Today, there is a growing interest in the physical properties of large arteries. Hence, an accurate characterisation of the behaviour of these arteries could contribute to a better understanding of the material and underlying dynamics. Viscoelastic properties of the arterial wall determine the response to dynamic forces such as pulsatile pressure and flow (Milnor, 1982; Nichols & O’Rourke, 1990; Simon et al. 1991) and hence potentially cellular/molecular biological responses (Davies, 1995). The presence of viscous components in the arterial wall suggests time- or frequency-dependent responses to changes in wall stress (Peterson et al. 1960; Bergel, 1961; Milnor, 1982; Nichols & O’Rourke, 1990). The dynamic behaviour of arteries at different frequencies is an important element in both theoretical and practical haemodynamics. When the Young’s modulus of the arterial wall is determined as a function of frequency, a particular behaviour is found. For frequencies from 0.01 to 1.5 Hz, the amplitude of the complex Young’s modulus rises steeply to a quite constant level (Fung, 1981; Cox, 1984). This value is reached at frequencies below heart rate. Hardung (1970) and Goedhard et al. (1973) have provided data in this low frequency range from in vitro experiments. Sipkema (1979) studied the viscoelastic behaviour by means of a servo- controlled occluder system in vivo. Gow & Taylor (1968) measured the viscoelastic properties of arteries in anaesthetised animals in both the low and the high frequency ranges using power spectrum analysis from pressure and diameter data. The power spectrum method provides frequency–response information, but it does not consider explicitly mechanical properties of the arterial wall. Moreover, tracking of time-varying dynamic properties is also difficult using a non-parametric transfer function estimation method. Identification of arterial wall dynamics in conscious dogs Lucas G. Gamero *†, Ricardo L. Armentano *§, Juan G. Barra *, Alain Simon ‡ and Jaime Levenson ‡ * Favaloro University, Solís 453, (1078) Buenos Aires, † Facultad de Ingeniería, Universidad Nacional de Entre Ríos y Universidad de Buenos Aires, Buenos Aires, Argentina and ‡ Cardiovascular, CRI (INSERM) Hôpital Broussais, 96 rue Didot, Paris 14, France (Manuscript received 17 November 2000; accepted 17 April 2001) Viscoelastic properties determine the dynamic behaviour of the arterial wall under pulsatile pressure and flow, suggesting time- or frequency-dependent responses to changes in wall stress and strain. The objectives of the present study were: (i) to develop a simplified model to derive simultaneously the elastic, viscous and inertial wall moduli; (ii) to assess Young’s modulus as a function of frequency, in conscious, chronically instrumented dogs. Parametric discrete time models were used to characterise the dynamics of the arterial system based on thoracic aortic pressure (microtransducer) and diameter (sonomicrometry) measurements in control steady state and during activation of smooth muscle with the a- adrenoceptor agonist phenylephrine (5 μg kg _1 min _1 , I.V.), in eight conscious dogs. The linear autoregressive model and a physically motivated non-linear model were fitted to the input–output (stress–strain) relationship. The aortic buffering function (complex Young’s modulus) was obtained in vivo from the identified linear model. Elastic, viscous and inertial moduli were significantly increased from control state ((44.5 ± 7.7) w 10 4 Pa; (12.3 ± 4.7) w 10 4 Pa s; (0.048 ± 0.028) w 10 4 Pa s 2 ) to active state ((85.3 ± 29.5) w 10 4 Pa, P < 0.001; (22.4 ± 8.3) w 10 4 Pa s, P < 0.05; (0.148 ± 0.060) w 10 4 Pa s 2 , P< 0.05). These moduli, obtained using the linear model, did not present significant differences compared with those derived using the non-linear model. In control conditions, the magnitude of the normalised complex Young’s modulus was found to be similar to that reported in previous animal studies ranging from 1 to 10 Hz. During vascular smooth muscle activation, this modulus was found to be increased with regard to control conditions (P < 0.01) in the frequency range used in this study. The frequency-dependent Young’s modulus of the aortic wall was obtained for the first time in conscious, unsedated dogs. The parametric modelling approach allows us to verify that vascular smooth muscle activation increases the elastic, viscous and inertial moduli with the advantage of being able to track their time evolution. Furthermore, under activation, the aortic wall remains stiff in the physiological frequency range, suggesting the impairment of the arterial buffering function. Experimental Physiology (2001) 86.4, 519–528. 2172 Publication of The Physiological Society § Corresponding author: armen@favaloro.edu.ar ) by guest on April 29, 2013 ep.physoc.org Downloaded from Exp Physiol (