Validity of Ultra-Short-Term Heart Rate Variability Derived from Femoral Arterial Pulse Waveform in a British Military Cohort

Various non-electrocardiogram (ECG) based methods are considered reliable sources of heart rate variability (HRV) measurement. However, the ultra-short recording of a femoral arterial waveform has never been validated against the gold-standard ECG-based 300s HRV and was the aim of this study.

A validity study was conducted using a sample from the first follow-up of the longitudinal ADVANCE study UK. The participants were adult servicemen (n = 100); similar in age, rank, and deployment period (Afghanistan 2003–2014). The femoral arterial waveforms (14s) from the pulse wave velocity (PWV) assessment, and ECG (300s) were recorded at rest in the supine position using the Vicorder™ and Bittium Faros™ devices, respectively, in the same session. HRV analysis was performed using Kubios Premium. Resting heart rate (HR) and root mean square of successive differences (RMSSD) were reported. The Bland-Altman %plots were constructed to explore the PWV-ECG agreement in HRV measurement. A further exploratory analysis was conducted across methods and durations.

The participants’ mean age was 38.0 ± 5.3 years. Both PWV-derived HR (r = 0.85) and RMSSD (r_s=0.84) showed strong correlations with their 300s-ECG counterparts (p < 0.001). Mean HR was significantly higher with ECG than PWV (mean bias: -12.71 ± 7.73%, 95%CI: -14.25%, -11.18%). In contrast, the difference in RMSSD between the two methods was non-significant [mean bias: -2.90 ± 37.82% (95%CI: -10.40%, 4.60%)] indicating good agreement. An exploratory analysis of 14s ECG-vs-300s ECG measurement revealed strong agreement in both RMSSD and HR.

The 14s PWV-derived RMSSD strongly agrees with the gold-standard (300s-ECG-based) RMSSD at rest. Conversely, HR appears method sensitive.