Contribution of Collagen, Elastin, and Smooth Muscle to In Vivo Human Brachial Artery Wall Stress and Elastic Modulus
Background The contributions of collagen, elastin, and smooth muscle to arterial mechanical properties in the in vivo human artery are not known.
Methods and Results We used a recently developed intravascular ultrasound technique to measure total brachial artery wall stress and incremental elastic modulus (Einc) in seven normal human subjects at baseline and after intra-arterial norepinephrine (1.2 μg) and nitroglycerin (100 μg). Then we applied a modified Maxwell model to estimate the elastic modulus of elastin (EE); the recruitment of collagen fibers supporting wall stress; and the differential contributions of collagen, elastin, and smooth muscle to wall stress and Einc over a wide range of pressure and smooth muscle tone. With this model, EE was 3×106 dynes/cm2. Collagen fibers were recruited increasingly as transmural arterial pressure increased and reached a value of ≈5% to 6% at 100 mm Hg under each of the conditions studied. Isobaric smooth muscle contraction resulted in a small decrease in total wall stress and no significant change in total Einc while shifting the predominant element contributing to these mechanical parameters from collagen in parallel with the smooth muscle to collagen in series with the smooth muscle. In contrast, isometric smooth muscle contraction produced large increases in total wall stress (from 0.11×106 dynes/cm2 after nitroglycerin administration to 1.35×106 dynes/cm2 after norepinephrine administration) and Einc (from 3.84×106 dynes/cm2 after nitroglycerin administration to 57.8×106 dynes/cm2 after norepinephrine administration) entirely as a result of the additional contribution of the smooth muscle and its associated series collagen.
Conclusions This study describes a technique for determining arterial elastic properties and a model that can be used to estimate a number of mechanical parameters of the human brachial artery in vivo. This technique may be useful in studies of the arterial elastic properties of arteries in patients with vascular pathology.
The arterial wall structures collagen, elastin, and smooth muscle bear the vast majority of wall stress and determine the stiffness of blood vessels. A number of in vitro studies with arterial wall strips or intact rings have been performed to assess the relative contribution of these three arterial wall components to stress and incremental elastic modulus (Einc).1 2 3 4 In most of these studies, the arterial wall is proposed to consist of a parallel elastic component (PEC) made up of collagen and elastin and a force-generating component of smooth muscle. A series elastic component (SEC) is hypothesized to be connected in series with either the smooth muscle itself (Maxwell model) or with the combined PEC and smooth muscle (Voigt model). Recently, the contribution of collagen, elastin, and smooth muscle to canine aortic wall stress and Einc has been assessed by application of a modified Maxwell model and the accompanying constitutive equations to stress-strain data acquired over a wide range of pressure and smooth muscle activity in vivo.5 6 This type of analysis has not been performed in human subjects in vivo, in part because no techniques have been available to measure arterial wall stress and Einc under the necessary conditions.
We have developed a technique to measure arterial wall elastic mechanics in human subjects in vivo over a wide range of pressure and smooth muscle tone.7 This technique used intravascular ultrasound to measure brachial artery cross-sectional area and wall thickness simultaneously with intra-arterial pressure. Inflation of a cuff overlying the segment of brachial artery being imaged reduces transmural brachial artery pressure and enables one to measure arterial elastic mechanics over a pressure range of ≈0 to 100 mm Hg. Furthermore, infusion of vasoactive drugs into the brachial artery in subsystemic doses produces changes in smooth muscle tone without changes in blood pressure and the associated indirect effects on arterial elastic properties. Our laboratory previously demonstrated the reliability and reproducibility of this technique and the direct effects of smooth muscle relaxation and contraction on brachial artery elastic mechanics.7 In the present study, we determined arterial stress, strain, and Einc at baseline and after changes in smooth muscle tone in normal human subjects. We then applied a modified Maxwell model to estimate the contributions of collagen, elastin, and smooth muscle to arterial wall stress and Einc under various levels of smooth muscle tone.
Seven normal male subjects 22 to 56 years of age (mean±SD, 42±5 years) were studied. They were recruited from newspaper advertisements and were free of heart disease, hypertension, hypercholesterolemia, diabetes mellitus, vascular disease, and other medical illness as assessed by history, physical examination, routine blood tests, and ECGs. None of the subjects smoked or was taking medications. Mean blood pressure was 136/78±6/2 mm Hg. Body weight was 79±3 kg, and forearm volume was 1122±71 mL. Written informed consent was obtained from all subjects. The study was approved by the Human Rights in Research Committee at the University of Minnesota.
Intravascular Ultrasound Technique
The intravascular ultrasound technique used to measure brachial artery elastic properties has been described in detail.7 The brachial artery was imaged with a commercially available intravascular ultrasound system (HP Sonos Intravascular Imaging System, Hewlett Packard Co) and a 3.5F or 3.0F catheter with a 30-MHz mechanical rotating transducer (SCIMED Life Systems, Inc). This catheter is 140 cm long and consists of a polyethylene shaft surrounding a central, moveable rotating transducer that produces a 360° tomographic image of the blood vessel. Images were obtained at 30 frames per second and at a radial depth display of 5 mm. Images were optimized by use of compression, time-gain compensation, and postprocessing controls to give the blood in the brachial artery lumen a slight degree of reflectivity. Axial resolution is ≈100 μm; lateral resolution is ≈150 μm (manufacturer's specifications).
The intravascular ultrasound catheter was placed in the brachial artery through a 3.5F or 4.0F arterial sheath with side arm. The catheter was positioned approximately at the middle of the upper arm and 5 to 6 cm past the end of the sheath. Adjustments of ultrasound catheter position, gain, and gray scale were made to obtain optimal images and were not changed over the course of the study. Intra-arterial pressure was measured through the side arm of the arterial sheath, which was connected to a pressure transducer. Arterial pressure waveforms were recorded on a four-channel chart recorder (Gould Inc), converted to a video signal, and overlaid onto the intravascular ultrasound video image with a video mixer (Panasonic). Thus, simultaneous intra-arterial pressure and intravascular images were displayed and recorded on videotape.
A 12-cm blood pressure cuff was placed around the upper arm and centered over the underlying intravascular ultrasound catheter. The cuff was connected to a compressed air system (Hokanson E-10, D.E. Hokanson) that was calibrated to a mercury manometer and used to rapidly inflate the cuff to any desired pressure. Transmural pressure was defined as intra-arterial pressure minus cuff pressure.
Video Image Analysis
Cross-sectional video images of the brachial artery were analyzed off-line to determine brachial artery area and wall thickness. Video frames were digitized by use of a frame grabber board (Data Translation). Digitized images were stored on optical disks and displayed on a Macintosh IIci computer. The software image analysis package NIH Image 1.52 was used to measure lumen area and wall thickness.
End-diastolic and mid-diastolic images and the simultaneously measured pressures were analyzed for each condition and external cuff pressure because we wanted to assess the elastic behavior of the arterial wall and not the viscous and inertial behavior that also is present during systole.5 The lumen-vessel wall border was manually traced, and the lumen cross-sectional area was automatically determined. Brachial artery area was measured and averaged for five consecutive images obtained 5 to 10 seconds after a change in cuff pressure to allow the vessel to equilibrate to the new conditions.
Arterial wall thickness was measured where a vein ran adjacent to the wall of the brachial artery. This location significantly improved our ability to discern the outer wall of the blood vessel because of the large difference in echo reflectivity of the wall tissue and the blood in the lumen of the adjacent vein. Ten measurements of wall thickness were made and averaged for each of three beats under baseline conditions with zero cuff pressure. We previously demonstrated low intraobserver and interobserver measurement variabilities using this technique.7 The vessel wall was assumed to be incompressible.3 8 9 Therefore, wall thickness for other conditions and cuff pressures could be calculated by use of the measured, constant, cross-sectional area of the wall for each individual.
Smooth Muscle Relaxation and Contraction
Smooth muscle activity was altered by administration of vasoactive drugs into the brachial artery sheath through the side arm. All drugs were administered as 5-mL boluses over ≈2 seconds. Norepinephrine (1.2 μg) was administered to produce smooth muscle contraction; nitroglycerin (100 μg) was given to produce smooth muscle relaxation. We previously demonstrated that these drugs at these doses cause local effects on the brachial artery (without changes in systemic pressure) that are sustained throughout the ≈2-minute measuring period.7 The dose of nitroglycerin was chosen to produce complete smooth muscle relaxation. Previous studies using intracoronary nitroglycerin10 and our previous data confirm that maximal or near-maximal smooth muscle relaxation is achieved with this dose.
Studies were performed in a quiet room of constant temperature (22°C to 23°C). Subjects refrained from caffeinated beverages and cigarettes on the day of the study. Forearm volume was measured with a truncated cone formula.11 An arterial sheath was placed into the nondominant brachial artery after sterile preparation and subcutaneous injections of lidocaine. The arm was positioned at the patient's side just above the level of the heart. Subjects rested quietly for 30 minutes. A sterile plastic sleeve was placed over the intravascular ultrasound catheter. The intravascular ultrasound catheter was placed through the sheath and positioned in the midbrachial artery beneath the center of the overlying blood pressure cuff.
Brachial artery images and intra-arterial pressure were recorded continuously throughout the study. The arm cuff was inflated by 10 mm Hg increments approximately every 10 seconds until the cuff pressure exceeded diastolic blood pressure. These procedures provided measurements of arterial elastic properties over a wide range of transmural pressure. Norepinephrine was then infused, and the above stepwise increase in cuff pressure was repeated 45 to 60 seconds after the drug was administered. After the brachial artery area had returned to its baseline area (≈15 minutes), nitroglycerin was administered, and the above sequence of measurements was recorded.
Transmural pressure versus area data points were plotted and fit to the formula of Langewouters et al,12 A=a [0.5+1/π tan−1 (P/c−b/c)], by use of nonlinear least-squares regression methods. In this formula, A is brachial artery area; P is transmural pressure; and a, b, and c are three independent parameters that characterize each pressure-area curve. Although logarithmic13 and other models have been used to describe the pressure-area relationship in human peripheral large arteries, we used the arctangent formula of Langewouters et al because all individual pressure-area data fit this arctangent model with r values >.90. Circumferential wall stress (σ) was calculated as where P is transmural pressure, rm is midwall radius, and h is wall thickness. Circumferential strain (ε) was calculated as where ro is the effective unstressed midwall radius (the midwall radius at 0 mm Hg transmural pressure) under baseline conditions. The reproducibility of ro measurements with this technique and the rationale for defining ro as described were discussed previously.7 Einc was calculated with the assumption of arterial isotropy, as was done by previous investigators3 14 15 : Data for all curves under each condition were averaged every 10 mm Hg between 0 and 100 mm Hg transmural pressure.
Arterial Wall Model
Models have been used to describe skeletal,16 17 cardiac,18 and smooth2 4 muscle mechanics in numerous studies of isolated muscle strips or rings. Recently, arterial wall models have been applied to studies of elastic mechanics in conscious dogs.5 6 In the present study, we used a modified Maxwell model (Fig 1⇓) to characterize the arterial elastic mechanics of the in vivo human brachial artery. In this model, the arterial wall consists of a contractile element, a PEC, and an SEC. The PEC represents the elastic behavior of the vessel wall when the contractile element is completely relaxed. The PEC is made up of one spring, which represents elastin, and a number of much stiffer springs of different lengths, which represent collagen. This model of collagen fibers is based on the disconnecting hook model of Wiederhielm.19 At very low transmural pressures, the wall stress is borne almost exclusively by elastin.20 As the fully relaxed vessel is stretched, increasing numbers of stiffer collagen fibers are recruited.
The contractile element is the smooth muscle. The model assumes that the smooth muscle offers no permanent resistance to stretch when relaxed. The smooth muscle is coupled in series with the SEC. In this model, the SEC is made up of collagen fibers of differing lengths that are increasingly recruited in proportion to smooth muscle activation. When the contractile element is activated, the combined smooth muscle–SEC increases isometric wall stress and wall stiffness.
On the basis of this model, the following equations can be formulated to estimate the contribution of various wall elements to arterial elastic properties. In these equations, the individual wall elements are assumed to behave in a hookean manner so that there is a linear relationship between stress and strain. Total circumferential wall stress (σT) is the sum of the individual wall stresses of the three parallel elements:|<|\sigma|>|_|<|T|>||<|=|>||<|\sigma|>|_|<|E|>||<|+|>||<|\sigma|>|_|<|C(p)|>||<|+|>||<|\sigma|>|_|<|SM-C(s)|>|where σE is the stress supported by elastin, σC(p) is the stress supported by parallel collagen, and σSM-C(s) is the active stress supported by the smooth muscle and collagen in series. The stress supported by the parallel elastic element (σPE) can be determined from the nitroglycerin stress-strain curve because the smooth muscle is assumed to be nearly completely relaxed under this condition:|<|\sigma|>|_|<|PE|>||<|=|>||<|\sigma|>|_|<|T(NTG)|>||<|=|>||<|\sigma|>|_|<|E|>||<|+|>||<|\sigma|>|_|<|C(p)|>|
We have previously demonstrated that the assumption of nearly complete smooth muscle relaxation with the dose of nitroglycerin given in this study is reasonable.7
The elastic modulus of elastin (EE) can be determined from the slope of the nitroglycerin stress-strain curve under 0 mm Hg transmural pressure because elastin has been shown to support nearly all the wall stress under very low pressures.20 The contribution of elastin to total stress in the fully relaxed vessel can thus be determined by the following equations:|<|\Delta|>||<|\sigma|>|_|<||<|E|>||>||<|=|>|E_|<|E|>||<|\times|>||<|\Delta|>||<|\epsilon|>|_|<||<|E|>||>||<|\sigma|>|_|<||<|E|>||>||<|-|>|0|<|=|>||<|\sigma|>|_|<||<|E|>||>||<|=|>|E_|<|E|>||<|\times|>|(|<|\epsilon|>||<|-|>||<|\epsilon|>|_|<|0E|>|)where ε is the vessel strain and ε0E is the strain-axis intercept for the fully relaxed vessel. In the fully relaxed vessel, the stress borne by collagen is the total stress (σPE) minus the stress borne by elastin:|<|\sigma|>|_|<|C(p)|>||<|=|>||<|\sigma|>|_|<|PE|>||<|-|>||<|\sigma|>|_|<|E|>|
The parallel collagen fibers are increasingly recruited to bear wall stress as pressure increases. This recruitment function of parallel collagen fibers [fC(p)] can be calculated from the elastic modulus of collagen (EC; ≈1×109 dynes/cm2),20 EE, and the Einc for the fully relaxed vessel (EPE):|<|E|>|_|<|PE|>||<|-|>|E_|<|E|>||<|=|>|E_|<|C|>||<|\times|>|f_|<|C(p)|>|The recruitment function for the series collagen fibers [fC(s)] can be determined in a manner analogous to that for the parallel collagen fibers if the stiffness of the combined smooth muscle and series collagen is assumed to be predominantly attributable to the collagen:E_|<|SM-C(s)|>||<|=|>|E_|<|C|>||<|\times|>|f_|<|C(s)|>|
The computation of active stress [σSM-C(s)] under different conditions of smooth muscle activity can be made by subtracting the passive stress generated by the PEC from the total stress at the same level of strain:|<|\sigma|>|_|<|SM-C(s)|>||<|=|>||<|\sigma|>|_|<|T|>||<|-|>||<|\sigma|>|_|<|PE|>|
Similarly, the elastic modulus of the combined smooth muscle and series collagen under different conditions of smooth muscle activity can be calculated as the difference between the measured total Einc and the EPE at constant strain:E_|<|SM-C(s)|>||<|=|>|E_|<|T|>||<|-|>|E_|<|PE|>|
Curves generated at baseline and after drug administration were compared by use of ANOVA with two repeated variables (drug and pressure). Statistical significance was accepted at a value of P<.05. Data are presented as mean±SEM.
Fig 2⇓ shows stress versus strain data. The addition of smooth muscle contraction under baseline and norepinephrine conditions to the nearly fully relaxed nitroglycerin vessel produced significant (P<.01) leftward shifts of the stress-strain curves so that wall stress for a given level of strain increased as smooth muscle contraction increased. EE, obtained by measuring the slope of a tangent to the nitroglycerin stress-strain curve at 0 mm Hg transmural pressure, was 3.04×106 dynes/cm2. Smooth muscle contraction under baseline and norepinephrine conditions produced similar significant (P<.01) leftward shifts in the Einc-strain curves (Fig 3⇓).
Fig 4⇓ shows total wall stress and the arterial wall components that bore wall stress over the 100 mm Hg pressure range for the three levels of smooth muscle tone. Total isobaric wall stress increased in direct proportion to the amount of smooth muscle relaxation owing to increases in radius and decreases in wall thickness after vasodilation. Total wall stress (106 dynes/cm2) at 50 and 100 mm Hg transmural pressure was 0.63±0.06 and 1.34±0.10 after norepinephrine administration, 0.74±0.05 and 1.56±0.10 at baseline, and 0.86±0.07 and 1.82±0.14 after nitroglycerin administration. Over the range of measurements made in this study, the wall elements bearing stress varied greatly with the level of smooth muscle activation. In the fully relaxed (nitroglycerin) vessels, wall stress was borne completely by parallel collagen and elastin. At most pressures under baseline and norepinephrine conditions, elastin continued to bear a small proportion of the wall stress. However, parallel collagen bore little stress; instead, smooth muscle and collagen in series with the smooth muscle bore most of wall stress.
Fig 5⇓ shows Einc versus pressure data and the arterial wall components that contributed to Einc. Einc (106 dynes/cm2) at 50 and 100 mm Hg transmural pressure was 24.5±8.2 and 57.8±16.8 after norepinephrine administration, 23.2±5.3 and 60.8±11.8 at baseline, and 19.9±3.1 and 58.6±9.0 after nitroglycerin administration. Alterations in smooth muscle tone did not significantly change total isobaric Einc, as we showed previously.7 At all strains equal to or greater than the strain at 0 mm Hg pressure after nitroglycerin administration, elastin contributed a constant amount to total Einc. Similar to the stress versus pressure data, smooth muscle activation shifted the predominant wall elements contributing to Einc from parallel collagen (in the nitroglycerin vessels) to smooth muscle and series collagen.
Fig 6⇓ demonstrates isometric wall stress and Einc at various levels of smooth muscle tone. Isometric data were obtained at a strain of 1.08 because a vertical line through this point on the stress-strain graph (Fig 2⇑) intersects all three curves. Isometric smooth muscle contraction increased wall stress from 0.11×106 dynes/cm2 after nitroglycerin administration to 0.28×106 dynes/cm2 under baseline conditions and 1.35×106 dynes/cm2 after norepinephrine administration. Similarly, Einc increased from 3.84 to 9.92 and 57.80×106 dynes/cm2, respectively, with increasing isometric smooth muscle contraction. According to the arterial wall model, the contributions of the PEC elements (elastin and parallel collagen) to wall stress and Einc are identical under isometric conditions. Therefore, the large increases in stress and Einc that occurred with isometric smooth muscle contraction were due entirely to the generation of tension in the smooth muscle and its associated SEC.
Fig 7⇓ shows the recruitment of collagen fibers as a function of pressure for the three levels of smooth muscle tone. Recruitment of collagen fibers increased in a curvilinear manner for each of the conditions. Between 5% and 6% of the total collagen fibers were recruited at 100 mm Hg for each condition according to our model. With the smooth muscle completely relaxed, only parallel collagen fibers were recruited. As smooth muscle contraction increased, there was a shift from recruitment of parallel collagen fibers to the recruitment of collagen fibers in series with the smooth muscle.
The goals of this study were to apply a model of the arterial wall to the measured mechanics of the normal in vivo human brachial artery and to estimate the contributions of collagen, elastin, and smooth muscle to arterial wall stress and stiffness. To accomplish this, an intravascular ultrasound technique was used that enabled us to determine arterial elastic mechanics in the brachial artery over a wide range of pressure and smooth muscle tone. Using a modified Maxwell model for the arterial wall, we developed equations describing the contributions of individual wall components to the measured elastic parameters and depicted how these contributions changed as a result of altered smooth muscle activity. Isometric smooth muscle contraction produced large (10- to 15-fold) increases in circumferential wall stress and Einc. Isobaric smooth muscle contraction also shifted the predominant wall element contributing to wall stress and Einc from collagen in parallel with the smooth muscle to collagen in series with the smooth muscle.
Brachial Artery Wall Model
Models of muscle have been used extensively to describe muscle mechanical function, beginning with Hill's16 concept of muscle as a system of elements whose mechanical properties could be described in terms of elasticity and viscosity. Models of the arterial wall have mostly been applied to studies of isolated vessel strips or rings, although some recent studies have been performed in vivo in conscious dogs.5 6 To apply most arterial wall models to in vivo data, one must be able to determine arterial elastic mechanics (1) under complete or near-complete smooth muscle relaxation, (2) under low (or zero) wall stress, (3) after smooth muscle contraction, and (4) over a wide pressure range. Near-complete smooth muscle relaxation is necessary so that the PEC mechanics can be ascertained. Measurement of arterial elastic mechanics under low wall stress is necessary so that the contribution of elastin (which bears most of the PEC wall stress under low stress states)20 can be separated from the contribution of parallel collagen. Once the contributions of elastin and collagen to the PEC are determined at each level of strain, the contribution of smooth muscle and its associated SEC can be calculated as the difference between the total stress or Einc and that of the PEC under isometric conditions. Smooth muscle contraction of a muscular artery produces large changes in arterial cross-sectional area. If arterial elastic mechanics were measured only between diastolic and systolic pressures, isometric properties under significantly different levels of smooth muscle tone could not be determined because there would be no strain value common to all conditions. By using a pressurized cuff to reduce transmural arterial pressure and intra-arterial drug infusions to alter smooth muscle tone, we were able to satisfy the above requirements and determine total and constitutive isometric wall stress and Einc for the first time in human arteries in vivo.
The PEC in our model consists of elastin and collagen. This composition of the PEC is based on data demonstrating that elastin bears most of the wall stress at low transmural pressure and collagen bears an increasing proportion of the wall stress as transmural pressure (and strain) rises.20 21 EE was defined as the slope of the stress-strain curve of the fully relaxed vessel at 0 mm Hg transmural pressure. EE calculated from our data was 3.0×106 dynes/cm2. This value is within the range of values that we obtained in a previous study (1.0×106 dynes/cm2) and those determined in vitro by Cox1 (2.8×106 dynes/cm2) and in vivo by Armentano et al5 (4.9×106 dynes/cm2) and Barra et al6 (5.0×106 dynes/cm2). Because elastin is presumed to be a hookean material (the stress-strain relationship is linear), the contribution of elastin to total Einc was identical at all strains greater than or equal to the strain of the fully relaxed (nitroglycerin) vessel at zero stress.
The fraction of collagen fibers within the PEC recruited to support wall stress at each pressure was calculated as Einc of the fully relaxed vessel minus EE divided by EC. Using this model, we calculated that ≈5% to 6% of the collagen fibers were recruited at 100 mm Hg. This human in vivo value is in general agreement with data obtained in animal blood vessels with a similar arterial wall model. Armentano et al5 found that 6.1% of the collagen fibers were recruited at an arterial pressure of 118 mm Hg in the in vivo canine aorta, and Cox1 calculated 10% recruitment of collagen fibers at an arterial pressure of 100 mm Hg for intact cylindrical segments of muscular canine arteries.1
The SEC has been defined as the algebraic sum of all the undamped compliances that are coupled in series with the force-generating apparatus.4 The SEC is thus stretched whenever the smooth muscle contracts isometrically. The morphological basis of the SEC is not known. Although we could not measure the mechanical characteristics of the SEC alone with the present technique, our data may provide some insight into this component of the arterial wall. Details regarding connective tissue–smooth muscle interactions are not well described. However, there is anatomic evidence for connections between smooth muscle cells and collagen.22 In addition, smooth muscle cells are surrounded by an intricate meshwork of collagen.23 Burton24 hypothesized that the SEC was collagen. In our model, collagen was placed in series with the smooth muscle. Our data and data from other studies of muscular arteries are consistent with this arrangement. The active stress-strain relationship is curvilinear, concave to the stress axis, and similar in shape to the passive stress-strain curve. Force-extension curves for the SEC in vertebrate smooth muscle also appear to be curvilinear and concave to the force axis.25 Increasing recruitment of parallel collagen fibers with increasing stress or strain has been shown to be responsible for the shape of the passive stress-strain curve.20 Increasing recruitment of parallel collagen with increasing stress or strain cannot explain the shape of the active stress-strain curve because the large decrease in vessel cross-sectional area with norepinephrine ensures that most if not all of the parallel fibers are slack, or not engaged, at the strains achieved between 0 and 100 mm Hg in this condition.
Isobaric Stress and Einc
Isobaric circumferential wall stress was 1.35, 1.56, and 1.82×106 dynes/cm2 at 100 mm Hg under norepinephrine, baseline, and nitroglycerin conditions, respectively. These modest differences in isobaric wall stress resulted solely from changes in the radius-to-wall-thickness ratio after alterations in smooth muscle tone. According to our model, however, the arterial wall elements that bore wall stress under the three conditions were quite different. In the fully relaxed (nitroglycerin) vessel, the two elements of the PEC, parallel collagen and elastin, bore all the stress, with collagen bearing increasing amounts up to ≈66% of the total wall stress at 100 mm Hg. In contrast, the large decrease in brachial artery area under baseline and norepinephrine conditions shifted the wall components that bore stress so that little if any of the parallel collagen fibers were engaged or contributing to wall stress under these two conditions.
Isobaric Einc was not significantly changed by alterations in smooth muscle tone. We previously demonstrated that contraction of the brachial artery smooth muscle decreases brachial artery compliance, predominantly because of a geometric effect rather than because of an alteration in intrinsic stiffness of the vessel wall.7 Although total isobaric Einc did not change with smooth muscle activation, the wall elements contributing to Einc changed in a manner similar to that described for wall stress. At transmural pressures <50 mm Hg, the magnitude of norepinephrine constriction was sufficient to prevent stretching of the elastin or collagen in the PEC. Under these conditions, vessel stiffness was due solely to smooth muscle and collagen in series with the smooth muscle. At pressures of ≥50 mm Hg, elastin and parallel collagen contributed a small amount to Einc, but the vast majority of wall stiffness was due to smooth muscle and series collagen.
Isometric Stress and Einc
The active circumferential wall stress produced by the brachial artery smooth muscle was determined by subtracting the stress of the relaxed (nitroglycerin) vessel from the stress of the activated (norepinephrine) vessel under isometric conditions. The greatest active stress generated by the smooth muscle in this study was 1.24×106 dynes/cm2. Histological studies have shown that 60% to 75% of the cross-sectional area of moderate-size conduit arteries consists of smooth muscle.25 26 If we assume that 70% of the brachial artery cross-sectional area is smooth muscle, then the active stress generated by the smooth muscle was 1.77×106 dynes/cm2. This compares favorably with computations of maximum active stress from in vitro studies of arterial rings from a variety of sites and animal species that show values of active stress between 2.2 and 3.5×106 dynes/cm2.26 Our value probably is somewhat lower than these values because we measured stress only up to 100 mm Hg transmural pressure. Also, it is possible that a higher dose of norepinephrine would have increased the maximum active stress measured in our study. Higher doses were not used because of concern about systemic effects and activation of systemic reflexes.
In the present study, Einc increased ≈15-fold with isometric smooth muscle contraction. Large increases in isometric Einc after smooth muscle activation have been shown in studies of muscular arteries in vitro. Speden27 found that constricted rabbit arteries (0.60 mm) were extremely stiff up to 200 mm Hg. The findings of Hinke and Wilson28 in rat tail arteries (0.260 mm) constricted with norepinephrine were similar, with marked stiffness up to 140 mm Hg. Finally, Warshaw et al29 found an ≈36-fold increase in Young's modulus with isometric smooth muscle activation of rat mesenteric resistance vessels (0.15 mm). The large increase in isometric Einc with smooth muscle activation is unlikely to be caused solely by stiffness generated in the contractile apparatus. More plausible is the explanation that the smooth muscle and series collagen are connected in a complex manner so that the smooth muscle has a large mechanical advantage in contracting the collagen jacket.24 30 Although our study could not directly address this issue, studies of atrial and ventricular myocytes are consistent with this general concept because intact myocardial tissue is very stiff compared with isolated myocytes.31 32
The technique described in this investigation is invasive and hence is not readily applicable to the study of large populations or the screening of asymptomatic individuals. Nevertheless, it can be used to study the effects of aging, hypertension, atherosclerosis, and other conditions associated with vascular pathology on arterial elastic properties in small groups of human subjects in vivo.
Arterial wall stress, strain, and Einc were measured in the human brachial artery in vivo at baseline and after smooth muscle contraction and relaxation. Using a modified Maxwell model to describe the arterial wall, we estimated the contributions of collagen, elastin, and smooth muscle to the arterial wall mechanical properties at different pressures and degrees of smooth muscle tone. We demonstrated that isometric smooth muscle contraction markedly increases (10- to 15-fold) circumferential wall stress and Einc. In contrast, isobaric smooth muscle contraction produces small changes in overall wall stress and no significant change in Einc while shifting the predominant element contributing to these elastic properties from collagen in parallel with the smooth muscle to collagen in series with the smooth muscle.
This study was supported in part by an American Heart Association grant-in-aid, program project grant P01-HL-32427 from the NIH, and General Clinical Research Center grant M01-RR-00400. We would like to thank Thomas S. Rector, PhD, for his assistance with data analysis and statistics.
- Received April 17, 1996.
- Revision received July 15, 1996.
- Accepted July 18, 1996.
- Copyright © 1996 by American Heart Association
Cox RH. Regional variation of series elasticity in canine arterial smooth muscles. Am J Physiol. 1978;234:H542-H551.
Cox RH. Passive mechanics and connective tissue composition of canine arteries. Am J Physiol. 1978;234:H533-H541.
Dobrin PB, Rovick AA. Influence of vascular smooth muscle on contractile mechanics and elasticity of arteries. Am J Physiol. 1969;217:1644-1652.
Dobrin P, Canfield T. Identification of smooth muscle series elastic component in intact carotid artery. Am J Physiol. 1977;232:H122-H130.
Armentano LA, Barra JG, Levenson J, Simon A, Pichel RH. Arterial wall mechanics in conscious dogs: assessment of viscous, inertial, and elastic moduli to characterize aortic wall behavior. Circ Res. 1995;76:468-478.
Barra JG, Armentano RL, Levenson J, Fischer EIC, Pichel RH, Simon A. Assessment of smooth muscle contribution to descending thoracic aortic elastic mechanics in conscious dogs. Circ Res. 1993;73:1040-1050.
Bank AJ, Wilson RF, Kubo SH, Holte JE, Dresing T, Wang H. Direct effects of smooth muscle relaxation and contraction on in vivo human brachial artery elastic properties. Circ Res. 1995;77:1008-1016.
Carew TE, Vaishnau RN, Patel DJ. Compressibility of the arterial wall. Circ Res. 1968;23:61-68.
Lawton RW. The thermoelastic behavior of isolated aortic strips in the dog. Circ Res. 1954;2:344-353.
Feldman RL, Marx JD, Pepine CJ, Conti CR. Analysis of coronary responses to various doses of intracoronary nitroglycerin. Circulation. 1982;66:321-327.
Longhurst J, Capone RJ, Mason DT, Zelis R. Comparison of blood flow measured by plethysmography and flowmeter during steady state forearm exercise. Circulation. 1974;49:535-540.
Armentano R, Megnien JL, Simon A, Bellenfant F, Barra J, Levenson J. Effects of hypertension on viscoelasticity of carotid and femoral arteries in humans. Hypertension. 1995;26:48-54.
Bergel DH. The static elastic properties of the arterial wall. J Physiol (Lond). 1961;156:445-457.
Patel DJ, DeFreitas FM, Greenfield JC, Fry DL. Relationship of radius to pressure along the aorta of living dogs. J Appl Physiol. 1963;18:1111-1117.
Hill AV. The heat of shortening and the dynamic constraints of muscle. Proc R Soc Lond B Biol Sci. 1938;126:136-195.
Sonnenblick EH. Series elastic and contractile elements in heart muscle: changes in muscle length. Am J Physiol. 1964;207:1330-1338.
Roach MR, Burton AC. The reason for the shape of the distensibility curves of arteries. Can J Biochem Physiol. 1957;35:681-690.
Apter JT. Correlation of visco-elastic properties with microscopic structure of large arteries: thermal responses of collagen, elastin, smooth muscle, and intact arteries. Circ Res. 1967;21:901-918.
Burton AC. Relation of structure to function of the tissues of the wall of blood vessels. Physiol Rev. 1954;34:619-642.
Murphy RA. Mechanics of vascular smooth muscle. In: Bohr DF, Somlyo AP, Sparks HV Jr, eds. Handbook of Physiology: The Cardiovascular System—Section 2. Bethesda, Md: American Physiological Society; 1980:325-351.
Dobrin PB. Mechanical properties of arteries. Physiol Rev. 1978;58:397-460.
Hinke JAM, Wilson ML. A study of elastic properties of a 550-μm artery in vitro. Am J Physiol. 1962;203:1153-1160.
Warshaw DM, Mulvany MJ, Halpern W. Mechanical and morphological properties of arterial resistance vessels in young and old spontaneously hypertensive rats. Circ Res. 1979;45:250-259.
Covell JW. Functional dynamics of the myocyte-connective tissue unit. In: Robinson TF, Kinne RKH, eds. Cardiac Myocyte: Connective Tissue Interactions in Health and Disease. Basel, Switzerland: Karger; 1990:99-112.
Tarr M, Trank JW, Leiffer P, Shepherd N. Sarcomere length-resting tension relation in single frog atrial cardiac cells. Circ Res. 1979;45:554-559.