|
|
||||||||
1Departamento de Ciencia de Materiales, E. T. S. de Ingenieros de Caminos, Universidad Politécnica de Madrid, and 2Departamento de Patología Cardiaca, Hospital Clínico Universitario San Carlos, Madrid, Spain; and 3Department of Mechanical Science, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka, Japan
Submitted 28 October 2004 ; accepted in final form 28 January 2005
| ABSTRACT |
|---|
|
|
|---|
blood vessels; inflation tests; thermal dilatation coefficient; mechanical properties; structural stiffness
The thermomechanical behavior of arteries, however, is not a secondary issue in cardiovascular research because many cardiac surgical procedures are performed at nonphysiological temperatures; coronary artery bypass surgeries are ordinarily performed in conditions of hypothermia (2631°C) (13), and hyperthermia over 50°C is routinely used in thermal balloon angioplasty (3, 10, 16, 22).
Still today, a lot of efforts are devoted to understanding the effects of changes in temperature on the final results of the treatments and the quality of life of the patients (13, 16). Also, the influence of temperature on vascular properties is important in situations like organ preservation, because modifications of temperature could affect intrarenal vasodilatation, bypass surgery, and extracorporeal circulation.
In addition, thermomechanical experimental data are needed for the development of appropriate constitutive equations for the arterial wall (9, 18) implemented in numerical models, which are proving to be a useful tool for surgical procedures as well as for other clinical cardiovascular issues such as the assessment of plaque vulnerability (9).
The first study on the effect of temperature on human blood vessels was brought out more than a century ago by C. S. Roy, who showed that unloaded arteries shrink when heated and expand when cooled (17). Roy tested human, cow, and sheep aortas in the passive state in the range of 1654°C.
Parallel results were reported by Lawton (15), who performed isometric tests on passive dog aortas at 1540°C and found that arterial elasticity was primarily entropic-like elastomers. Nevertheless, other data suggest that heat-induced contraction of blood vessels is not so apparent; Dobrin and Canfield (2), working on dog carotid arteries at the 3339°C interval, did not notice a clear influence of temperature on passive behavior. In addition, Kang et al. (10) observed a weak effect of temperature on the multiaxial mechanical behavior of passive bovine coronary arteries in the range of 2155°C and reported a significant stiffening of the arterial wall only above 60°C.
In a way similar to passive mechanics, the influence of temperature on the active behavior of arteries is still in question. In vivo tests on rabbit mesenteric arteries by Gorisch and Boergen (4) showed no contraction under 75°C, and Deng et al. (1) found that temperatures between 25 and 37°C had a small effect on the shear modulus of rat aortas under multiaxial loading. Herrera et al. (7) reported contradictory effects of temperature on isometric tests performed in the range of 537°C, where a rise of temperature induced contraction in rat aortas while relaxing pig renal arteries.
The cross-effect of temperature and vasoconstrictor drugs was studied by Keatinge (12), who found that temperatures under 10°C inhibited the response of bullock ulnar arteries to adrenaline. Toda et al. (19) showed that the contractile response of rabbit aortas to vasoconstrictor drugs reached a maximum between 33 and 37°C.
From the literature presented above, it appears that temperature may influence both the passive and active behavior of arteries, although the detailed conditions where it occurs are still controversial. Further studies are needed to determine and quantify its effects.
The purpose of the present study was twofold: 1) to offer experimental results on the effect of temperature on the passive mechanical behavior of human carotid arteries that can be profitably used for clinical treatment and cardiovascular research, and 2) to provide valuable data to serve as an experimental benchmark for checking and tuning numerical models on human arteries. We present data from 11 inflation tests performed on arteries stretched to in vivo length at temperatures ranging from 17 to 42°C.
Carotid arteries have been characterized by two general parameters that reflect the overall vessel behavior from the mechanical as well as the thermal point of view: the thermal dilatation coefficient (
) and Hayashi's stiffness parameter (
). Both parameters give useful values that can be directly applicable to clinical procedures and easily used for research purposes. Other structural parameters such as the pressure-strain elastic modulus (16a) or vascular compliance (4a) can be derived from
, as described in Ref. 6.
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
Histological analysis. Before the testing, a histological analysis was performed on 5-mm edge portions of the samples to determine their structure and condition by light microscopy. Specimens were fixed by immersion in formaldehyde in no-load conditions and embedded in paraffin to obtain the sections. They were then deparaffinized, hydrated, and stained by hematoxylin-eosin and orcein to resolve the elastic fibers.
Mechanical tests. In vitro pressure-diameter tests at fixed (in vivo) length were carried out at four different temperatures (17, 27, 37, and 42°C).
Arterial segments were cannulated at both ends by means of veterinary needles of appropriate diameter. To prevent sliding between the sample and the cannula, the tips of the needles were machined to form a flange.
Two stainless steel fixtures joined the needles to the grips of an electromechanical tensile testing machine (Instron 4411) equipped with a 100N load cell (Instron 2525-806). The lower fixture was designed to permit internal pressurization of the vascular segment through the needle. The experimental device is sketched in Fig. 1.
|
Once mounted on the holders, each specimen was stabilized at 37°C and stretched to the in vivo length, equal, on average, to 1.13 times the length at rest. The specimen was then brought to the prescribed temperature and stabilized for at least 15 min. Intravascular pressure was then applied via the lower needle and measured with a pressure transducer with 0.2-mbar accuracy (Druck PMP 4000). Proper zero adjustment of the intraluminal pressure was achieved by subtracting the external pressure due to the fluid in the PMMA chamber. The outer diameter of the specimen, measured at the midpoint section, was continuously recorded using an optical extensometer with 1-µm accuracy (Keyence LS-7500).
All the specimens were mechanically preconditioned before final testing to remove initial stress-relaxation effects and to obtain a stable response. Transmural pressure, i.e., the pressure difference between the inside and the outside of the vessel, was cycled from 0 to 200 mmHg at a loading rate of 3 mmHg/s until stable and reproducible pressure-diameter hysteresis loops were formed. All the arteries tested achieved a stable response before 10 cycles. After being preconditioned, each specimen was then subjected to a new pressurizing cycle between 0 and 200 mmHg at 3 mmHg/s while pressure and external diameter were continuously monitored. Only the ascending branch of the inflation test was used for the analysis, because all the carotid segments displayed similar behavior with a small hysteresis loop.
Test temperatures were sequentially applied on every specimen starting at the lowest level (17°C). The average total test duration per sample (tests at the four temperatures) was 120 min. To check reproducibility, some samples previously tested at the four temperatures (17, 27, 37, and 42°C) were tested once again at 27°C. Because that no appreciable differences were observed between pressure-diameter curves of the two tests at 27°C, it was concluded that pressure-diameter tests of carotid arteries were not affected by the duration of the test.
Thermomechanical analysis.
The effect of temperature on the mechanical properties of the carotid arteries was evaluated through pressure-diameter curves measured at different temperatures. Nevertheless, to facilitate data analysis and discussion of the results, two simple "global" parameters were computed from the experimental curves: the thermal dilatation coefficient
and Hayashis
stiffness parameter.
measures the change in size caused by temperature. Positive values of
correspond to substances that expand when heated, whereas negative ones indicate that the material shrinks as the temperature increases. The fractional variation in diameter of a vessel is (in first order) proportional to the temperature increment (
T) and given by
![]() | (1) |
![]() | (2) |
![]() | (3) |
TRef, and
is readily obtained from experimental data by direct fitting of Eq. 3.
The structural stiffness of the arterial wall has been characterized by the so-called stiffness structural parameter
, introduced by Hayashi and co-workers (5). They proposed an exponential equation to describe the pressure-diameter relationship in the physiological range, which is given by
![]() | (4) |
values have been found to increase gradually with age and arterial diseases such as atherosclerosis (8). | RESULTS |
|---|
|
|
|---|
|
|
|
|
at different pressure levels were computed for all the arteries by least-square fitting of Eq. 3 to diameter-temperature pairs corresponding to the same pressure. The mean values of
and the standard errors are given in Fig. 6 as a function of pressure.
|
displays negative values from zero pressure up to
10 mmHg, where an inversion takes place and
becomes positive. The
curve increases monotonically with pressure level, although a progressively decreasing slope suggests that it could approach a saturation value close to +103 (°C)1.
Structural stiffness parameter.
Figure 7 shows the mean
value and standard error for human carotid arteries as a function of temperature. Individual
values were computed for every specimen at specific temperatures by least-square fitting of Eq. 4 to recorded pressure-diameter curves.
|
seemed to decrease moderately with temperature, this variation was not statistically significant due to the large experimental scatter. Paired Student's t-test between
values obtained at the two extreme temperatures (17 and 42°C) gave P = 0.3, as shown in Fig. 7. | DISCUSSION |
|---|
|
|
|---|
In this work, we focused on the influence of temperature on the passive mechanical behavior of human carotid arteries, with all the tests being performed on cryopreserved specimens. By this procedure, we separated the ground behavior of the arterial wall (mainly dictated by elastin and collagen components) from that influenced by the muscular response. Nevertheless, it is worth noting that preliminary results by the authors on living carotid arteries suggest that as long as no stimulation is present, findings reported in this study could also apply to living vessels.
Broadly speaking, temperature seems to have a softening effect on the passive response of carotid arteries when measured by pressure-diameter tests. Figure 3, which shows the mean behavior registered in all the tests, illustrates how higher temperatures shift pressure-diameter curves to the right at large enough values of the pressure, whereas the opposite (a shift to the left) happens for low-pressure values. As a result, pressure-diameter curves become more compliant. This trend is more easily observed in Fig. 5 by the introduction of diameters relative to zero-pressure level at each temperature [D/D(0,T)].
The reverse effect of temperature at high and low pressures gives rise to a pressure-dependent behavior of
. When a carotid artery is subjected to pressure as low as P = 0 mmHg, its outer diameter shrinks as temperature increases (Fig. 4), and the least-square fitting of (D,T) values to Eq. 3 results in a mean negative
equal to 0.63 x 103 (°C)1.
However, diameter dependence on temperature reverses above a certain pressure threshold, and the artery dilates noticeably as the pressure level rises. In the case of P = 100 mmHg, the reference physiological pressure, the slope of the diameter versus temperature curve is positive (Fig. 4) and the fit of Eq. 3 to experimental points gives a mean
of +0.73 x 103 (°C)1.
Figure 6 illustrates the evolution of the mean
with transmural pressure. Despite the experimental scatter, it is shown that
increases at a decreasing rate as pressure rises, from a negative value close to 0.7 x 103 (°C)1 for P = 0 mmHg to almost +103 (°C)1 at P = 200 mmHg. The mean
becomes positive at a pressure threshold of about 10 mmHg and reaches +0.7 x 103 (°C)1 at P = 100 mmHg. Although the data shown in Fig. 6 could suggest
coming close to a saturation value at high enough pressure levels, this point is still controversial and deserves further research.
The "atypical" (negative)
found at pressures under 10 mmHg fully agrees with the early findings by Roy (17), who in 1880 first reported this behavior in human blood vessels. He studied length changes in 9.25-cm strips (1 cm in diameter) taken from the human aorta under no internal pressure. Interestingly, the
that can be deduced from his measurements is equal to 0.7 x 103 (°C)1, a value close to that we have obtained in this work for P = 0 mmHg (see Fig. 4). Other studies on mammalians have reported similar results in unloaded vessels, such as the work by Lawton in 1954 (15) on dog aortas, which indicated the entropic basis of this behavior as primarily dictated by arterial wall composition (elastin and collagen).
Nevertheless, Herrera et al. (7, 8) have shown opposite trends in tests performed on living pig renal arteries and rat aortas (the former contracted when heated whereas the latter expanded), which they explained by the different nature of the vessels (muscular vs. elastic). They put forward the hypothesis that elastic fibers (prevalent in elastic arteries such as the aorta) were responsible for the cooling-induced relaxation, whereas smooth muscle cells largely found in muscular vessels contract due to the increase of intracellular calcium induced by temperature. This explanation, however, cannot be applied to our results because the human common carotid artery is mainly an elastic vessel, as shown in Fig. 2, and, in addition, our specimens were not alive. Consequently, the observed pressure dependence of
has to be considered as produced by passive constituents of the arterial wall.
In accordance with the results presented in this work, one of the authors discovered a similar behavior in Wistar rats (K. Hayashi, unpublished observations). The thermal expansion of their common carotid artery has been found to be pressure dependent, changing from negative to positive at about P = 120 mmHg. Although both types of artery show an inversion point, the inversion pressure in human carotid arteries is significantly lower than that measured by Hayashi in Wistar rats. One possible reason is the observed strong dependence of
on the mechanical stress state. Rat arteries were axially stretched up to 1.4 and over (in vivo length for rat carotids), in contrast to the 1.13 elongation ratio for human carotid arteries used in our work, so the combined action of axial stress (due to axial stretching) and circumferential stress (due to pressure) could be responsible for the shifting of the inversion point. In addition, other factors could came into play such as differences in arterial wall composition and the distribution of elastin and collagen.
With regard to arterial stiffness, Fig. 7 shows the values of
, which does not seem to change appreciably within the range of temperature analyzed in this work.
, on average, seems to diminish moderately with temperature and is consistent with the observed increment of compliance of pressure-diameter curves, as deduced from Figs. 3 and 5. Nevertheless, this tendency cannot be taken as statistically significant due to the large scatter of experimental points.
The high
values obtained in our work are comparable with results from other human arteries of elderly patients (11, 14), which have shown that arterial stiffness increases gradually with age, specially above 70 yr (6).
In conclusion, this work demonstrates that the combined effect of temperature and stress has a striking effect on the dilatation coefficient of the human carotid arterial wall, which can shift from negative to positive depending on the stress state. On the other hand, the structural stiffness of carotid arteries seems not to change appreciably within the range of temperatures tested.
As the available data reveal, the combined effect of temperature and stress can dramatically affect arterial properties, and consequently has to be properly taken into account for clinical treatments and when developing constitutive equations for the arterial wall. Further research is required to complement and enlarge the reported thermomechanical data on human arteries.
| GRANTS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. J. Dougherty, M. J. Davis, D. C. Zawieja, and M. Muthuchamy Calcium sensitivity and cooperativity of permeabilized rat mesenteric lymphatics Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2008; 294(5): R1524 - R1532. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |