New algorithms for real-time, 24 hr continuous and noise-adjusted power spectral analysis of heart rate and blood pressure fluctuations in conscious rats

The effective combination of C++ and assembler could realize real-time power spectral analysis of various fluctuation indexes simultaneously. The blood pressure (BP) waveform indexes (WIs) analyzed simultaneously were heart rate (HR), systolic blood pressure (SEP), mean blood pressure (MBP) and diastolic blood pressure (DBP). Power amplitudes (very low frequency, VLFamp; low frequency, LFamp; and high frequency, HFamp) were evaluated by accurate BP waveform recognition, accurate automatic rejection of outliers, baseline adjustment in periodograms and digital filtering of each amplitude. In the in vivo experiments, the amplitudes were changed in a dose-dependent manner by methylatropine (HR-VLFamp, HR-LFamp and HR-HFamp), phentolamine (HR-LFamp, SBP-VLFamp, SBP-LFamp and SBP-HFamp) and propranolol (HR-LFamp and SBP-LFamp). The absolute correlation coefficients of the amplitude and the change in each parameter were more than 0.96. In conclusion, this real-time, noise-adjusted power spectral analysis system for investigating HR and BP fluctuations enabled us to accurately evaluate autonomic nerve activity in conscious rats. Moreover, unlike other systems, this system was able to detect the biphasic changes in SBP-HFamps caused by phentolamine.