Technical pitfalls of heart rate variability analysis

Heart rate variability (HRV) analysis is considered a popular, non-invasive method for investigating autonomic regulation of cardiovascular system. However, inappropriate technical conditions can result in false and inaccurate measurement. Present booklet (based on the author’s still applicable PhD Dissertation from 2005) details the effects of analogue corner frequencies, AC notch filtering and sampling rate of uncorrupted and noise-corrupted ECG signals on the accuracy and precision of heart rate interval detection and HRV indices based on hardware and software simulation. Power line notch filtering is highly recommended and a bandwidth of 0.5-20Hz is considered sufficient, since neither of them deteriorate the accuracy and precision of heart rate detection, conversely, they improve it by suppressing unwanted noises and interferences. A sampling rate of 1ms is necessary for correct measurement of variability especially in patients with seriously reduced HRV; however, lower sampling frequency may be satisfactory when higher variability or larger differences are expected. Taking above technical considerations into account can improve the validity of HRV measurement.