research output: Contribution to journalArticle

42 Citations (Scopus)

Abstract

Heart rate variability (HRV) analysis is considered a simple method for investigating neurocardiac regulation. However, measures of HRV may be corrupted by technical artifacts. In order to investigate the consequences of digitization errors on the time domain parameters, HRV measures from model tachograms resampled at different rates were compared. Two 375-element tachograms from human ECG tracings were shifted to a mean of 800 ms and stretched to standard deviations (SD) of 5-120 ms in 5 ms steps. All were resampled at 1-10 ms in 1 ms steps, 10 times repetitively at each interval. The mean, SD, relative accuracy error (RAE) and relative precision error (RPE) were calculated from the mean RR-interval, SD (SDNN), root mean square of successive RR-differences (RMSSD) and the percentage of consecutive RR-interval differences greater than 50 ms (pNN50). The RAE and RPE of the mean heart rate were below 0.1%. In the series with 5 ms SD, the SDNN-RAE exceeded 30% at 10 ms SI, its RPE was lower than 2% all through. Resampling the 15 ms SD tachogram at 1,2,4, or 10 ms resulted in RMSSD-RAE of 0.7%, 2.5%, 7.8% and 45.1%, respectively, while its RPE remained below 5%. The pNN50 shows poor accuracy and precision. An ECG sampling interval of 1 ms is recommended for HRV analysis without interpolation in order to get accurate time domain measures even in seriously reduced-variability samples. However, a lower sampling rate may be satisfactory in cases where higher variability is expected.

Original languageEnglish
Pages (from-to)1405-1411
Number of pages7
JournalPhysiological measurement
Volume25
Issue number6
DOIs
Publication statusPublished - Dec 1 2004

    Fingerprint

Keywords

  • AD conversion
  • Heart rate variability
  • Sampling error
  • Sampling interval
  • Time domain

ASJC Scopus subject areas

  • Biophysics
  • Physiology
  • Biomedical Engineering
  • Physiology (medical)

Cite this