Written and maintained by CASRAI Editorial Board
Published
On this page
Informational reference for cardiovascular clinical trialists, autonomic physiology investigators, and digital biomarker engineers. This guide documents published ESC/NASPE electrophysiological measurement standards and does not provide individual medical diagnosis or treatment advice.
Cardiology researchers, psychophysiology investigators, and decentralized clinical trial biostatisticians use this guide to compute and interpret Heart Rate Variability (HRV) metrics across time-domain (`SDNN`, `SDANN`, `RMSSD`, `pNN50`), frequency-domain (`ULF`, `VLF`, `LF`, `HF`, `LF/HF`), and non-linear (`Poincaré SD1/SD2`, `SampEn`, `DFA α1`) frameworks, enforce ESC/NASPE Task Force (1996) sampling and epoch standards, and harmonize ECG versus wearable photoplethysmography (PPG) telemetry.
What Is Heart Rate Variability Metrics and How Is It Used in Clinical Research and Healthcare?
Heart Rate Variability (HRV) metrics are quantitative mathematical indices that measure the physiological beat-to-beat fluctuations in the time intervals between consecutive normal sinus heartbeats—termed normal-to-normal (`NN`) intervals or R-R intervals on an electrocardiogram (ECG). Standardized globally by the joint Task Force of the European Society of Cardiology (ESC) and the North American Society of Pacing and Electrophysiology (NASPE) (Circulation, 1996; updated by Shaffer & Ginsberg, 2017), HRV metrics quantify the dynamic interplay between the parasympathetic (vagal) and sympathetic branches of the autonomic nervous system (ANS) acting on the sinoatrial (SA) node.
In clinical research and healthcare, HRV metrics serve as validated prognostic biomarkers for post-myocardial infarction mortality risk stratification, diabetic cardiovascular autonomic neuropathy (CAN) screening, neonatal sepsis early warning, neuropsychiatric stress/depression phenotyping, and digital endpoints in wearable health monitoring trials.
Time-Domain, Frequency-Domain, and Non-Linear HRV Metrics Reference Table
All HRV calculations begin by detecting R-wave peaks on a continuous ECG, removing ectopic beats (premature ventricular or atrial contractions) and motion artifacts, and constructing the discrete time series of adjacent sinus intervals NN1, NN2, ..., NNN (measured in milliseconds, ms):
| Metric & Domain | Units & Mathematical Definition | Standard Recording Epoch | Primary Autonomic & Physiological Interpretation |
|---|---|---|---|
SDNN (Time-Domain) |
ms — Standard deviation of all normal-to-normal (NN) intervals: √[1/(N−1) ∑(NNi − NN̄)²] |
24-hour Holter (gold standard) or 5-minute short-term epoch (never compare across different durations) | Reflects total cyclic variability across the recording period; 24-h SDNN < 50 ms is a classic post-MI mortality risk predictor (Kleiger et al.) |
SDANN (Time-Domain) |
ms — Standard deviation of the averages of NN intervals calculated over all 5-minute segments of an entire 24-hour recording |
24-hour recording only | Quantifies ultra-low-frequency circadian and neurohormonal heart rate oscillations |
RMSSD (Time-Domain) |
ms — Root mean square of successive differences between adjacent NN intervals: √[1/(N−1) ∑(NNi+1 − NNi)²] |
5-minute (or ultra-short 60-s to 24-h) | Primary time-domain index of beat-to-beat vagally mediated parasympathetic tone; less affected by slow respiratory rate changes than HF |
pNN50 (Time-Domain) |
% — Percentage of successive NN interval pairs that differ by more than 50 ms (NN50 / [N − 1] × 100) |
5-minute or 24-hour | Closely correlated with RMSSD and HF power; reflects high-frequency parasympathetic vagal activity |
HF Power (Frequency-Domain) |
ms² or nu — Power spectral density (FFT or AR) in the High-Frequency band (0.15–0.40 Hz, corresponding to 9–24 breaths/min) |
Minimum 1–5 minutes | Reflects efferent parasympathetic (vagal) activity and Respiratory Sinus Arrhythmia (RSA) |
LF Power (Frequency-Domain) |
ms² or nu — Power spectral density in the Low-Frequency band (0.04–0.15 Hz, centered near the ~0.1 Hz Mayer wave) |
Minimum 2–5 minutes | Reflects baroreceptor reflex activity and mixed sympathetic and parasympathetic modulation (predominantly baroreflex/vagal during resting paced breathing) |
VLF & ULF Power (Frequency-Domain) |
ms² — Very-Low-Frequency (0.0033–0.04 Hz) and Ultra-Low-Frequency (≤ 0.003 Hz) bands |
VLF ≥ 5–15 min; ULF requires 24 hours |
Driven by thermoregulation, renin-angiotensin-aldosterone system, intrinsic cardiac afferents, and circadian rhythm |
Poincaré SD1 & SD2 (Non-Linear) |
ms — Ellipse width perpendicular (SD1 = RMSSD / √2) and length along (SD2) the line of identity on the NNi vs. NNi+1 scatter plot |
5-minute or 24-hour | SD1 quantifies short-term beat-to-beat vagal variability; SD2 quantifies long-term sympathetic/baroreflex variability |
What Standards or Regulatory Requirements Govern Heart Rate Variability Metrics?
Rigorous HRV measurement in clinical trials requires strict adherence to signal-processing and regulatory validation standards:
- ESC/NASPE Task Force (1996) Sampling Rate and Epoch Standardization:
- ECG Sampling Frequency: Because R-peak fiducial point jitter degrades high-frequency (`RMSSD` and `HF`) accuracy, the Task Force specifies an optimal ECG sampling rate of
250 Hz to 500 Hzor higher (sampling at100–250 Hzis acceptable only if a parabolic interpolation algorithm is applied to refine the R-wave peak). - Incomparability of Recording Lengths: Because total variance (`SDNN`) grows mathematically with the duration of the recording,
SDNNvalues computed from 5-minute resting segments must never be compared directly with 24-hour Holter `SDNN` values. - Ectopic Beat & Artifact Filtering: Even a single unedited premature ventricular contraction (PVC) or missed beat in a 5-minute epoch can inflate `RMSSD` and `HF` power by more than 50%; clinical pipelines require automated QRS morphology classification and cubic-spline interpolation when artifact burden is
< 5%.
- ECG Sampling Frequency: Because R-peak fiducial point jitter degrades high-frequency (`RMSSD` and `HF`) accuracy, the Task Force specifies an optimal ECG sampling rate of
- ECG Heart Rate Variability vs. PPG Pulse Rate Variability (PRV): Wristband and ring optical sensors measure peripheral blood volume pulse waves via photoplethysmography (`PPG`), yielding Pulse Rate Variability (PRV) rather than true electrical R-R intervals. Because vascular compliance and blood pressure alter Pulse Transit Time (`PTT`), and wrist motion introduces optical artifacts, FDA Digital Health Technology (DHT) guidance and the Digital Medicine Society (DiMe) V3 framework require analytical validation against a simultaneous reference ECG patch (such as an iRhythm Zio patch) across resting nocturnal windows.
How Does Heart Rate Variability Metrics Integrate with EHR and Health Informatics Workflows?
In hospital telemetry, ambulatory Holter systems, and decentralized clinical trials, HRV metrics are standardized using Regenstrief LOINC and HL7 FHIR:
- LOINC and IEEE 11073 Coding: Standardized LOINC codes—including
80404-7(R-R interval.standard deviation [SDNN] by ECG) and80406-2(R-R interval differences.root mean square [RMSSD])—allow diverse biosensors to transmit semantically uniform values. - HL7 FHIR Observation Resources: Remote patient monitoring hubs encode epoch-level `SDNN`, `RMSSD`, and `HF` values as HL7 FHIR Observation resources with explicit `effectivePeriod` timestamps (recording the exact 5-minute or nocturnal window), `device` references, and `bodyPosition` / sleep-stage context extensions.
Common Misunderstandings
- Interpreting the
LF/HFratio as a pure “sympathovagal balance” meter: Physiological blockade studies (e.g., atropine parasympathetic blockade and propranolol beta-adrenergic blockade, summarized by Billman, 2013 and Shaffer & Ginsberg, 2017) demonstrate that `LF` power (`0.04–0.15 Hz`) is heavily influenced by vagally mediated baroreceptor activity and slow breathing (`< 9` breaths/min), invalidating simplistic interpretations of `LF/HF` as a direct ratio of sympathetic to parasympathetic nerve traffic.
Frequently Asked Questions
What is heart rate variability metrics and how is it used in clinical research and healthcare?
Heart Rate Variability (HRV) metrics—such as time-domain `SDNN`, `RMSSD`, and `pNN50` and frequency-domain `LF` and `HF` power—are standardized mathematical measures of the beat-to-beat variation in normal sinus R-R intervals (`NN` intervals), used in clinical research and cardiology to quantify autonomic nervous system function, vagal tone, and cardiovascular mortality risk.
What standards or regulatory requirements govern heart rate variability metrics?
HRV measurement is governed by the 1996 Task Force of the European Society of Cardiology (ESC) and North American Society of Pacing and Electrophysiology (NASPE) guidelines (specifying `≥ 250–500 Hz` ECG sampling rates, ectopic beat removal, and standardized 5-minute vs. 24-hour epochs) and FDA Digital Health Technology (DHT) V3 validation rules for wearable sensors.
How does heart rate variability metrics integrate with EHR and health informatics workflows?
Ambulatory ECG monitors and clinical wearables map calculated HRV indices to Regenstrief LOINC codes (such as `80404-7` for `SDNN` and `80406-2` for `RMSSD`) and transmit them into EHRs and clinical trial databases as HL7 FHIR `Observation` resources with explicit recording duration and posture metadata.
Sources
- Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation. 1996;93(5):1043–1065. https://pubmed.ncbi.nlm.nih.gov/8598068/ (2026-11-09).
- Shaffer F, Ginsberg JP. An Overview of Heart Rate Variability Metrics and Norms. Frontiers in Public Health. 2017;5:258. https://pmc.ncbi.nlm.nih.gov/articles/PMC5624990/ (2026-11-09).
- Regenstrief Institute. LOINC 80404-7: R-R interval.standard deviation (Heart rate variability) by ECG. https://loinc.org/80404-7/ (2026-11-09).







