Functional properties of fresh and cryopreserved carotid and femoral arteries, and of venous and synthetic grafts: comparison with arteries from normotensive and hypertensive patients

Abstract

The ideal arterial graft must share identical functional properties with the host artery. Surgical reconstruction of the common carotid artery (CA) is performed in several clinical situations, using expanded polytetrafluoroethylene prosthesis (ePTFE) or saphenous vein (SV) grafts. At date there is interest in obtaining an arterial graft that improves the results of that nowadays available. The use of a fresh or cryopreserved/defrosted artery appears as an interesting alternative. However, if the fresh and cryopreserved/defrosted arteries allow an adequate viscoelastic and functional matching with the host arteries needs to be established. The aims were to compare the viscoelastic and functional performance of: (1) conduits used in CA reconstruction (SV and ePTFE) with those of the fresh and cryopreserved/defrosted CA and femoral arteries (FA), and (2) normotensive and hypertensive patients’ arteries with those of the arterial substitutes in vitro analyzed. Pressure, diameter and wall thickness of the CA were recorded in 15 normotensive and 15 hypertensive patients (in vivo studies), and in SV, fresh and cryopreserved/defrosted CA and FA (obtained from 15 donors), and ePTFE segments (in vitro studies). From stress–strain relationship we calculated elastic and viscous modulus, and the characteristic impedance. The local buffer and conduit functions were quantified as the viscous/elastic quotient and the inverse of the characteristic impedance. Fresh and cryopreserved/defrosted CA and FA were more alike, both in viscoelastic and functional levels, respect to normotensive and hypertensive patients’ arteries, than the ePTFE and SV grafts. CA and FA cryografts could be considered an important alternative for carotid reconstruction.

Appendix

Data analysis

In this work diameter waveforms were assessed by two different methodologies: echography and sonomicrometry. Both techniques have been validated and used to evaluate the vascular biomechanical properties (Graf et al. 1999). Pressure waveforms were also assessed by different methodologies: applanation tonometry and intravascular pressure microtransducer (Konigsberg). The high fidelity of the pressure waveforms obtained with these techniques has been demonstrated (Armentano et al. 1995b, 1998; Nichols and O’Rourke 1998).

The wall thickness of the segments studied in vitro was calculated as the difference between the external radius (r _e=diameter/2), and the internal radius (r _i), estimated as previously reported (Armentano et al. 1995a). Patients’ CA wall thickness was assumed to be similar to the intima-media wall thickness index (Astrand et al. 2003). Strain ($\varepsilon$) and circumferential wall stress (σ) were calculated according to previous works (Armentano et al. 1995a; Bia et al. 2005a).

Viscoelastic properties

The viscoelastic properties were evaluated using a Kelvin–Voigt viscoelastic model (spring-dashpot) (Bia et al. 2005a, b). According to it, the σ developed in the wall (σ_total) can be divided into elastic␣(σ_elastic) and viscous (σ_viscous) components (Armentano et al. 1995a).

$$\sigma_{\rm total}=\sigma_{\rm elastic}+\sigma_{\rm viscous} $$

(1)

As the viscous component is proportional to the first derivative of $\varepsilon$ respect to time ($d\varepsilon /dt$), the equation is written as:

$$\sigma_{\rm elastic}=\sigma_{\rm total}-\eta.d\varepsilon/dt $$

(2)

where η is the viscous modulus of the arterial wall. To obtain the pure elastic σ component, the viscous term must be subtracted from the σ_total. To do this, the area of the σ–$\varepsilon$ hysteresis loop was reduced until the minimum value, that still preserves the clockwise direction of the loop (Armentano et al. 1995a; Bia et al. 2005b). The viscous value needed to reach this condition was considered the viscous modulus (η). When the elastic component of σ_Total had been isolated, the incremental elastic modulus (E) was calculated by means of the slope of the linear regression curve, evaluated at the mean prevailing pressure (Bia et al. 2005a).

Local buffer function

To characterize the local buffer function, the σ–$\varepsilon$ relationship was established using Eqs. (1) and (2), and the computed E and η:

$$\sigma (t)=\sigma_0 \quad t> 0 \,\Rightarrow\, \varepsilon (t)=\frac{\sigma_{0}}{\eta/E}\left({1-e^{-t/\eta/E}}\right)\quad t> 0 $$

(3)

where the η/E ratio characterizes the exponential temporal response of strain due to a stress change. This ratio, the time constant of the Kelvin–Voigt model or “time retardation” (Westerhof and Noordergraaf 1970), evaluates the intrinsic capacity of the material to cushion the stress exerted over its surface. Recently, our group proposed the quantification of the wall local buffer function by means of this time constant (Bia et al. 2005a). An elevated value is related with a slow response, suggesting an augmented buffering effect with an increased attenuation of stress or pressure oscillations (Bia et al. 2005a).

The η/E has the elastic modulus as denominator, responsible of the storage capacity of the arterial wall, and the viscous modulus as numerator, responsible of the arterial wall energy dissipation (Bia et al. 2005a). The local buffer function, influences the propagation and interaction of pressure and flow waves throughout the arterial tree and, consequently, the energetic demands placed on the heart (Shadwick 1999; Morita et al. 2002), and inhibits the sharp peaks of the pressure and flow pulses (Pontrelli and Rossoni 2003). Thus, vascular viscosity and elasticity, and in consequence the local buffer function, are important determinants of the whole cardiovascular system performance.

Local conduit function

Similar to our previous work, the local conduit function was calculated as 1/Z _C, where Z _C is the characteristic impedance (Bia et al. 2005a). Assuming a cylindrical geometry for the arterial vessels, Z _C was in vivo and in vitro estimated by using the Water–Hammer equation (Nichols and O’Rourke 1998)

$$ Z_{\rm C} =\frac{\rho\cdot\hbox{PWV}}{\hbox{diastolic area}} $$

(4)

where ρ is the blood density (assumed equal to 1.055 g/ml), and PWV is the pulse wave velocity estimated from the Moens–Korteweg equation (Nichols and O’Rourke 1998):

$$\hbox{PWV} =\sqrt{\frac{E\cdot h_{\rm m}} {2\rho_{\rm T}\cdot R_{\rm m}}} $$

(5)

where ρ_T is the wall tissue density (ρ_T=1.06 g/cm³) and h _m is the wall thickness. Midwall radius (R _m) was calculated as:

$$R_{\rm m}=(r_{\rm e} +r_{\rm i})/2 $$

(6)

As can be seen in Eqs. (4) and (5) the Z _C correlates directly with the elastic properties and inversely with the cross-sectional area of the conduit. An increased Z _C would determine an increase in local impedance against blood flow, resulting in a decreased capability to conduct blood (Cholley et al. 2001). Therefore, the local conduit function was computed as 1/Z _C. The local conduit function of the arteries allows the blood transference from the heart to the peripheral vessels. To maintain an adequate high level of mean pressure and to minimize ventricular work and wave reflections, low arterial impedance must be offered to the pulsatile blood flow ejected by the heart (Pepine and Nichols 1982; Bia et al. 2005a). The implantation of a rigid graft into the arterial tree – comparable to the introduction of an impedance into an oscillating electrical circuit – would diminish the perfusion efficiency and, in low-flow situations, may lead to further flow stagnation and graft thrombosis (Tai et al. 2000), with concomitant increase in ventricular afterload when the rigid graft is implanted in a large artery (Tai et al. 2000; Morita et al. 2002).