Reducing the effect of respiration in baroreflex sensitivity estimation with adaptive filtering

Area of Science:

• Cardiovascular Physiology
• Autonomic Nervous System Function
• Biomedical Signal Processing

Background:

• Baroreflex sensitivity (BRS) quantifies autonomic cardiovascular regulation using blood pressure and heart rate variability.
• Respiration influences blood pressure and heart rate through non-baroreflex mechanisms, potentially confounding BRS estimates.
• Controlled breathing protocols are common for BRS but may alter natural autonomic function.

Purpose of the Study:

• To develop and validate an adaptive filter to remove respiratory influences on BRS.
• To enable accurate BRS estimation during spontaneous breathing.
• To improve the reliability of BRS measurements in cardiovascular research.

Main Methods:

• An adaptive Least Mean Squares (LMS)-based filter was designed to isolate baroreflex sensitivity (BRS) from respiratory effects.
• Electrocardiogram (ECG), continuous blood pressure, and respiration were recorded during spontaneous and metronome-guided breathing (0.25 Hz).
• BRS was calculated using spectral methods in the low-frequency (LF) band, with and without the adaptive filter.

Main Results:

• BRS(LF) values were overestimated in subjects with spontaneous breathing rates below 0.15 Hz without filtering.
• The adaptive filter significantly reduced this overestimation, correcting for respiratory bias.
• Filtered BRS estimates showed improved accuracy compared to unfiltered data.

Conclusions:

• The proposed adaptive filter effectively mitigates the distorting effects of respiration on BRS.
• This method allows for more accurate BRS estimation using spontaneous breathing protocols.
• The findings support the use of spontaneous breathing for BRS assessment with the developed filtering technique.