THE POWER SPECTRA RESPONSE OF STROKE VOLUME AND

ARTERIAL BLOOD PRESSURE VARIABILITY SIGNALS TO

AUTONOMIC NERVOUS SYSTEM MODULATION OF THE HEART

Abdul-Hakeem H. AlOmari

1,3

, Andis Graudins

2,3,4

School of Electrical Engineering and Telecommunications;

Prince of Wales Clinical School, Faculty of Medicine

The University of New South Wales;

Clinical and Experimental Toxicology Unit and Department of Emergency Medicine

Prince of Wales Hospital , Sydney, Australia

Andrey V. Savkin

School of Electrical Engineering and Telecommunications, The University of New South Wales, Sydney, Australia

Keywords:

Stroke volume variability (SVV), Systolic blood pressure variability (SBPV), Power spectral analysis, Sym-

pathovagal balance, Levosimendan, Verapamil, Calcium channel sensitizers, Calcium channel blockers.

Abstract:

This study presents results that describe the short term oscillations in SBPV and SVV signals due to calcium

channel blockers poisoning with verapamil treated with continuous infusion of levosimendan. In addition, we

used average spectra of these oscillations to observe the activity and sympathovagal balance of the autonomic

nervous system. Then, we compared the average spectra obtained from both signals. The frequency contents

of the average spectra of SVV and SBPV signals to levosimendan treatment of verapamil-poisoned rats were

analyzed and related to the activity of sympathetic and parasympathetic tones. In control group, average

spectra of SVV and SBPV exhibited a low-frequency band (LF: 0.03− 0.8 Hz) peaked at ∼ 0.4 Hz and a high-

frequency (HF: 0.8 − 3.0 Hz) peaked at ∼ 1 Hz. LF peak was abolished after verapamil infusion. The LF

component of both spectra was observed to recover after continuous infusion of levosimendan. Additionally,

a new frequency component was observed at 1.5 Hz in the spectrum of SBPV. Signiﬁcant correlations were

found between bands of the average spectra in both signals in all groups of treatment studied in this paper. Our

results revealed that, like SBPV, SVV can herald useful information regarding the sympathovagal balance and

cardiac output improvements.

1 INTRODUCTION

Power spectra analyses of blood pressure and heart

rate variability (HRV) signals have previously been

used as a window to monitor the autonomic ner-

vous activity in the cardiovascular system (see e.g.

(Julien et al., 2003; Cerutti et al., 1994) and the ref-

erences therein). Using power spectral techniques,

systolic blood pressure variability (SBPV) can be de-

constructed into variabilities at different frequencies.

These variabilities can be divided into respiratory and

vasomotor components in a similar way to that seen

with HRV signals. Previously, it has been shown that

vasomotor-related SBPV correlates well with vascu-

lar sympathetic function of α-adrenoceptors (Brown

et al., 1994).

In rats, SBPV exhibits low-frequency(LF), known

as Mayer waves (Julien et al., 2003; Brown et al.,

1994), and high-frequency (HF) oscillations. The LF

oscillations are related to the activity of sympathetic

nervous system and the HF one is related to respira-

tory activity to be mediated by parasympathetic ac-

tivity (Cerutti et al., 1994; Japundzic et al., 1990).

Recently, seeking new methods to provide a clini-

cally useful information about the autonomic nervous

system, power spectrum of stroke volume variability

(SVV) signal has been employed. Previous studies

(Liu et al., 2004; Siebert et al., 2004) have compared

the spectra responses of SVV and HRV in healthy

subjects.

In the present study, we analyzed the spectral

properties of the systolic blood pressure variability

411

H. AlOmari A., Graudins A. and V. Savkin A. (2009).

THE POWER SPECTRA RESPONSE OF STROKE VOLUME AND ARTERIAL BLOOD PRESSURE VARIABILITY SIGNALS TO AUTONOMIC

NERVOUS SYSTEM MODULATION OF THE HEART.

In Proceedings of the International Conference on Bio-inspired Systems and Signal Processing, pages 411-415

DOI: 10.5220/0001775804110415

 SciTePress

(SBPV) signal to assess the effects of levosimendan

on the autonomic nervous activity in verapamil poi-

soning in a rodent model and then compare it with the

results obtained from applying the same procedure to

the estimated stroke volume variability signal. To our

knowledge, there are no studies that have compared

the spectra of SBPV and SVV estimated from contin-

uous blood pressure recordings.

2 Materials and Methods

The methods used in surgery, instrumentation, treat-

ment and data collection for this research has been

described in detail elsewhere (Graudins et al., 2008).

To summarize, experiments were conducted on

healthy, anaesthetized, intubated, and ventilated male

Wistar rats weighing between 300 and 500 grams.

Jugular, femoral venous and carotid catheters were in-

serted. Continuous recordings of ECG, arterial blood

pressure was performedusing a PowerLab data acqui-

sition system and Chart Version 5.0 software (ADIn-

struments, Castle Hill, Australia). After an equilibra-

tion period, all rats were then administered verapamil

hydrochloride(AbbottAustralasia, Botany, Australia)

6µg/kg/h (IsoptinInjection)until systolic blood pres-

sure (SBP) fell to 50% of baseline values. Then, an-

imals were randomized to one of the following treat-

ment groups:

Group 1. Control: rats received a loading dose of

normal saline (1 mL) followed by a continuous infu-

sion (1 mL/h). n = 7 rats.

Group 3. Levosimendan: levosimendan (donated by

Abbott Australasia, Sydney, Australia) was adminis-

trated to rats by loading dose of 6µg/kg, then levosi-

mendan infusion at 0.4µg/kg/min. n = 6 rats.

3 SIGNAL PROCESSING AND

SPECTRUM ANALYSIS

The blood pressure signal was processed by an al-

gorithm that extracts the cyclical features of the sig-

nal such as SBP (P

), diastolic pressure (P

), pulse

pressure (P

), integrated mean blood pressure (MAP),

stroke volume (SV), cardiac output (CO), heart rate

(HR), systolic time (T

), and diastolic time (T

). To

summarize, ABP signal, sampled at 400 Hz, has to

pass through four stages of processing which are: 1)

low-pass ﬁltering to remove the high frequency noise,

2) a windowed and weighted slope sum function to

support and enhance the up slope of the arterial blood

pressure signal and remove the remainder of the sig-

nal, 3) a logic circuit that detects the edges of the pro-

cessed slope sum function, and 4) threshold and deci-

sion rule that is used to suppress the edges resulting

from the dicrotic notch and noise. To obtain SBPV

signal, SBP was detected from the continuous record-

ing of ABP signal as follows: after low-pass ﬁltering

of ABP signal, the signal is processed using the slope

sum function (Moody et al., 2003) deﬁned as follows:

∑

j= k−w

∆y

, (1)

where

∆y

(

∆x

if ∆x

> 0,

0 if ∆x

≤ 0.

(2)

Here w is the number of samples representing the

analyzing window of the ﬁrst derivative of the BP sig-

nal ∆x

= x

−x

j− 1

, and x

is the ABP signal. w is cho-

sen to be 32 samples which is equal to the upslope of

the ABP peaks. Pulse contour method (Kouchoukos

et al., 1970) was used to estimate beat-by-beat values

of SV.

The time series of SBPV and SVV signals were

extracted from the raw beat-by-beat ABP and divided

into epochs after down and equidistant sampling were

applied to the signals. For each block of discrete data,

power spectral density was calculated directly from

the signal itself using the FFT algorithm. Average

spectral densities of SBPV and SVV obtained from

all rats within each group was then calculated and

compared. For each data block, the spectral density

estimates were smoothed. In this study, the spectral

estimate was performed on 312 data blocks of 512

samples overlapping by a half to reduce the loss of

stability.For each data set, the linear trends were re-

moved to reduce their contributions to the LF compo-

nent. Average spectra were evaluated approximately

every 5 minutes for each block of data. Spectral esti-

mates were found in each groupbeforeand after drugs

infusion has started.

4 Results

The extracted physiological parameters observed

from control group (baseline), after verapamil poi-

soning, and after continuous infusion of levosimen-

dan are summarized in Table 1.

Figure 1 shows the average spectra of SBPV and

SVV signals obtained from control group before ad-

ministration of any of the study drugs. Spectra of

SBPV and SVV exhibited LF band (0.03 − 0.8 Hz),

which corresponds to the activity of the sympathetic

BIOSIGNALS 2009 - International Conference on Bio-inspired Systems and Signal Processing

412

Table 1: Cardiovascular parameters extracted form continuous recording of ABP signals. Values included are the average

values obtained within each group. ABP: arterial blood pressure.

Variable Control Verapamil Levosimendan

(mmHg) 124.138±8.218 56.149±4.086

∗

57.108±5.087

†

MAP (mmHg) 92.731±7.258 63.709±12.537

66.643±2.953†

SV (µL/beat) 271.178±5.43 141.773±2.583

182.178±0.827

CO (mL/min) 76.089±1.89 24.278±1.670

56.598±6.975

HR (beats/min) 328.393±6.12 262.503±4.366

∗

301.737±10.293

∗

Values shown as Mean±SD. Paired t-test

p < 0.05,

∗

p < 0.01, and

†

: not signiﬁcant.

part of the autonomic nervous system, and a HF one

(0.8 − 3) Hz. Two clear peaks were detected in each

frequency band; the HF peak at ∼ 1 Hz which syn-

chronized by respiration and related to the parasym-

pathetic tone, and the LF peak at ∼ 0.4 Hz which

is related to sympathetic nervous activity. Note the

dominance of the vagal tone in the non-stressed state

prior to the induction of verapamil toxicity. This is

evidenced by the HF component being larger than

the LF one. Signiﬁcant positive correlation between

spectra of SBPV and SVV are seen in both LF and

HF bands (ﬁgure 2).

0 0.5 1 1.5 2 2.5 3 3.5

200

400

600

mmHg

/Hz

0 0.5 1 1.5 2 2.5 3 3.5

Frequency (Hz)

/Hz

Figure 1: Average spectra for both signals SBPV (top

panel), and SVV (bottom panel) estimated from control

group prior administration of any drugs.

0 50 100 150 200 250

SBPV (LF)

SVV (LF)

0 100 200 300 400 500 600

SBPV (HF)

SVV (HF)

r = 0.967, p < 0.05

r = 0.988, p < 0.05

Figure 2: Correlation analysis between LF bands in SBPV

and SVV (top panel), and between HF band (bottom panel)

in control group.

Shock state was introduced by an overdose infu-

sion of verapamil. As shown in ﬁgure 3, this com-

pletely abolished the LF component and shift the HF

component to ∼ 0.9 Hz of both SBPV and SVV spec-

tral traces. The HF peaks in both spectra were en-

hanced suggesting that sympathovagal balance was

shifted to the parasympathetic predominance. A

highly signiﬁcant correlation (R

= 0.977, p < 0.05)

between the two frequency bands LF and HF in both

spectra was obtained.

0 0.5 1 1.5 2 2.5 3 3.5

500

1000

1500

2000

mmHg

/Hz

0 0.5 1 1.5 2 2.5 3 3.5

Frequency (Hz)

/Hz

Figure 3: Average spectra for both signals SBPV (top

panel), and SVV (lower panel) estimated after verapamil

infusion.

The shape of power spectra of SBPV and SVV

observed after the continuous infusion of levosimen-

dan were markedly altered (ﬁgure4). Clearly, com-

pared with spectra obtained immediately after the in-

duction of verapamil, the LF component of both spec-

tra was observed to recover with concomitant reduc-

tion in the HF band peak. These results suggest that

there is a recovery of sympathetic tone after the con-

tinuous infusion of levosimendan. This may be due

to the improvements of the contractility proﬁle of the

heart. Additionally, a new frequency component was

observed at ∼ 1.5 Hz in the spectrum of the SBPV

which does not appear in SVV spectrum. Signiﬁ-

cant correlation were noted between LF (R

= 0.943,

p < 0.05) and MF (R

= 0.978, p < 0.05) bands from

both spectra, while a poor correlation was noticed be-

tween HF (R

= 0.295, p = 0281) bands.

THE POWER SPECTRA RESPONSE OF STROKE VOLUME AND ARTERIAL BLOOD PRESSURE VARIABILITY

SIGNALS TO AUTONOMIC NERVOUS SYSTEM MODULATION OF THE HEART

413

0 0.5 1 1.5 2 2.5 3 3.5

100

200

300

mmHg

/Hz

0 0.5 1 1.5 2 2.5 3 3.5

Frequency (Hz)

/Hz

Figure 4: Average spectra for both signals SBPV (top

panel), and SVV (lower panel) estimated after a continuous

infusion of levosimendan.

5 DISCUSSION

Spectral analysis of HRV has been used to study the

sympathovagalbalance of the autonomic nervous sys-

tem due to therapeutic verapamil infusion in humans

after acute myocardial infarction (Pinar et al., 1998),

in hypertensive patients (Sahin et al., 2004), and nor-

mal humans (Fauchier et al., 1997).

In the present study, verapamil overdose resulted

in a sharp drop in systolic, diastolic, mean blood pres-

sure, SV, HR, and CO. Additionally, it completely

abolished the LF component and enhanced the HF

one in both spectra suggesting that verapamil has

an anti sympatholytic properties contributing to its

negative inotropic effects and its vasodilatory proper-

ties. Hemodynamically, compared with period maxi-

mal verapamil toxicity seen prior to the administra-

tion of levosimendan, it signiﬁcantly improved CO

with no improvements in blood pressure while sig-

niﬁcant improvements were noticed in SV and HR. In

this study, levosimendan produced improvements in

cardiac function in heart failure induced by verapamil

poisoning. The results of ourstudy correlate well with

the hemodynamicparameters reported by Graudins et

al. (2008) in verapamil poisoned rats. Levosimendan

helped restore the LF component and reduced the HF

component suggesting that both drugs restored sym-

pathovagal balance seen prior to the administration of

verapamil. This dominance of sympathetic tone may

be the reason for the improvements of the myocardial

muscle contractility which cause the improvementsin

cardiac output heralded by a new frequency compo-

nent at ∼ 1.5 Hz in both spectra of SBPV and in SVV.

6 CONCLUSIONS

Spectral analysis of SVV signal may provide, along

with SBPV, useful information to clinicians regarding

the activity of the autonomic nervous system, cardiac

output, and responses to therapies aimed at improving

hemodynamic stability in hypotension patients.

ACKNOWLEDGEMENTS

This work was supported in part by The Australian

Research Council. It is also supported by an Amer-

ican College of Medical Toxicology Antidotal Re-

search Grant. Levosimendan was kindly donated by

Abbott Australasia.

REFERENCES

Brown, D., Brown, L., Patwardham, A., and Randall, D.

(1994). Sympathetic activity and blood pressure are

tightly coupled at 0.4 hz in conscious rats. Am J Phys-

iol, 267(5 pt 2):R1378–84.

Cerutti, C., Berr´es, C., and Paultre, C. (1994). Baroreﬂex

modulation of blood pressure and heart rate variabil-

ities in rats: assessment by spectral analysis. Am J

Physiol, 266(35):H1993–H2000.

Fauchier, L., Babuty, D., Autret, M., Cosnay, P., and

Fauchier, J. (1997). Effects of verapamil on heart rate

variability in subjects with normal hearts. Am J Car-

diol, 80(9):1234–5.

Graudins, A., Najaﬁ, J., and Perry, M. (2008). Treatment of

experimental verapamil poisoning with levosimendan

utilizing a rodent model of drug toxicity. Clin Toxicol,

46(1):50–6.

Japundzic, N., Grichois, M., Zitoun, P., Laude, D., and El-

ghozi, J. (1990). Spectral analysis of blood pressure

and heart rate in concious rats:effects of autonomic

blockers. J Auton Nerv Syst, 30(2):91–100.

Julien, C., Chapuis, B., Cheng, Y., and Barr´es, C. (2003).

Dynamic interactions between arterial pressure and

sympathetic nerve activity: role of arterial barore-

ceptors. Am J Physiol Regul Integr Comp Physiol,

285(4):R834–41.

Kouchoukos, N., Sheppard, L., and McDonald, D. (1970).

Estimation of stroke volume in the dog by a pulse con-

tour method. Circulation Research, 26(5):611–23.

Liu, H., Yambe, T., Sasada, H., Nanka, S., Tanaka, A., Na-

gatomi, R., and Nitta, S. (2004). Comparison of heart

rate variability and stroke volume variability. Auton

Neuro: Basic and Clinical, 116(1-2):69–75.

Moody, G., Mark, R., Zong, W., and Heldt, T. (2003). An

open-source algorithm to detect onset of blood pres-

sure pulses. IEEE Computers in Cardiology, 30:259–

262.

BIOSIGNALS 2009 - International Conference on Bio-inspired Systems and Signal Processing

414

Pinar, E., Garc´ıa-Alberola, E., LIamas, C., Vicente, T.,

L´opez-Candel, J., Rojo, J., Fern´andez, R., and Vald´es,

M. (1998). Effects of verapamil on indexes of heart

rate variability after acute myocardial infarction. Am

J Card, 81(9):1085–9.

Sahin, I., Kosar, F., Altunkan, S., and G¨unaydin, M. (2004).

Comparison of the effects of amlodipine and vera-

pamil on autonomic activity in hypertensive patients.

Eur J Intern, 15(4):225–230.

Siebert, J., Drabik, P., Lango, R., and Szyndler, K. (2004).

Stroke volume variability and heart rate power spec-

trum in relation to posture changes in healthy subjects.

Med Sci Monit, 10(2):MT31–7.

THE POWER SPECTRA RESPONSE OF STROKE VOLUME AND ARTERIAL BLOOD PRESSURE VARIABILITY

SIGNALS TO AUTONOMIC NERVOUS SYSTEM MODULATION OF THE HEART

415