Assessing heart rate variability from real-world Holter reports

Research output: Contribution to journalArticlepeer-review

33 Scopus citations


Real world clinical Holter reports are often difficult to interpret from a heart rate variability (HRV) perspective. In many cases HRV software is absent. Step-by-step HRV assessment from clinical Holter reports includes: making sure that there is enough usable data, assessing maximum and minimum heart rates, assessing circadian HRV from hourly average heart rates, and assessing HRV from the histogram of R-R intervals and from the plot of R-R intervals or heart rate vs. time. If HRV data are available, time domain HRV is easiest to understand and less sensitive to scanning errors. SDNN (the standard deviation of all N-N intervals in ms) and SDANN (the standard deviation of the 5-min average of N-N intervals in ms) are easily interpreted. SDNN < 70 ms post-MI is a cut point for increased mortality risk. Two times ln SDANN is a good surrogate for ln ultra low frequency power and can be compared with published cut points. SDNNIDX (the average of the standard deviations of N-N intervals for each 5-min in ms) < 30 ms is associated with increased risk in patients with congestive heart failure. RMSSD (the root mean square of successive N-N interval difference in ms) < 17.5 ms has also been associated with increased risk post-myocardial infarction. Frequency domain HRV values are often not comparable to published data. However, graphical power spectral plots can provide additional information about whether the HRV pattern is normal and can also identify some patients with obstructive sleep apnea.

Original languageEnglish
Pages (from-to)239-244
Number of pages6
JournalCardiac Electrophysiology Review
Issue number3
StatePublished - Sep 2002


  • Ambulatory ECG monitoring
  • Heart rate variability
  • Risk stratification


Dive into the research topics of 'Assessing heart rate variability from real-world Holter reports'. Together they form a unique fingerprint.

Cite this