| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Department of Physiology, School of Medical Sciences, University of Bristol, Bristol BS8 1TD, UK 2 Experimental Animal Center, Fukushima Medical University School of Medicine, Fukushima 960-1295, Japan 3 Henry Wellcome Laboratories for Integrative Neurosciences and Endocrinology, University of Bristol, Dorothy Hodgkin Building, Bristol BS1 3NY, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 4 August 2005;
accepted after revision 18 October 2005; first published online 20 October 2005)
Corresponding author H. Waki: Department of Physiology, School of Medical Sciences, University of Bristol, Bristol BS8 1TD, UK. Email: h.waki{at}bristol.ac.uk
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Evaluating alterations in the balance of cardiovascular sympathetic/parasympathetic motor activity is important for gaining insights into neural mechanisms of cardiovascular disease. In addition, many diseases are associated with a change in baroreceptor reflex gain (sBRG). For example, essential hypertension is related to an increase in sympathetic tone (Grassi et al. 1998) and an attenuated cardiac baroreceptor reflex (e.g. Korner, 1995), which is an important prognostic indicator of autonomic dysfunction (La Rovere et al. 2001). Autonomic nerve activities, such as renal nerve or lumbar sympathetic chain activity, can be measured directly via implantable electrodes from freely moving rodents (Miki et al. 2002). However, in the rat, a widely used animal model for cardiovascular science, this technique is not suitable for long-term evaluation of autonomic function (i.e. greater than 3 weeks) because the signal/noise ratio can deteriorate dramatically (Miki et al. 2002). Traditionally, baroreceptor reflex function is measured using vasoactive agents but this technique is not suitable for continuous, chronic evaluation because repeated doses of these agents may result in down-regulation of vascular responses. Moreover, animals have to be catheterized, which can cause stress, and agents can directly affect both receptor and end organs (Peveler et al. 1983; Casadei & Paterson, 2000; Meyrelles et al. 2003). In summary, none of these means of analysis is appropriate for long-term measurements in chronically instrumented awake rodents, such as rats or mice.
An alternative method is a mathematical analysis of the BP signal. It is generally accepted that the high-frequency (HF) component of power spectral density in heart rate (HR) variability reflects the activity of the cardiac parasympathetic innervation whereas the ratio of low-frequency (LF) to HF (LF:HF) component of HR variability is an index of cardiac sympathetic tone (Pagani et al. 1986). Similarly, the LF power in the spectral density of BP variability reflects vasomotor sympathetic tone (deBoer et al. 1987; Madwed et al. 1989) whereas the very LF (VLF) component includes modifications by hormonal agents (Akselrod et al. 1985; Cerutti et al. 1991). Also, changes in BP and corresponding responses in either pulse interval (PI) or HR allow analysis of the spontaneous cardiac baroreceptor reflex gain (sBRG; Oosting et al. 1997; Waki et al. 2003a,b).
Although there are a number of software packages commercially available at present, none can automatically acquire BP signals and calculate all of the cardiovascular-related parameters described above. Moreover, no data have been published which thoroughly validated any of these packages. In this study, we introduce and validate a new software package – ‘Hey-Presto’ (Mizuno software, Miyagi, Japan) – designed to evaluate automatically cardiovascular autonomic function from the BP signal in freely moving animals. The software is written in Visual C++ Ver. 6.0 and runs under Windows 2000 or XP operating systems. The software acquires pulsatile BP signals and calculates variability in both HR and BP as well as sBRG and rate of respiration. We confirm the validity of the results generated by the software using conventional cardiovascular pharmacological tools in unrestrained conscious rats fitted with a radio-telemetery BP transducer. Moreover, using the software, we clarify the dynamic characteristics of the spontaneous cardiac baroreceptor reflex in conscious rats. We also demonstrate its versatility by demonstrating its application to pigs and humans.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Data acquisition. The pulsatile BP signal is recorded and then stored on a computer (Ideal specification: central processing unit (CPU), Pentium4, 2.66 GHz; hard disk drive (HDD), 40 Gbytes; Memory, 512 Mbytes) via an analog input box (IP-810 A, GigaTex Co, Ltd, Miyagi, Japan) connected to an analog input/output (I/O) PC card (AD 12–8(PM), CONTEC Co Ltd, Japan) for digitization. Four separate BP signals can be recorded simultaneously at a sampling rate of up to 4 kHz per channel. Each BP channel is accompanied by an additional analog input channel available for a signal marker or electromyograph. The exact timing of data sampling, its duration and periodicity of collection are all adjustable and controlled by the software (Fig. 1). During data acquisition the program displays the pulsatile BP traces, systolic (SBP), mean (MBP) and diastolic (DBP) pressures as well as PI, HR and respiratory frequency (Fig. 1).
|
|
|
Parameters used to calculate sBRG can be changed by the user (Fig. 2A). For example, either SBP or DBP and either PI or HR can be used. The threshold level for detection of a change in BP to calculate sBRG can be set randomly. Both the moving average function (to filter out respiration-induced fluctuations) and the delay time from a change in BP to a reflex response in PI can be set within a certain range. These adjustments may be needed when sBRG is calculated from different species.
In Hey-Presto, PI and SBP are used as the parameters for determining the variability in HR and BP, respectively. The beat-to-beat PI or SBP data are re-sampled with 1, 5 or 10-Hz frequency using a spline interpolation. After the linear trend is removed, power spectral density is computed using the FFT algorithm. Variable setting of the window for the FFT is shown in Fig. 2A.
The software contains a function for measuring respiratory rate. This is detected through the respiratory dependent changes in BP waves. Moving averages of SBP over 10 beats are calculated which provides a threshold line. Respiratory related oscillations in the original SBP data that exceed this threshold will be detected as a single respiratory cycle.
Validating the software
Experimental animals and animal care. The procedures were carried out according to the UK Home Office Guideline on Animals (Scientific Procedures) Act 1986. Male Sprague-Dawley rats between 275 and 325 g were used and housed individually, allowed normal rat chow and drinking water ad libitum, and kept on a 12 h light–12 h dark cycle.
Measurement of BP using radio-telemetry. For full details of the method see Waki et al. (2003b). Briefly, the radio-telemetry system (Data Sciences International) was used for recording BP. The system consists of three basic elements: (1) a transmitter for monitoring BP (TA11PA-C40); (2) a receiver (RPC-1); and (3) an adapter (R11CPA) with an ambient pressure monitor (APR-1) that produces analog output signals of pulsatile BP. Telemetry data were acquired by Hey-Presto through an analog input box and analog I/O PC card as described above, displayed on the computer screen and stored on hard drive. The transmitter was implanted at least 7 days before any experimental protocol began. PI and HR were measured from the pressure pulses.
Validation of sBRG analysis.
A polyethylene catheter (SP-31, Natsme) was inserted into the right jugular vein under anaesthesia (ketamine, 60 mg kg–1 and medetomidine, 250 µg kg–1, I.M.). The baroreceptor reflex gain was determined pharmacologically in conscious rats 48 h after installation of the venous line. Either phenylephrine (10–20 µg kg–1, total volume 2.75–6.50 µl; Sigma-Aldrich) or nitroprusside (10–20 µg kg–1, total volume 2.75–6.50 µl; Sigma-Aldrich) was injected to raise or lower BP by about 25–50 mmHg, respectively. Both pressor and depressor challenges were performed at least twice. In order to assess the baroreceptor reflex gain, the peak changes in SBP (
SBP) and the corresponding peak reflex changes in PI (PI) were measured manually and the PI/SBP ratios obtained by both phenylephrine and nitroprusside injection were averaged and used as an index of reflex gain. Tests were made before and after either cardiac vagal blockade (atropine, 1 mg kg–1, I.V., n
= 4) or cardiac sympathetic blockade (atenolol, 1 mg kg–1, I.V., n
= 4). These data were compared with the sBRG obtained using the software where the inputs were SBP and either PI or HR.
Validation of HR variability analysis. PI variability was analysed using the FFT function of the new software. For each 5-min recording period, the beat-to-beat pulse intervals were converted into data points every 100 ms using a spline interpolation. The resulting time series were divided into half-overlapping sequential sets of 512 data points. For each data set, after the linear trend was removed and Hanning window applied, power spectral density was computed using the FFT algorithm (Cerutti et al. 1994). According to Murasato et al. (1998), the magnitude of power was integrated in both the LF band between 0.27 and 0.75 Hz and the HF band (0.75–3.3 Hz). To validate HR variability analysis, we measured changes in HF power of the PI before and after vagal blockade (atropine, 1 mg kg–1, I.V., n = 4). LF:HF ratio of the PI was also calculated before and after cardiac sympathetic blockade (atenolol, 1 mg kg–1, I.V., n = 4).
Validation of BP variability analysis. We assessed whether the LF power of BP variability correlated with changes in vasomotor sympathetic nerve activity recorded simultaneously in a conscious rat. We measured lumbar sympathetic nerve activity using techniques described by others (see Miki et al. 2004) in rats fitted with a radio transmitter. The electrode for lumbar chain recording was implanted under ketamine (60 mg kg–1) and medetomidine (250 µg kg–1) anaesthesia, as described above. The electrical activity was amplified using an amplifier (AB-651 J, Nihon Kohden, Japan) with a band-pass filter of 300–3000 Hz. The signal was further filtered at 300 Hz with high-pass filter (E-320B Decade Filter, NF Electronic Instruments, Yokohama, Japan). Sympathetic nerve activity and BP were measured continuously for 60 min from a conscious animal 24 h after surgery. During this measurement, the animal was relatively quiet. Pulse count values of nerve activity (counts min–1) were analysed by Maclab system (M-8260 8S, AD Instruments) and the values were compared to LF power of SBP derived by the software for 5-min epochs of data (12 sets in total).
Validation of respiratory rate analysis. When animals were relatively quiet, respiratory related thoracic movements were counted manually by visual inspection for 1 min and those values were then compared with the computer-derived values (1-min average, n = 7).
Circadian rhythm analysis. To validate whether the new software could detect a circadian rhythm of cardiovascular related parameters, HR and BP variability, and sBRG were measured for 24 h during which time the BP signal was measured continuously for 5 min in every hour from 09.00 h until 08.05 h on the following day (n = 8).
Versatility of Hey-Presto
To test the robustness and versatility of Hey-Presto, data from different species were analysed.
Application to a mature pig. A 34-kg pig was used. The procedures were carried out according to the UK Home Office Guideline on Animals (Scientific Procedures) Act 1986. Anaesthesia was induced with an intramuscular injection of ketamine (10 mg kg–1) and the pig was anaesthetized with 2.25–2.5% halothane in 40% O2–60% N2O. The left common carotid was exposed in the supine position and the tip of the catheter from the transmitter (TA11PA-C40) was inserted into the artery. The carotid arterial pressure was measured using the telemetry system (Data Sciences International). Data were acquired for 5 min via 1401 CED interface to a computer running Spike2 software (Cambridge Electronic Designs, UK) and the BP signal was converted into a text file, and transferred to the Hey Presto program. The power spectra of PI and SBP were analysed. For analysis, 5-Hz re-sampling, 512 data sets, half-overlapping, Hanning window were selected from the program (Fig. 2A).
Application to humans. The Portapres Model-2 (TNO TPD Biomedical Instrumentation, Netherlands; see Imholz et al. 1998) was employed for measurement of the BP from seven healthy men and two healthy women (22–40 years old). BP was measured for 10 min in the standing position following to 10 min recording in the supine position. During the recording period, respiratory rhythm was controlled at 12 breaths min–1 (Sandercock et al. 2004). Using the Spike 2 software, the BP signal for each 10-min recording period was extracted and converted into a text file, and transferred to the data analysis component of the Hey Presto program. For each 10-min recording period, HR and BP variability and sBRG were measured using the analysis program. For power spectral analysis, the beat-to-beat data were converted into data points every 1 s using a spline interpolation. The resulting time series were divided into half-overlapping sequential sets of 256 data points. For each data set, Hanning window was applied and the spectra obtained for the different data sets were averaged. The magnitude of power was integrated in both the LF band between 0.04 and 0.15 Hz and the HF band between 0.15 and 0.4 Hz band (Sandercock et al. 2004). The study was approved by an institutional review committee and the subjects gave informed written consent.
Data analysis
Group data were expressed as means ± S.E.M. Differences were compared by either paired t test or one-way ANOVA followed by Bonferroni's procedure for multiple comparisons. Data were taken as significant at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The sBRG and HR variability analysis function was validated by vagal and cardiac sympathetic blockade in conscious rats. MBP did not change significantly after application of either atropine or atenolol (control, 103 ± 2 mmHg; atropine, 110 ± 4 mmHg; atenolol, 104 ± 1 mmHg, n = 4).
Validation of sBRG. The sBRG, as calculated by Hey-Presto, decreased by 50% after vagal blockade (control, 0.74 ± 0.08 ms mmHg–1; after vagal blockade, 0.37 ± 0.03 ms mmHg–1, P < 0.001, n = 4) and decreased further by 43% after cardiac sympathetic blockade (0.21 ± 0.01 ms mmHg–1, P < 0.05, n = 4). Similar reductions were found in the baroreceptor reflex gain as determined using conventional vasoactive drugs: after vagal blockade it was decreased significantly by 41% (control, 1.11 ± 0.08 ms mmHg–1; after vagal blockade, 0.65 ± 0.11 ms mmHg–1, P < 0.01) and further decreased by 57% after cardiac sympathetic blockade (0.28 ± 0.04 ms mmHg–1, P < 0.05). It is important to note that there was a high correlation coefficient between the automated and manually calculated data sets (r = 0.99, Fig. 3). The control values of sBRG calculated by Hey-Presto were significantly lower than those measured by using vasoactive agents (P < 0.01). A similar result was obtained after vagal blockade (P < 0.05) but there was no difference following cardiac sympathetic blockade.
|
The number of pressure ramps required for calculating sBRG before and after addition of atropine and atenolol was calculated as well as their average size. Although the average range of magnitudes of the pressure ramps had a tendency to be decreased by the pharmacological agents, there were no significant differences in those values among the three conditions (i.e. control, 4.8 ± 0.5 mmHg; after vagal blockade, 3.8 ± 0.3 mmHg; after subsequent sympathetic blockade, 4.0 ± 0.7 mmHg). On the other hand, the number of ramps was significantly decreased after vagal blockade (control, 79 ± 10; after vagal blockade, 41 ± 4, P < 0.01). Following subsequent blockade of cardiac sympathetic transmission, the number of ramps did not change (i.e. 40 ± 2). These findings may reflect an attenuation of baroreceptor reflex by blockade of cardiac autonomic transmission.
Compared with spontaneous fluctuation of BP for sBRG measurements, average changes in BP by vasoactive drugs were significantly (P < 0.001) larger in all three conditions (i.e. control, 36.6 ± 4.1 mmHg; after administration of atropine, 27.6 ± 4.7 mmHg; after subsequent administration of atenolol, 37.0 ± 8.4 mmHg).
The control values of sBRG measured at the different delay time from a change in BP to a reflex response in PI (i.e. three, four and five beats delay, see Methods) were 0.79 ± 0.10, 0.77 ± 0.11, and 0.70 ± 0.11 ms mmHg–1, respectively, and there were no differences among those delay times. This result indicates that the peak slope was evenly distributed in these three delay times. Similarly, there were no differences among those delay times after administration of atropine (three beats delay, 0.33 ± 0.02; four beats delay, 0.35 ± 0.01; five beats delay, 0.36 ± 0.01 ms mmHg–1) and atenolol (three beats delay, 0.23 ± 0.05; four beats delay, 0.24 ± 0.05; five beats delay, 0.23 ± 0.05 ms mmHg–1).
Validation of HR variability. Hey-Presto produced typical spectral patterns of PI power (Fig. 2D). The respiratory-dependent peak power was detected within the HF range (Fig. 2D). After vagal blockade, HR was significantly increased (control, 345 ± 10 beats min–1; after vagal blockade, 406 ± 8 beats min–1, P < 0.001, n = 4, Fig. 4) and, as expected, HF power of the PI was significantly decreased (control, 11.7 ± 1.1 ms2; after vagal blockade, 7.4 ± 0.5 ms2, P < 0.01, n = 4, Fig. 4). Following subsequent sympathetic blockade, HR and LF:HF of the PI were both significantly decreased (HR: control, 406 ± 8 beats min–1; after subsequent sympathetic blockade, 324 ± 7 beats min–1, P < 0.001; LF:HF of the PI: control, 0.79 ± 0.08, after subsequent sympathetic blockade, 0.43 ± 0.02, P < 0.01, n = 4, Fig. 4) while HF power of the PI did not change (control, 7.4 ± 0.5 ms2; after subsequent sympathetic blockade, 8.8 ± 0.5 ms2). Respiratory rate calculated by Hey-Presto (see below for the validation of respiratory rate) did not change after administration of atropine (control, 57.5 ± 2.4; atropine, 64.7 ± 2.4) while the value after subsequent blockade of cardiac sympathetic transmission was significantly higher than the control value (control, 57.5 ± 2.4; atenolol, 69.2 ± 3.4, P < 0.05). It is important to note that the respiratory frequency was within the range of HF in all three conditions (i.e. control, 0.96 ± 0.04 Hz; after blockade by atropine, 1.08 ± 0.04 Hz; after blockade by atenolol, 1.15 ± 0.06 Hz).
|
A highly significant positive correlation between spontaneous changes in lumbar sympathetic nerve activity and LF power of SBP was observed (see Fig. 5, r = 0.88, P < 0.001).
|
The new software detected circadian patterns in sBRG, HF power of the PI, LF:HF ratio of the PI and LF power of the SBP as well as BP and HR (n = 8, Fig. 6). BP, HR, LF:HF ratio of the PI and LF power of the SBP were increased significantly during the dark phase when averaged over 12 h (MBP: light phase, 105 ± 2 mmHg; dark phase, 111 ± 2 mmHg, P < 0.001; HR: light phase, 343 ± 5 beats min–1; dark phase, 391 ± 3 beats min–1, P < 0.001; LF:HF ratio of the PI: light phase, 0.41 ± 0.03; dark phase, 0.73 ± 0.03, P < 0.001; LF power of the SBP: light phase, 3.17 ± 0.27 mmHg2; dark phase, 4.33 ± 0.16 mmHg2, P < 0.001). In contrast, the sBRG and HF of the PI were decreased significantly during the dark phase compared to the light phase when averaged over this 12-h period (sBRG: light phase, 0.75 ± 0.10 ms mmHg–1; dark phase, 0.58 ± 0.05 ms mmHg–1, P < 0.05; HF power of the PI: light phase, 11.0 ± 0.4 ms2; dark phase, 9.2 ± 0.3 ms2, P < 0.001).
|
Respiratory rate was measured by thoracic movements and varied from 65 to 109 in conscious, unrestrained rats. A significant positive correlation between the number of respiratory depended thoracic movements and the computer-derived respiratory rate was found (see Fig. 7; n = 7, r = 0.99, P < 0.001).
|
With the computer used (CPU, Pentium4, 2.00 GHz; Memory, 512 Mbytes), the time to analyse a 5-min BP trace was about 10 s and it took approximately 4 min to evaluate a 24-h trace (5 min x 24 data sets) per channel. This indicates that Hey-Presto can analyse a large number of cardiovascular and respiratory variables extremely efficiently.
Versatility of Hey-Presto
Application to humans. In the supine position, average MBP and HR from nine human subjects were 65.3 ± 3.1 mmHg and 66.3 ± 2.6 beats min–1, respectively. The MBP measured by the Portapres from the finger showed a lower than expected value. This is consistent with previous studies when intra-arterial BP measured from a brachial artery was compared to Portapres-measured values (Imholz et al. 1998; Omboni et al. 1998). These values were significantly higher when standing (MBP, 79.1 ± 4.1 mmHg, P < 0.001; HR, 80.3 ± 3.8 beats min–1, P < 0.001). Hey-Presto produced typical spectral patterns of PI and SBP power (i.e. the respiratory dependent peak power was detected at 0.2 Hz within the HF range in both supine and standing positions; Fig. 8). This HF peak of PI was decreased by standing (a in Fig. 8) while the peak within the LF range was increased in both PI and SBP (b and c in Fig. 8). Consistent with these findings, averaged HF power of the PI was significantly decreased by standing (supine, 54.8 ± 7.2 ms2; standing, 32.2 ± 4.7 ms2, P < 0.01) while LF power of SBP and LF:HF ratio of the PI were both significantly increased by standing (LF power of SBP: supine, 2.79 ± 0.25 mmHg2; standing, 5.97 ± 0.57 mmHg2, P < 0.001; LF:HF of the PI: supine, 0.63 ± 0.07; standing, 1.33 ± 0.15, P < 0.001, Fig. 9). Although the sBRG was slightly decreased by the standing position, there was no significant difference between supine (3.98 ± 0.24 ms mmHg–1) and standing (3.26 ± 0.22 ms mmHg–1).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Validation of HR and BP variability analysis with Hey-Presto
Using Hey-Presto, we found that HR was increased while HF power of the PI was decreased after vagal blockade with atropine. Following subsequent blockade with atenolol of cardiac sympathetic transmission, HR and LF:HF of the PI were both decreased. These data support the recognized concept that HF power of the PI reflects levels of cardiac parasympathetic activity while LF:HF of the PI reflects levels of cardiac sympathetic activity (e.g. Pagani et al. 1986). It is important to note that our new software captured these differences accurately.
We found a highly significant positive correlation between LF power of BP and lumbar sympathetic nerve activity, suggesting that LF power of BP reflects the level of vasomotor sympathetic nerve activity as previously reported (deBoer et al. 1987; Madwed et al. 1989). Because of the technical limitations of chronically measuring sympathetic nerve activity from conscious rodents (Miki et al. 2002), we suggest that power spectral analysis of BP variability can be used as a valid alternative method for long-term assessment of changes in vasomotor sympathetic nerve activity. Hey-Presto also detected a circadian rhythm in cardiovascular parameters, which corresponded well to HR circadian patterns: for example, HR increased during the dark phase, which coincided with a decreased HF power of PI and an increased LF:HF of PI consistent with the data of Iellamo et al. (2004). Also, LF of SBP showed a circadian pattern: with peaks during the dark phase and troughs during the light phase presumably reflecting circadian changes in vasomotor sympathetic tone. Thus, Hey-Presto can adequately and efficiently detect and evaluate spontaneously occurring changes in cardiovascular autonomic function based on chronic recordings of BP.
Validation of sBRG analysis
The spontaneous fluctuation of BP changes were transient and ± 5 mmHg around the operating point in conscious rats. This value was consistent with our previous data (Waki et al. 2003a) and was slightly higher than values obtained by Oosting et al. (1997). On the other hand, the BP changed between 25 and 50 mmHg by using vasoactive agents to measure the baroreceptor reflex gain manually. However, the baroreceptor reflex gain calculated by Hey-Presto and that calculated manually both showed a decrease after vagal blockade, which was further reduced after subsequent cardiac sympathetic blockade. There was a high correlation coefficient between the automated and manually calculated data sets (r = 0.99, see Fig. 3). These results suggest that even if the fluctuation is transient, measuring the spontaneous fluctuation of BP and PI (or HR) allows us to evaluate the barorefex function around the operating point, and of most importance, the new software was able to derive the spontaneous baroreceptor reflex gain correctly in conscious rats.
There were quantitative differences in the absolute values in the baroreceptor reflex gain using time-series versus pharmacological analysis. The time-series method produced lower sBRG values compared to those obtained using vasoactive compounds. There may be a number of reasons to explain this. First, using time-series analysis it is not possible to construct the full reflex–response curve because of the limited range of the spontaneous changes in BP. The technique only allows a measure of reflex gain around the operating point and does not indicate ‘total gain’ or ‘maximal gain’. Second, the sBRG is the average of three different measurements made at delays of three, four and five beats (see Methods), indicating that the value of baroreceptor reflex gain could be dampened.
The inability to measure total gain or maximal gain of the baroreceptor reflex might be seen as a disadvantage of the sBRG measurement. However, the physiological relevance of the values obtained should not be underestimated because they are taken from around the operating point of this homeostatic reflex. Indeed, we argue that this may be essential for immediate reflex control of BP and HR. Moreover, the sBRG measurement allows chronic tracking of the baroreceptor reflex gain in freely moving, unrestrained, conscious animals, which is most insightful, and technically very difficult using vasoactive agents. Further, conventional methods normally require pharmacological agents (e.g. 
-agonists and NO donors), which may have direct effects at both peripheral (e.g. Peveler et al. 1983; Casadei & Paterson, 2000) and/or central sites (Paton et al. 2001) thereby influencing baroreceptor reflex data (Burger et al. 1998). We believe that using time-series analysis alleviates the complications of delivering vasoactive drugs, which dictate the need for restraining animals and inevitable induction of stress caused by indwelling catheters.
It should also be emphasized that the time needed to compute the data was negligible compared with manual analysis. For example, time to derive circadian pattern of all cardiovascular parameters, which are shown in Table 1, was less than 4 min in each animal whereas, in our experience, this would have taken more than 1 hour if performed manually.
Validation of additional functions relating to baroreceptor reflex function including -index, coherence and phase function are currently under examination.
Dynamic characteristics of spontaneous cardiac baroreceptor reflex in conscious rats
We found that there are circadian patterns of sBRG in rats (i.e. decreased in dark phase and increased in light phase). These findings are consistent with previous reports in a human study indicating that baroreceptor reflex function is up-regulated during sleep when the level of parasympathetic activity is highest (Iellamo et al. 2004). Of interest, unlike results in humans (Parlow et al. 1995), we found that the sBRG was significantly decreased after cardiac sympathetic blockade, indicating that the sBRG also reflects the cardiac sympathetic component of the baroreceptor cardiac reflex in conscious rats. This is probably due to a high level of resting sympathetic tone in rats compared with dogs and humans (Burger et al. 1998). We also found that the sensitivity of the cardiac sympathetic component is higher than the vagal component (around 80 mmHg) in the working heart brainstem preparation in rats (Simms et al. 2004). Altogether, the dynamic characteristics of sBRG in rats may be due to the control of both the sympathetic and parasympathetic component of the baroreceptor cardiac reflex.
Validation of respiratory rate analysis
In addition, the software successfully measured breathing rate during the periods when the animals were quiet and thus allowed evaluation of chronic changes of respiratory function as well as cardiovascular autonomic function. Whether the software can detect the breathing rate under a stressful condition, such as exercise or mental stress, needs further investigation.
Versatility of Hey-Presto
In human subjects, we found that HF power of PI decreased on standing, while both LF:HF of the PI and LF power of SBP increased. These results suggest that on standing, cardiac vagal activity decreased but both cardiac and vasomotor sympathetic tone increased. This resulted in an increase in both HR and BP compared with values measured in the supine position. These findings are consistent with previous reports (Vybiral et al. 1989; Lipsitz et al. 1990; Hayano et al. 1993). Thus, Hey-Presto successfully evaluated human cardiovascular autonomic function. Although for our human studies further development of the software is required to permit acquisition of the pulsatile BP signal, we have confirmed that Hey-Presto can compute cardiovascular-related parameters from data acquired using different software, such as Spike2.
In a mature pig, we found relatively similar patterns of power spectra of PI and SBP compared with the typical pattern of humans and rats. These findings support the versatility of Hey-Presto to evaluate cardiovascular autonomic functions from different species other than rats. High level of power spectral density in PI within 0.2–0.5 Hz was also found in the pig. Whether this reflects high cardiac sympathetic tone in this species or an anaesthetic-related bias in autonomic motor activity requires further investigation.
In summary, Hey-Presto is a powerful analytical tool for analysis and quantification of long-term changes in cardiovascular autonomic function without the need for repeated interventions such as intravenous infusion of drugs. Moreover, we have confirmed its versatility by demonstrating its application to larger animals and man.
Limitations
Since power spectral analysis of HR and BP variability was introduced, some researchers have expressed concerns about the validity of calculating both LF:HF of PI (or R–R interval) as an index of the cardiac sympathetic tone and LF power of BP as an index of the vasomotor sympathetic tone (Stauss et al. 1995; Sloan et al. 1996; van de Borne et al. 1997; Taylor et al. 1998). These concerns arose because the results of the power spectrum analyses did not match data obtained from direct measurements of either muscle sympathetic nerve activity or changes in plasma levels of adrenaline (epinephrine) and noradrenaline (norepinephrine). However, in these studies, data were compared between subjects and were therefore confounded by differences in basal levels between individuals. Thus, the use of power spectral analysis for evaluating autonomic nerve activity may be restricted to comparisons within subjects before and after a perturbation only. In addition, LF power of BP may also include changes in the cardiac output during states when circulating blood volume, or its distribution, are changed. Use of power spectral analysis to evaluate the vasomotor sympathetic tone therefore is limited under certain physiological conditions. In a related issue, whether the power spectrum analysis for evaluating autonomic nerve activity can be used under stressful conditions, such as exercise or mental stress, needs further investigation.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Brockway BP, Mills PA & Azar SH (1991). A new method for continuous chronic measurement and recording of blood pressure, heart rate and activity in the rat via radio-telemetry. Clin Exp Hypertens 13, 885–895.
Burger HR, Chandler MP, Rodenbaugh DW & Dicarlo SE (1998). Dynamic exercise shifts the operating point and reduces the gain of the arterial baroreceptor reflex in rats. Am J Physiol 275, R2043–R2048.
Casadei B & Paterson DJ (2000). Should we still use nitrovasodilators to test baroreceptor reflex sensitivity? J Hypertens 18, 3–6.[Medline]
Cerutti C, Barres C & Paultre C (1994). Baroreceptor reflex modulation of blood pressure and heart rate variabilities in rats: assessment by spectral analysis. Am J Physiol 266, H1993–H2000.
Cerutti C, Gustin MP, Paultre CZ, Lo M, Julien C, Vincent M & Sassard J (1991). Autonomic nervous system and cardiovascular variability in rats: a spectral analysis approach. Am J Physiol 261, H1292–H1299.
deBoer RW, Karemaker JM & Strackee J (1987). Hemodynamic fluctuations and baroreceptor reflex sensitivity in humans: a beat-to-beat model. Am J Physiol 253, H680–H689.
Goldman S & Raya TE (1995). Rat infarct model of myocardial infarction and heart failure. J Card Fail 1, 169–177.[CrossRef][Medline]
Grassi
G, Cattaneo
BM, Seravalle
G, Lanfranchi
A
&
Mancia
G (1998). Baroreceptor reflex control of sympathetic nerve activity in essential and secondary hypertension. Hypertension
31, 68–72.
Hayano J, Taylor JA, Yamada A, Mukai S, Hori R, Asakawa T, Yokoyama K, Watanabe Y, Takata K & Fujinami T (1993). Continuous assessment of hemodynamic control by complex demodulation of cardiovascular variability. Am J Physiol 264, H1229–H1238.
Iellamo
F, Placidi
F, Marciani
MG, Romigi
A, Tombini
M, Aquilani
S, Massaro
M, Galante
A
&
Legramante
JM (2004). Baroreceptor reflex buffering of sympathetic activation during sleep: evidence from autonomic assessment of sleep macroarchitecture and microarchitecture. Hypertension
43, 814–819.
Imholz
BP, Wieling
W, van Montfrans
GA
&
Wesseling
KH (1998). Fifteen years experience with finger arterial pressure monitoring: assessment of the technology. Cardiovasc Res
38, 605–616.
Kato
N, Nabika
T, Liang
YQ, Mashimo
T, Inomata
H, Watanabe
T, Yanai
K, Yamori
Y, Yazaki
Y
&
Sasazuki
T (2003). Isolation of a chromosome 1 region affecting blood pressure and vascular disease traits in the stroke-prone rat model. Hypertension
42, 1191–1197.
Korner PI (1995). Cardiac baroreceptor reflex in hypertension: role of the heart and angiotensin II. Clin Exp Hypertens 17, 425–439.[Medline]
La Rovere
MT, Pinna
GD, Hohnloser
SH, Marcus
FI, Mortara
A, Nohara
R, Bigger
JT
Jr, Camm
AJ, Schwartz
PJ
&
ATRAMI Investigators. Autonomic Tone and Reflexes After Myocardial Infarcton. (2001). Baroreceptor reflex sensitivity and heart rate variability in the identification of patients at risk for life-threatening arrhythmias: implications for clinical trials. Circulation
103, 2072–2077.
Lipsitz
LA, Mietus
J, Moody
GB
&
Goldberger
AL (1990). Spectral characteristics of heart rate variability before and during postural tilt. Relations to aging and risk of syncope. Circulation
81, 1803–1810.
Madwed JB, Albrecht P, Mark RG & Cohen RJ (1989). Low-frequency oscillations in arterial pressure and heart rate: a simple computer model. Am J Physiol 256, H1573–H1579.
Meyrelles
SS, Sharma
RV, Mao
HZ, Abboud
FM
&
Chapleau
MW (2003). Modulation of baroreceptor activity by gene transfer of nitric oxide synthase to carotid sinus adventitia. Am J Regul Integr Comp Physiol
284, R1190–R1198.
Miki K, Kosho A & Hayashida Y (2002). Method for continuous measurements of renal sympathetic nerve activity and cardiovascular function during exercise in rats. Exp Physiol 87, 33–39.[Abstract]
Miki K, Oda M, Kamijyo N, Kawahara K & Yoshimoto M (2004). Lumbar sympathetic nerve activity and hindquarter blood flow during REM sleep in rats. J Physiol 15, 261–271.
Murasato Y, Hirakawa H, Harada Y, Nakamura T & Hayashida Y (1998). Effects of systemic hypoxia on R-R interval and blood pressure variabilities in conscious rats. Am J Physiol 275, H797–H804.
Omboni
S, Parati
G, Castiglioni
P, Di Rienzo
M, Imholz
BP, Langewouters
GJ, Wesseling
KH
&
Mancia
G (1998). Estimation of blood pressure variability from 24-hour ambulatory finger blood pressure. Hypertension
32, 52–58.
Oosting J, Struijker-Boudier HAJ & Janssen BJA (1997). Validation of a continuous baroreceptor reflex sensitivity index calculated from spontaneous fluctuations of blood pressure and pulse interval in rats. J Hypertens 15, 391–399.[CrossRef][Medline]
Pagani
M, Lombardi
F, Guzzetti
S, Rimoldi
O, Furlan
R, Pizzinelli
P, Sandrone
G, Malfatto
G, Dell'Orto
S
&
Piccaluga
E (1986). Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympatho–vagal interaction in man and conscious dog. Circ Res
59, 178–193.
Parlow
J, Viale
JP, Annat
G, Hughson
R
&
Quintin
L (1995). Spontaneous cardiac baroreflex in humans. Comparison with drug-induced responses. Hypertension
25, 1058–1068.
Paton
JFR, Deuchars
J, Ahmad
Z, Wong
L-F, Murphy
D
&
Kasparov
S (2001). Adenoviral vector demonstrates that angiotensin II-induced depression of the cardiac baroreceptor reflex is mediated by endothelial nitric oxide synthase in the nucleus tractus solitarii of the rat. J Physiol
531, 445–458.
Peveler RC, Bergel DH, Robinson JL & Sleight P (1983). The effect of phenylephrine upon arterial pressure, carotid sinus radius and baroreflex sensitivity in the conscious greyhound. Clin Sci (Lond) 64, 455–461.[Medline]
Sandercock GR, Bromley P & Brodie DA (2004). Reliability of three commercially available heart rate variability instruments using short-term (5-min) recordings. Clin Physiol Funct Imaging 24, 359–367.[CrossRef][Medline]
Simms A, Paton JFR & Pickering AE (2004). Differences in the arterial pressure responsiveness of the sympathetic and parasympathetic baroreceptor reflex. J Physiol 560.P, C6.
Sloan
RP, Shapiro
PA, Bagiella
E, Bigger
JT
Jr, Lo
ES
&
Gorman
JM (1996). Relationships between circulating catecholamines and low frequency heart period variability as indices of cardiac sympathetic activity during mental stress. Psychosom Med
58, 25–31.
Stauss HM, Mrowka R, Nafz B, Patzak A, Unger T & Persson PB (1995). Does low frequency power of arterial blood pressure reflect sympathetic tone? J Auton Nerv Syst 54, 145–154.[CrossRef][Medline]
Taylor JA, Williams TD, Seals DR & Davy KP (1998). Low-frequency arterial pressure fluctuations do not reflect sympathetic outflow: gender and age differences. Am J Physiol 274, H1194–H1201.
van de Borne
P, Montano
N, Pagani
M, Oren
R
&
Somers
VK (1997). Absence of low-frequency variability of sympathetic nerve activity in severe heart failure. Circulation
95, 1449–1454.
Vybiral T, Bryg RJ, Maddens ME & Boden WE (1989). Effect of passive tilt on sympathetic and parasympathetic components of heart rate variability in normal subjects. Am J Cardiol 63, 1117–1120.[CrossRef][Medline]
Waki H, Kasparov S, Katahira K, Shimizu T, Murphy D & Paton JF (2003a). Dynamic exercise attenuates spontaneous baroreceptor reflex sensitivity in conscious rats. Exp Physiol 88, 517–526.[Abstract]
Waki
H, Kasparov
S, Wong
LF, Murphy
D, Shimizu
T
&
Paton
JF (2003b). Chronic inhibition of endothelial nitric oxide synthase activity in nucleus tractus solitarii enhances baroreceptor reflex in conscious rats. J Physiol
546, 233–242.
Yamori Y (1999). Implication of hypertensive rat models for primordial nutritional prevention of cardiovascular diseases. Clin Exp Pharmacol Physiol 26, 568–572.[CrossRef][Medline]
|
Acknowledgements |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Fazan Jr, D. A. Huber, C. A. A. Silva, V. J. Dias da Silva, M. C. O. Salgado, and H. C. Salgado Sildenafil acts on the central nervous system increasing sympathetic activity J Appl Physiol, June 1, 2008; 104(6): 1683 - 1689. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Duale, H. Waki, P. Howorth, S. Kasparov, A. G. Teschemacher, and J. F.R. Paton Restraining influence of A2 neurons in chronic control of arterial pressure in spontaneously hypertensive rats Cardiovasc Res, October 1, 2007; 76(1): 184 - 193. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Waki, B. Liu, M. Miyake, K. Katahira, D. Murphy, S. Kasparov, and J. F.R. Paton Junctional Adhesion Molecule-1 Is Upregulated in Spontaneously Hypertensive Rats: Evidence for a Prohypertensive Role Within the Brain Stem Hypertension, June 1, 2007; 49(6): 1321 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. R. Antunes, S. T. Yao, A. E. Pickering, D. Murphy, and J. F. R. Paton A spinal vasopressinergic mechanism mediates hyperosmolality-induced sympathoexcitation J. Physiol., October 15, 2006; 576(2): 569 - 583. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Waki, D. Murphy, S. T. Yao, S. Kasparov, and J. F.R. Paton Endothelial NO Synthase Activity in Nucleus Tractus Solitarii Contributes to Hypertension in Spontaneously Hypertensive Rats Hypertension, October 1, 2006; 48(4): 644 - 650. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M. Stauss, J. A. Moffitt, M. W. Chapleau, F. M. Abboud, and A. K. Johnson Baroreceptor Reflex Sensitivity Estimated by the Sequence Technique is Reliable in Rats Am J Physiol Heart Circ Physiol, July 1, 2006; 291(1): H482 - H483. [Abstract] [Full Text] [PDF]
|
|
|
|
