Signal processing
pNN50: Simple Parasympathetic HRV Metric
pNN50 is the percentage of successive NN interval pairs that differ by more than 50 milliseconds, providing a simple, intuitive measure of parasympathetic cardiac modulation that is highly correlated with RMSSD and HF power but expressed as an easily interpretable percentage.
How it works
Schematic summary of this guide. Waveforms are illustrative, not measured data.
pNN50 is computed as
The 50 ms threshold is the convention listed in the 1996 ESC/NASPE Task Force standards (doi:10.1161/01.CIR.93.5.1043).
The advantage of pNN50 is its simplicity and interpretability — it directly answers "what fraction of heartbeats show large beat-to-beat changes?" This makes it useful for patient communication and quick clinical screening. The disadvantage is that it discards magnitude information: a 51 ms difference and a 200 ms difference are treated identically. Variants such as pNN20 and pNN10 use lower thresholds for increased sensitivity in populations with low baseline HRV.
For PPG applications, pNN50 is sensitive to IBI detection accuracy. A single misdetected beat creating a spurious large IBI difference can significantly inflate pNN50 in short recordings. Quality-controlled IBI extraction with ectopic beat detection is essential before pNN50 computation.
How long to record: a review of HRV norms lists 2 minutes as the minimum recording for pNN50, and notes that 60-second ultra-short recordings have been proposed (Shaffer and Ginsberg, 2017; doi:10.3389/fpubh.2017.00258).
pNN50 versus RMSSD: pNN50 is closely correlated with parasympathetic activity, RMSSD and HF power, but RMSSD generally gives a better assessment of respiratory sinus arrhythmia, especially in older people, and most researchers prefer it (Shaffer and Ginsberg, 2017).
PPG versus ECG: in one study summarized in the same review, PPG- and ECG-derived HRV differed by less than 6% for most measures but by 29.9% for pNN50, consistent with its sensitivity to small beat-timing errors described above.
Breathing and age: changes in breathing rate or depth can shift pNN50 and RMSSD without any change in vagal tone. In 24-hour recordings from 1,743 people aged 40 to 100, both followed a U-shaped pattern with age, falling between 40 and 60 and rising again from about 70 (as summarized by Shaffer and Ginsberg, 2017).
Frequently Asked Questions
How does pNN50 compare to RMSSD?
RMSSD is preferred for research because it uses continuous magnitude information. pNN50 is more intuitive for clinical communication.
What is pNN20 and when should it be used?
pNN20 uses a 20 ms threshold instead of 50 ms, providing higher sensitivity for detecting parasympathetic changes in populations with low HRV (elderly, heart failure). It is less robust to noise than pNN50.
Is pNN50 affected by heart rate?
Yes. Higher heart rates produce shorter IBI and smaller absolute IBI differences, reducing pNN50 even without autonomic changes. Heart rate correction or using relative thresholds can mitigate this dependency.