Smart Smooth Muscle Spring-Dampers

BY RICARDO L. ARMENTANO, JUAN GABRIEL BARRA, FRANCO MARTIN PESSANA, DAMIAN O. CRAIEM, SEBASTIAN GRAF, DANIEL BIA SANTANA, AND RAMIRO A. SANCHEZ

P
Smooth Muscle Smart Filtering Helps to More Efficiently Protect the Arterial Wall

ulsatile pressure in arteries produces mechanical oscillations. Highfrequency vibrations tend to produce mechanical structure injures. Vascular smooth muscle (VSM) could react modulating viscoelasticity to protect the arterial wall, filtering the highest harmonics component present in the large and rapid slope of blood pressure. The aim of this study was to evaluate the filtering performance exerted by VSM in the human common carotid artery (CCA) in normotensive (NT, smart springdampers turned on), hypertensive (HT, smart spring-dampers in action) and angiotensin converting enzyme (ACE) inhibitors-treated hypertensive patients (HT-treated, smart spring-dampers tuned), and in harvested human CCA segments (smart spring-dampers turned off). Human carotid arteries in vitro experiments (n = 14) and in vivo studies (n = 24) were performed, using adaptive modeling techniques to calculate mechanical impedance and creep (C ) and stress relaxation (SR ) time constants. This adaptive procedure was tested in vitro in harvested CCA mounted in a circulation mock. A confirmatory measure of damping was achieved by using the half-power bandwidth method ( fC ) derived from the pressure-diameter frequency dependence using Bode diagrams, i.e., a compliance transfer function (diameter/pressure). Energy dissipation was calculated from the imaginary part of this function. Low-pass frequency responses were verified with a flat plateau up to a relatively stable frequency corner fC in the Bode diagram of the complete third-order model. Simplified first-order model cutoff frequencies were 2.7, 2.8, and 3.0 Hz for NT, HT, and HT-treated, respectively, showing an interesting constancy between groups. Smooth muscle tonus proved to preserve fC as well as C . Energy dissipation in hypertensive patients (n = 12) three-folded NT values and tended to be restored in HT-treated by means of a decrease in SR . VSM acts as smart spring-dampers, dissipating high-frequency components that might have damaging effects. VSM alterations found in HT patients could have a mandatory directive of preserving dynamic range near NT values, suggesting that VSM modulates its degree of activation and/or the vessel wall remodeling in order to ensure a suitable protection role.
Overview

A structure subject to oscillatory deformation contains a combination of kinetic and potential energy. In the case of real structures, there is also an energy dissipation element per cycle of motion. The amount of energy dissipated is a measure of the structures damping level. All the energy is lost as pure damping once the load is removed. In this case, the stress is proportional to the rate of the strain, and the ratio of stress to strain rate is known as viscosity. These materials have no stiffness component, only damping. On the contrary, purely elastic materials permit all the stored energy to be restored during the unloading phase. Biological tissues in general, and arteries in particular, are not purely
62 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE
0739-5175/07/$25.002007IEEE JANUARY/FEBRUARY 2007

elastic, that is, they exhibit marked viscous behavior [1], [2]. Thus, whereas part of the energy stored by the arterial wall during elastic distension is fully restored, the remaining part of the energy corresponding mainly to the viscous deformation is dissipated within the arterial wall [1], [3][5]. During both physiological and pathological situations, VSM could play an important role in modulating arterial stretching, which is ultimately responsible of wall mechanics. Moreover, VSM tone modifies not only the viscous but also the elastic properties of arterial walls [3]. When a system has vibration or oscillation, the simplest second-order model includes elastic, viscous, and inertial terms. The frequency response of this model can be modified by both the elastic and viscous responses. We believe VSM adjusts both to probably restrain the frequency response and to prevent vascular walls from being injuredjust like the role of spring-damper parameters of a car for protection of the chassis and driver from knocks of the wheels. This protects the system, just like the role viscoelasticity plays in arterial walls against the large and rapid slope of blood pressure. A total characterization of arterial mechanic behavior expressed in terms of mechanical impedance, that is, elastic, viscous, and inertial properties, has been proposed based on invasive recordings of pressure and diameter waveforms [3]. An improved parametric modeling-identification approach taking into account these viscoelastic properties and assessing the arterial frequency response was further developed [6]. In previous works, we found that viscoelastic parameters followed the level of VSM activation [3], [4], [7], [8]. Also, high-frequency vibrations and accelerating oscillations were reported to produce structure injures in arterial walls [9], [10]. We suggest that smooth muscle cells, as smart viscoelastic springdampers, exert a protective effect against high-frequency stretching, adjusting energy dissipation. B The aim of this study was to evaluate the filtering performance exerted by VSM in the human CCA in NT (smart spring-dampers turned on), HT (smart spring-dampers in action), and ACE inhibitors-treated hypertensive patients (smart spring-dampers tuned) and in harvested human CCA segments (smart spring-dampers turned off).
Materials and Methods In Vitro Study

Each CCA segment (maintained at 4 C during 2448 h postmortem) was nontraumatically mounted on specifically designed cannulae of a circulating loop (in vitro system), immersed and perfused with oxygenated Tyrodes solution (37 C, pH = 7.4). The perfusion line consisted of polyethylene tubing and a reservoir chamber, powered by a hydraulic pump electronically regulated to allow fine adjustments of pumping rate, pressure values, and waveforms. Bath temperature and pressure waveforms were controlled by a computer. This experimental setup manages waveforms, controlling mean and diastolic values, with the aim of maintaining controlled mean blood flow bounding endothelial function. This permits the in vitro mimicking of the conditions observed in vivo for preserving mechanical homeostasis (Figure 1). Pressure was measured with a solid-state microtransducer (flat at 01,200 Hz, Konigsberg Instruments, Inc., Pasadena, California) laterally placed in the proximal cannula. The sensor measures lateral pressure, avoiding flow disturbances. The pressure microtransducer had been previously calibrated using a mercury manometer. The arterial diameter was measured with a pair of ultrasonic dimension gauges (5 MHz, 2 mm diameter) sutured to the adventitia of the vessel at 58 mm from the pressure microtransducer. Each crystal weighed 7.5 mg, and its mass was considered negligible. This distance was set to avoid disturbances between sensors and ensures a neglected spatial delay with respect to typical wave propagation velocity [2], [3], [7]. The transit time of the ultrasonic

D A

D A B

E E E C

F F

Tissue procurements agree with the guides of the transplant program of the National Organ and Tissue Bank of Uruguay. All procedures of vascular tissue procurement and processing conform to ethical and safety concerns for therapeutic use, including written consent. General exclusion criteria took into account the International Standards on Tissue Banks issued by the International Atomic Energy Agency. Donors were 2345 years old (mean 29.6). Right and left CCA taken from seven donors in brain death condition were procured with a surgical aseptic technique during multiple organ and tissue harvesting (approximately 1 h postmortem). Donors were considered suitable for transplantation, and no traces of atherosclerosis were detected in the harvested arterial segments.
IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Fig. 1. A view of the in vitro setup, consisting of (a) a closed circulating loop powered by a pulsate pump. During measurements, the instrumented arterial segment, mounted in (c) adaptable supports, is immersed in (b) a rectangular chamber containing Tyrodes solution. The flow (arrows) returns from (d) a reservoir chamber. Bath temperature and pressure waveforms are controlled by a computer. The insert panel shows a pair of (e) piezoelectric crystals used to measure arterial diameter and (f) the solid-state pressure microtransducer.

JANUARY/FEBRUARY 2007

63

During both physiological and pathological

situations, VSM could play an important role in modulating arterial stretching, which is ultimately responsible of wall mechanics.

signal between crystals (1,580 m/s) was converted into distance by means of a sonomicrometer (flat at 01,000 Hz, Triton Technology Inc. San Diego, California) and was registered as arterial diameter. Once placed in the organ chamber, the segments were allowed to equilibrate for a period of 15 min under a steady-state flow (150 mL/min), mean pressure ( 83 mmHg), and stretching rate (70 cycles/min). Pressure and diameter signals of approximately 1020 consecutive cycles were sampled every 5 ms. Pressure and pumping rate levels were chosen to be similar to those observed in normotensive patients.
In Vivo Noninvasive Study

The CCA pressure waveform was calibrated by assigning the brachial diastolic value to the minimum CCA value and the mean brachial waveform value to the CCA average value. We assumed that mean pressure does not change in large conduit arteries and that diastolic pressure (as opposed to systolic pressure) does not substantially differ between the brachial and the carotid artery [8]. A surface electrocardiogram was acquired and stored together with the diameter and pressure signals. The pressure and diameter waveforms were identified according to the QRS complex of the electrocardiogram.
Data Analysis

Twelve normotensive subjects [NT group: 50 13 years; six males; body mass index (bmi) 25 1 kg/m2 ] and 12 patients with mild-to-moderate essential hypertension (HT group, 51 7 years; six males; bmi 27 3 kg/m2 ; blood systolic/diastolic pressure between 140179/90109 mmHg) were included in the study. After a washout period of four weeks, the HT patients received the ACE inhibitor ramipril as antihypertensive treatment (510 mg/day during three months). All individuals gave informed consent for the study according to the guidelines of the institutional committee for human investigations of the Cardiology and Cardiovascular Surgery Institute of the Favaloro Foundation. Echographic studies were performed with a real-time Bmode ultrasound imager (ATL HDI 5000, Miami Lakes, Florida) [11]. The left CCA was examined with a 7.5-MHz probe, 3 cm proximal to the bifurcation of the vessel. The sound beam was adjusted perpendicular to the arterial surface of the far wall of the vessel to obtain two parallel echogenic lines corresponding to the lumen-intima and media-adventitia interfaces. Once the two parallel echogenic lines of the far wall were clearly visible on the monitor, along at least 1 cm of the segment to measure, a fixed image (end-diastolic electrocardiogram triggering) to assess intimamedia thickness (IMT) and a sequence of images to determine the instantaneous waveform of arterial diameter were acquired [11]. The image analysis involved automatic detection of the anterior and posterior walls interfaces (Itec System, Paris, France). This procedure was previously validated against the sonomicrometric technique [12]. CCA pressure waveforms were recorded with a tonometer (Millar Instruments Inc.) at the same site as the diameter wave immediately after the echographic recording [8], [11]. In the same way, tonometric pressure waveforms were registered at the level of the brachial artery. The instantaneous pressure waveforms were digitized every 1 ms. The brachial pressure waveform was calibrated using the respective systolic and diastolic values assessed by sphygmomanometry.
64 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

A computerized procedure, developed in our laboratory, was used to determine the pressure-diameter loops and to calculate the viscoelastic parameters [3], [6]. In vitro and noninvasive pressure and diameter waveforms were interpolated in time to obtain the same number of data points, allowing calculation of the averaged cardiac cycle. A single-beat parametric system identification approach using a linear autoregressive with exogenous input (ARX) model was performed to calculate the arterial compliance Bode plot [6]. The averaged beat was extracted from stable periods, in each case, to ensure a representative segment. Period deviations never exceeded 1%. Such procedure characterizes the whole pressure-diameter hysteresis loop by estimation of the viscoelastic behavior of the arterial wall dynamics providing their individual contribution through the model [13]. This adaptive procedure was tested in-vitro in harvested human carotid arteries mounted in a circulation mock (See Figure 1). An initial third-order model was chosen based on the Akaike information criterion statistic. Applying this model to pressure-diameter waveforms, creep (C ) and stress relaxation (SR ) time constants were calculated [13]. Time constants in creep experiments are given by the poles, while those appearing in stress relaxation experiments are defined by the zeros of the compliance Bode plots. The arterial wall can be described using three main parameters that resume the elastic, viscous, and inertial behavior. Accordingly, a second-order simplification was further tested to reduce the model order, verifying that the frequency remained without significant changes and concluding in a second-order differential equation more suitable for clinical purposes, H(s) = 1 D(s) = , P(s) Ms2 + s + E (1)

where M, , and E are the inertial, viscous, and elastic modules. This simple model leads to two alternative characteristic
JANUARY/FEBRUARY 2007

parameters, that is, the damping ratio ( ) and the natural frequency ( fn ), whose mechanical expressions are, 1 fn = 2 E . and = M 2 EM (2)

WD = 4 |Eimaginary | A,

(3)

The damping ratio (or normalized damping) is often presented as a percentage or fraction of a critical value, called critical damping. This critical value (2 EM) is the critical viscous damping for a system, defined as the smallest level of viscous damping at which the mass will exhibit no oscillation when displaced from equilibrium. A system is classified as underdamped if the damping ratio is lower than unity. Because artery wall volume does not change in vivo, assuming incompressibility of the wall [2], CCA mass was considered a more appropriate variable than IMT to examine vascular remodeling. Thus, CCA mass, expressed as mass per length (L) unit (g/cm), was calculated as L(R2 R2 ) , where wall density was assumed as e i = 1.066 g/cm3 [2], and Re and Ri are external and internal radii, respectively. Using viscous damping to model energy dissipation, we will underestimate damping at low frequencies and overestimates it at high frequencies [14]. To circumvent this problem and to leverage experimental data available, we used the complex modulus approach [14]. The real part (storage modulus, Ereal ) of this complex term relates to the elastic behavior of the material and defines its stiffness. The imaginary part (loss modulus, Eimaginary ), related to the materials viscous behavior, defines the energy dissipative ability of the material. The energy dissipation (WD ) in such a system during the heartbeat was computed as [14],

where |Eimaginary | is the mean magnitude of the loss modulus, and A is mean pulsatile cross-sectional area. The behavior of a viscoelastic material is better described using the hysteretic model, in which the damping is proportional to strain and is independent of rate. This is achieved by eliminating the viscous dashpot and representing the energy dissipation in the system by a complex spring element. A quantitative measure of damping is achieved by using the half-power bandwidth method ( fC ) derived from the pressurediameter frequency dependence using Bode diagrams, i.e., a compliance transfer function (diameter/pressure) largely used in human striated muscle [15]. Damping is proportional to the reciprocal of fC .
Statistical Analysis

All data are reported as mean standard deviation. The inherent effect of the VSM was assessed comparing the specimens studied in vitro (null tonus, smart spring-dampers turned off) versus NT subjects (normal tonus, smart spring-dampers turned on) using unpaired Students t-tests. The effect of hypertension was assessed comparing the values obtained in NT subjects versus those obtained in HT patients (smart spring-dampers in action) using unpaired Students t-tests. The effects of treatment in HT patients (smart spring-dampers tuned) were established by paired Students t-tests. When multiple comparisons were needed, an analysis of variance (ANOVA) followed by Bonferroni correction was performed. For all statistical analyses, p < 0.05 was adopted as a significant difference. All calculations were performed with SPSS software (version 10.0, Statistical Package for the Social Sciences).

160 CCA Pressure (mmHg) 140 120 100 80 60 40 20 0 Systolic 12 CCA Diameter (mm) 10 Diastolic Mean

NT 240 HT Wall Mass (mg/cm) HT-t In Vitro 200 180 160 140 120 100 Pulse 1.4 1.2 Heart Rate (Hz) 1.0 0.8 0.6 0.4 0.2 0.0 Systolic Diastolic Mean NT HT HT-t NT HT HT-t In Vitro 220

0.9 Wall Thickness (mm)

0.8

0.7

0.6

0.5

8 6 4 2 0

In Vitro

Fig. 2. Averaged CCA hemodynamic and morphological parameters.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

JANUARY/FEBRUARY 2007

65

Results

Figure 2 shows averaged CCA hemodynamic parameters for noninvasive and in vitro studies. Initial values for IMT in the HT group were significantly higher than in the NT group. These remained unchanged after ACE inhibition. Representative pressure-diameter loops among the different conditions are shown in Figure 3.
Smart Spring-Dampers Turned Off

The effect of vasomotor tone was studied under virtually isobaric conditions between NT and in vitro specimens.

150 140 130 120 110 100 90 80 70 60 50 7.0

Viscoelastic results are presented in Figure 4. The presence of VSM tone in NT subjects caused an elastic decrease (43%, p < 0.05) concomitant with increased arterial wall viscosity (112%, p < 0.01), modifying damping ratio (22%, p < 0.05), and dramatically decreasing fC (67%, p < 0.01). Figure 5 shows the derived viscoelastic constants. The presence of VSM tone was evidenced through a sixfold increase (p < 0.01) in the stress relaxation time constant shifting pressure-diameter loops to the left (Figure 3). Bode responses are shown in Figure 6, where the transfer function H(s) was assumed as a frequency compliance function from the complete third-order model. Low-pass frequencies responses can be verified with a flat plateau up to a relatively stable frequency corner fC .
Smart Spring-Dampers in Action

CCA Pressure (mmHg)

NT HT HT-t In Vitro

7.5

8.0 8.5 9.0 CCA Diameter (mm)

9.5

10.0

Fig. 3. Graphs showing a representative pressure-diameter loop obtained in each study situation.

Hypertension increased pressure, diameter, IMT, circumferential stress (37%, p < 0.01), and CCA wall mass per unit length (36%, p = 0.01; see Figures 2 and 3) concomitantly with M increase (40%, p < 0.01; see Figure 4), shifting a more steep pressure-diameter loop to the right (Figure 3). Both elastic (79%, p < 0.01) and viscous (82%, p < 0.01) behavior increased with hypertension as shown in Figure 4. At the same time, energy dissipation almost tripled the normal values (p < 0.01), whereas all damping parameters ( fC , C , ) remained constant (Figures 46).

250 200 (mmHg.s/mm) E (mmHg/mm) 150 100 50 0 NT HT HT-t 18 Natural Frequency (Hz) 15 12 9 6 3 0 NT HT In Vitro

16 14 12 10 8 6 4 2 0 NT HT HT-t 1.0 0.8 Damping Factor 0.6 0.4 0.2 0.0 HT-t In Vitro NT HT In Vitro M (mmHg.s2/mm)

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 NT HT HT-t In Vitro

HT-t

In Vitro

Fig. 4. Averaged biomechanical parameters obtained from application of the viscoelastic model. E, and M: elastic, viscous and inertial parameters.

66 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

JANUARY/FEBRUARY 2007

We suggest that smooth muscle cells,

as smart viscoelastic spring-dampers, exert a protective effect against high-frequency stretching, adjusting energy dissipation.

Smart Spring-Dampers Tuned

ACE inhibition had a modulating effect on local VSM, independently from its peripheral action, evidenced by a 35% diminution in SR (p < 0.05; Figure 5), restoring the pressure-diameter loop to the normal location (Figure 3). Wall elasticity (26%, p < 0.05), viscosity (27%, p < 0.01), and energy dissipation (41%, p < 0.01) decreased with ACE inhibitor treatment, maintaining enhanced damping parameters ( fC , C , and ; see Figures 46).
Discussion

Damping refers to the extraction of mechanical energy from a vibrating system usually by conversion of this energy into heat. Damping serves to control the steadystate resonant response and to attenuate traveling waves in the structure [5]. Passive supplemental damping strategies are well understood and are recognized as a means for mitigating the effects of dynamic loadings. Smart damping (also known as semiactive control) strategies are particularly promising in addressing many of the challenges to this technology, offering the reliability of passive devices but maintaining the versatility and adaptability of fully active systems [16]. In this sense, there is a strong similitude with the smart smooth muscle spring-damping mechanism such as pressure-dependent myogenic response, flow-dependent response mediated by endothelium, or vascular remodeling [17]. These adaptive mechanisms are also described to maintain circumferential wall stress, within a normal and

uniform level. It is well known that hypertensive wall thickening is one of the main compensatory mechanisms to preserve circumferential wall stress [18]. Moreover, arterial wall viscosity is associated with higher IMT, suggesting that media thickening might be related to the smooth muscle alterations found in hypertension [11]. In light of this finding, we postulate that VSM dissipates energy to avoid high-frequency components in the pressure wave to be transferred to vascular diameter. During hypertension, energy dissipation is highly increased, protecting arterial wall but producing an additional load to the heart. Besides, the VSM modulates its degree of activation and vessel wall remodeling, thus preventing highfrequency vibrations from damaging the wall constituents. Recent works argue that over 30 Hz, noticeable damaging effects can be evidenced in certain tissues [9], [10]. We considered that oscillations around and beyond 30 Hz might represent injurious high frequencies for arteries showing a progressive reduction of finger blood flow after exposure to vibrations of frequencies ranging from 31.5250 Hz [10]. In vitro tests of harvested human arteries, allowing a null tonus analysis, could imply some limitations: 1) the absence of surrounding tissues could modify energy estimation, 2) some possible differences in the pressure/flow levels and waveforms, including the absence of wave reflections, 3) the elimination of neural innervations mainly affects vascular smooth muscle (warranting null tonus state), whereas endothelin, nitric oxide, and other

120 100 Time Constant (ms) 80 60 40 20 0

Creep Stress Relaxation

40

30

WD (105 erg/cm2)
NT HT HT-t In Vitro

20

10

0 NT HT HT-t In Vitro

Fig. 5. Time constants obtained from the application of the model. WD : beat-to-beat energy dissipation.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

JANUARY/FEBRUARY 2007

67

Smart damping strategies are particularly

promising in addressing many of the challenges to this technology, offering the reliability of passive devices but maintaining the versatility and adaptability of fully active systems
autocrine, paracrine, and endocrine substances could be conditioned to play their respective roles. All these enumerated factors indicate that the present results should be carefully analyzed. In spite of these restrictions, pressure-diameter loops and their corresponding viscoelastic parameters were suitable for a proper comparison, as can be verified in Figure 3, where changes in VSM tone or structural properties shift the pressure-diameter relation, moving the pressure-diameter loop to the right or to the left. The extension of a third-order model allows the inclusion of classic hysteresis, i.e., the behavior in the frequency domain beyond the corner frequency [13]. Hysteresis is usually demonstrated by the relationship between the internal pressure and diameter of a vessel segment. In experiments dealing with stress relaxation, the restoring force is recorded as a function of time, after a step change in length is applied. In studies regarding creep, the length of a strip of material is observed as a function of time, after a stepwise force is applied. Creep and stress relaxation time constants from our in vitro CCA specimen experiments were very close to those reported by Westerhof and Noordergraaf [13], suggesting that noninvasive technology may modify time constants values but maintains the same order of magnitude. However, some limitations of the noninvasive method should be issued. Pressure and diameter cycles were reduced to an averaged representative period. As the selected periods were stable for all cases, no significant differences were detected with respect to a beat-to-beat identification. Regarding pressure calibration, diastolic and mean pressures were matched to the tonometric waveform, assuming they remain constant throughout the arterial tree and to reduce all possible deviations [8], [12]. Representing the mechanical wall system by a secondorder differential equation, two cutoff frequencies emerged. Because fC2 (related to inertial effects) appears to be larger than fC1 and is beyond the physiological range, we can assume that a first-order system may be enough to represent arterial wall mechanics. In this sense, in a firstorder model, the cutoff frequency is only governed by the quotient E/, fC1 rst order system) = 1 E . 2

Our mean values calculated for fC1 were 2.7, 2.8, and 3.0 Hz for NT, HT, and HT-treated, respectively. They were very close to our fC values derived from the 3-dB method and showed an interesting constancy between groups (Figure 6). Because fC1 is always lower than fn (Figure 4), the influence of viscoelasticity may attenuate the high-frequency pulsation to counterbalance possible instability phenomena [19]. At low frequency, in the range of arterial sys5 tem acts as in Windkessel [20], pressure-flow waveforms are governed by total arterial compli0 ance and total peripheral resistance. But beyond fc 5 this range, both waveforms are influenced by pulse wave propagation and reflection, recently NT 10 characterized as negative feedback [20]. Working HT HT-t with the apparent compliance framework pro15 In Vitro posed by Quick et al., [20], higher dynamic ranges 20 than arterial impedance were found, including lumped and spatial distributed characteristics, and 25 seem very close to the low-pass pattern observed in our compliance Bode plot. However, in our 30 case, it must be stated that the end point was the 35 characterization of a local functional unit where 101 100 101 102 the stretching-pressure relationship become tranFrequency (Hz) scendent since wall fatigue or wall protection mechanisms are evidenced. The effect of blood pressure on measurements Fig. 6. A schematic representation of normalized Bode plots obtained from the application of the viscoelastic model. Circles indicate the aver- must be considered. In previous works, we anaage value of the half power bandwidth (f C) calculated in each group. lyzed in detail the influence of pressure on the The reference values for normalization were 39.8 dB in NT, 44.8 dB in viscoelastic properties [3], [4], [7], [8]. Accordingly, both arterial wall viscosity and HT, 42.2 dB in HT-treated, and 44.71 dB in vitro.
Compliance (dB)
JANUARY/FEBRUARY 2007

68 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

elasticity are strongly influenced by changes in the local smooth muscle status, independently of the prevailing pressure level. Besides, pressure levels can be more easily compared in Figure 3, which presents the corresponding pressure-diameter loops. In vitro measurements were designed to match normotensive pressures in a virtual isobaric comparison. The treated group shows readably the reversion process towards the normotensive group. Nevertheless, considering wall viscosity as a pressure-independent parameter of the VSM status [4], we would be able to conclude that the different viscous behaviors (and energy dissipation) found in HT and HT-treated are determined by intrinsic changes in the arterial wall. Local alterations of E cause pulse pressure augmentation. In the pressure waveform, this hyperpulsatility is manifested by a higher systolic level and by a steeper slope at its onset. The pulsatile work of the left ventricle is increased. The hyperpulsatility also increases the frequency content in the dynamics of the arterial wall. The interrelationship between E and could help to preserve the arterial wall, avoiding mechanical degradation. After ACE inhibitor treatment, the wall viscosity and cross-sectional area were reduced, resulting in a dramatic decrease of WD (Figure 5). ACE inhibition forced a pressure lowering, and the arterial wall reacted by decreasing loss energy, suggesting that it could both diminish energy dissipation (maintaining local protection) and decrease additional load to the heart. The vascular system appears to preserve the half-power bandwidth (Figure 6), that is, the VSM modulates its degree of activation and/or the vessel wall remodeling to prevent the deleterious high-frequency components to damage wall constituents.
Conclusions

(PEDECIBA/Uruguay); the Fundacin Favaloro (Buenos Aires, Argentina); and the Fundacin Universitaria Dr. Ren G. Favaloro (Buenos Aires, Argentina). Ricardo L. Armentano is currently dean of the Facultad de Ingeniera, Ciencias Exactas y Naturales, of the Universidad Favaloro; Class 1 researcher of the Ministry of Education, Culture and Technology of the Repblica Argentina; and Grade 5 Investigator of the PEDECIBA (United Nations Program for the Development of Basic Sciences), within the project URU/84/002 of the Ministry of Education and Culture and the Universidad de la Repblica, Montevideo, Uruguay. He received the engineering degree in 1984, and by the end of 1994, he qualified as a doctor of the University of Buenos Aires in physiological sciences and in 1999 he obtained the Ph.D. degree from Universit de Paris VII Denis Diderot, for the Doctorat de Biomecanique: Mecanique de Systmes Biologiques. In 1994 he was appointed principal investigator of the Basic Science Research Institute of the Favaloro Foundation. Since then he has been the leader of the Project Arterial System Dynamics, to which he has dedicated most of his efforts. In 1996, he was appointed director of the Master of Biomedical Engineering. In 2005, he was chosen director of the Ph.D. programs in signal processing in the National Technological University of Buenos Aires, Argentina. He was president of the Argentine Society of Bioengineering (1996-1999). In 2001, he was elected as a Senior Member and, in 2005, chair of the Argentine Chapter of the IEEE Engineering in Medicine and Biology Society . Juan Gabriel Barra received a D.V.M. degree and Ph.D. from the Universidad de Buenos Aires in animal physiology, in 1984 and 2002, respectively. He is preparing his thesis to obtain the M.Sc. degree in molecular biology and genetic engineering. He is currently an associate professor of the Department of Physiology, Biochemistry, and Pharmacology and a member of the Academic Council of the Faculty of Engineering and Exact and Natural Sciences of the Universidad Favaloro, where is head in the biological sciences area of the biomedical engineering career. His current research focuses on cardiovascular system dynamics and arterial hypertension. Franco Martin Pessana was born in Santa Fe, Argentina. He received the engineering degree in electronic engineering from the Mar del Plata National University (UNMdP- FI) and the M.Sc. degree in Biomedical Engineering from the Favaloro University (UF-FICEN), in Mar del Plata and Buenos Aires, Argentina, respectively. He is currently director of the Information Technology Department and Academic
JANUARY/FEBRUARY 2007

Pulsatile pressure stretches arteries, causing mechanical oscillations. VSM might act as smart spring-dampers, dissipating high-frequency components that might have damaging effects. This energy dissipation mechanism positively contributes to maintaining the cushioning exerted by the arterial wall, avoiding extra load to the heart in HT-treated. VSM alterations found in established hypertensive patients could have a mandatory directive of preserving dynamic range near NT values, despite the higher levels of arterial pressure present in the circulation, suggesting that VSM modulates its degree of activation and/or the vessel wall remodeling in order to ensure a suitable protection role.
Acknowledgments

The authors gratefully acknowledge the staff of Instituto Nacional de Donacin y Transplante de Clulas, Tejidos y rganos (INDT) of Repblica Oriental del Uruguay, and Dr. Alberto Crottogini for revising the manuscript. This article is dedicated to the memory of my father (Ricardo L. Armentano). This work was performed within a cooperation agreement between Universidad de la Repblica (Uruguay) and Universidad Favaloro (Argentina), and it was supported by BID 1201/OC-AR/PICT 14334 of the Secretara de Ciencia, Tecnologa e Innovacin Productiva (Argentina); the Programa para el Desarrollo de las Ciencias Bsicas
IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

69

Secretary in the Engineering Faculty of Favaloro University and is a qualified professor and researcher there. His main fields of research are focused in electronics, cardiovascular dynamics, arterial vasomotricity, digital signal processing, system modelization simulation, and tissue rheology. He is currently devoted to his doctorate program in arterial biomechanics to be presented in the Buenos Aires University. Damian O. Craiem was born in Buenos Aires, Argentina. He received the engineering degree in electronic engineering from the National Technologic University (UTN-FRBA) and the M.Sc. degree in biomedical engineering from the Favaloro University, both in Buenos Aires, Argentina. He is currently a qualified teacher and researcher in both universities. His research interests are in electronics, biomedical instrumentation, simulation, cells, and tissues rheology. He is currently devoted to his doctorate program in arterial mechanics to be presented in the Buenos Aires University. Sebastin Graf received the bachelor degree in biomedical engineering from the Universidad Nacional de Entre Ros, Argentina in 1995. Currently, he is completing his Ph.D. program in physiological sciences at the Buenos Aires University. He is a professor and head of the Electronic Department at the Favaloro University and a researcher Class 3 of the Ministry of Education and Culture and Technology of the Repblica Argentina. He is also a research professor at the National Technological University of Buenos Aires. His research interests include cardiovascular engineering, biomedical signal and image processing, and noninvasive biomedical devices. He is a Member of the IEEE Engineering in Medicine and Biology Society and a member of the Argentinean Cardiology Society and the Argentinean Bioengineering Society (SABI). Daniel Bia Santana received the M.Sc. degree in biological sciences from the PEDECIBA Program (United Nations Program for the Development of Basic Sciences) of Universidad de la Repblica in Uruguay. He is currently a researcher and teacher in the Physiological Depart\ment of the School of Medicine of Uruguy. His research interests are in arterial and vein physiology, vascular grafts, and tissue rheology. He is now dedicated to his doctorate project in vascular cryopreserved/defrosted homografts biomechanics and functional capability (PEDECIBA-Uruguay). Ramiro A. Snchez is the head of Hypertension and Metabolic Unit of the Favaloro Foundation and is a member of the Scientific Council, International Society of Hypertension (WHF), and of the American Society of Hypertension, the European Society of Hypertension, the International Society of Hypertension, the Argentina Society of Cardiology, and the Argentina Society of Nephrology. His main research interests include: smooth muscle energy dissiption in large arteries: the effect of blocking renin angiotensin system; silent brain white matter lesions in older

hypertensives: diferential effect of treatment; and nonmodulating salt sensitive hypertension: evidence for the involvement of kallikrein/kinin activity associated with overactivity of the renin-angiotensin system. Address for Correspondence: Ricardo Luis Armentano, Facultad de Ingeniera y Ciencias Exactas y Naturales, Av. Belgrano 1723, (C1093AAS), Buenos Aires, Argentina. Phone: +54 11 4378 1117. Fax: +54 11 4384 0782. E-mail: [email protected].
References
[1] C.D. Bertram, Energy dissipation and pulse wave attenuation in the canine carotid artery, J. Biomech., vol. 13, no. 12, pp. 10611073, 1980. [2] W.W. Nichols and M.F. ORourke, McDonalds Blood Flow in Arteries. Theoretical, Experimental and Clinical Principles. London, UK: Edward Arnold, 1998. [3] R.L. Armentano, J.G. Barra, J. Levenson, A. Simon, and R.H. Pichel, Arterial wall mechanics in conscious dogs. Assessment of viscous, inertial, and elastic moduli to characterize aortic wall behavior, Circ. Res., vol. 76, no. 3, pp. 468478, 1995. [4] D. Bia Santana, J.G. Barra, J.C. Grignola, F.F. Gines, and R.L. Armentano, Pulmonary artery smooth muscle activation attenuates arterial dysfunction during acute pulmonary hypertension, J. Appl. Physiol., vol. 98, no. 2, pp. 605613, 2005. [5] T.E. Kelly, In-structures damping and energy dissipation, Design Guidelines. Wellington: New Zealand: Holmes Consulting Group, 2001. [6] L.G. Gamero, R.L. Armentano, J.G. Barra, A. Simon, and J. Levenson, Identification of arterial wall dynamics in conscious dogs, Exp. Physiol., vol. 86, no. 4, pp. 519528, 2001. [7] J.G. Barra, J. Levenson, R.L Armentano, E.I. Cabrera Fischer, R.H. Pichel, and A. Simon, In vivo angiotensin II receptor blockade and converting enzyme inhibition on canine aortic viscoelasticity, Am. J. Physiol., vol. 272, no. 2, pp. H859H868, 1997. [8] R. Armentano, J.L. Megnien, A. Simon, F. Bellenfant, J. Barra, and J. Levenson, Effects of hypertension on viscoelasticity of carotid and femoral arteries in humans, Hypertension, vol. 26, no. 1 pp. 4854, 1995. [9] B.D. Curry, J.L. Bain, J.G. Yan, L.L. Zhang, M. Yamagouchi, H.S. Matloub, and D.A. Riley, Vibration injury damages arterial endothelial cells, Muscle Nerve, vol. 25, no. 4, pp. 527534, 2002. [10] M. Bovenzi, C.J. Lindsell, and M.J. Griffin, Acute vascular responses to the frequency of vibration transmitted to the hand, Occup. Environ. Med., vol. 57, no. 6, pp. 422430, 2000. [11] R.L. Armentano, S. Graf, J.G. Barra, G. Velikovsky, H. Baglivo, R. Sanchez, A. Simon, R.H. Pichel, and J. Levenson, Carotid wall viscosity increase is related to intima-media thickening in hypertensive patients, Hypertension, vol. 31, no. 1, pp. 534539, 1998. [12] S. Graf, J. Gariepy, M. Massonneau, R.L. Armentano, S. Mansour, J.G. Barra, A. Simon, and J. Levenson, Experimental and clinical validation of arterial diameter waveform and intimal media thickness obtained from B-mode ultrasound image processing, Ultrasound Med. Biol., vol. 25, no. 9, pp. 13531363, 1999. [13] N. Westerhof and A. Noordergraaf, Arterial viscoelasticity: a generalized model. Effect on input impedance and wave travel in the systematic tree, J. Biomech., vol. 3, pp. 357379, 1970. [14] K.K. Varanasi, Models of damped systems, in On the Design of a Precision Machine for Closed-Loop Performance, ch. 3, M.S. thesis, MIT, Cambridge, MA, 2002. [15] A. Desplantez, C. Cornu, and F. Goubel, Viscous properties of human muscle during contraction, J. Biomech., vol. 32, no. 6, pp. 555562, 1999. [16] B.F. Spencer Jr. and M.K. Sain. Controlling building: A new frontiers in feedback, IEEE Control Syst. Mag., vol. 17, no .6, pp. 1935, 1997. [17] P.F. Davies, J. Zilberberg, and B.P. Helmke, Spatial microstimuli in endothelial mechanosignaling, Circ. Res., vol. 92, no. 4, pp. 359370, 2003. [18] P. Fridez, A. Makino, H. Miyazaki, J.J. Meister, K. Hayashi, and N. Stergiopulos, Short-term biomechanical adaptation of the rat carotid to acute hypertension: contribution of smooth muscle, Ann. Biomed. Eng., vol. 29, no. 1, pp. 2634, 2001. [19] G. Pontrelli and E. Rossoni, Numerical modelling of the pressure wave propagation in the arterial flow, Int. J. Numer. Meth. Fluids, vol. 43, no. 6, pp. 651671, 2003. [20] C.M. Quick and D.S. Berger. Arterial pulse wave reflection as feedback, IEEE Trans. Biomed. Eng., vol. 49, no. 5, pp. 440445, 2002.

70 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

JANUARY/FEBRUARY 2007