Spontaneous Baroreflex Analysis Through the Sequence Method Quantifies the Respiratory Influences of Baroreflex

Abstract

The sequence method is one of the most used techniques to evaluate baroreflex function by means of spontaneous fluctuations of arterial pressure (AP). While other methods, such as cross‐spectral analysis, provide only the estimation of gain between AP and heart rate fluctuations, the sequence method offers an estimation of both gain and effectiveness (BEI) of the baroreflex. On the other hand, spectral methods allow the quantification of baroreflex gain in separated frequency bands, such as those associated with Mayer waves (low frequencies) or respiration (high frequencies). In this study, we aimed to evaluate whether the sequence method is able to quantify both low and high‐frequency control of the baroreflex. AP recordings of 61 conscious healthy rats were obtained from previous studies from our laboratory. Raw and low‐pass filtered (cutoff 0.8Hz) series of systolic pressure (SAP) and pulse intervals (PI) were generated from, at least, ten minutes of AP recording. Cross‐spectral analysis (transfer function) was calculated from original series, whereas the sequence method was calculated from both original and filtered series. For cross‐spectral analysis, baroreflex gain (transfer function between SAP and PI) was estimated at low (0.2 to 0.8Hz) and high (0.8 to 3.0Hz) frequencies. The sequence method was calculated for sequences of 3 beats with delays from 0 to 12 beats, with no thresholds for AP and PI and 0.8 as the threshold for correlation of values within sequence candidates. Gain of the transfer function, estimated from original series, was 0.62 ± 0.05 ms/mmHg (low frequency) and 2.27 ± 0.17 ms/mmHg (high frequency). The gain estimated by the sequence method was 3.13 ± 0.24 ms/mmHg (original series) and 1.19 ± 0.08 ms/mmHg (filtered series), whereas BEI was 18 ± 1 % (original series) and 84 ± 1 % (filtered series). Those values refer to delay of one beat, which seems to be the most reasonable timing between AP and PI. We also calculated the correlation coefficient (p) between the gain estimated from the sequence method and the gain measured at low and high frequencies of the transfer function. For original time series, we found p = 0.21 (sequence method vs. low frequency) and p = 0.71 (sequence method vs. high frequency). For filtered time series, we found p = 0.81 (sequence method vs. low frequency) and p = 0.44 (sequence method vs. high frequency). Correlation values are also represented for delay of one beat. In conclusion, the sequence method indices are poorly correlated with the low‐frequency components and highly correlated with the high‐frequency components of the baroreflex; wherein the latter is strongly related to respiration. This is evident in the construction of sequence candidates. Since only monotonic AP ramps (up or down) are accepted for the calculations in the sequence method, AP oscillations due to respiration limit the existence of sequences from slow AP changes. Thus, in order to quantify the low‐frequency components of the baroreflex using the sequence method, it is recommended to filter out the high frequencies prior to the application of the method.Support or Funding InformationFAPESP (grant 2015/20463‐0) and CAPES (grant PNPD 20131672)This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.