The areas ratio of normal arterial junctions and its implications in pulse wave reflections.

STUDY OBJECTIVE: The aim of the study was to establish measurements for the area ratios of normal arterial junctions at various anatomical sites in order to estimate the significance of pulse wave reflections in the healthy arterial system. DESIGN: Coronary cineangiograms were used for the coronary junctions measurements and routine arteriograms for the rest of the junctions studied. A pair of digital calipers served to take measurements at a distance of one vessel diameter from the junction centre. MEASUREMENTS AND MAIN RESULTS: A bell shaped distribution of arterial area ratios was found with mean value for coronary bifurcations of 1.179 (95% confidence limits, 1.138-1.220). Slightly higher values were found for the higher order coronary junctions. Other junctions had similar area ratio values with the exception of the aortoiliac junctions where the area ratio was found 0.848 (0.775-0.920). Overall area ratio was 1.14 (1.113-1.167). Asymmetrical coronary bifurcations had higher area ratio than symmetrical ones (p less than 0.05). CONCLUSIONS: In the majority of cases the area ratios are in close agreement with the theoretically predicted values for forward matched junctions. Minimal pulse wave reflections are therefore expected to arise from the junctions of a healthy arterial system. We therefore consider that, contrary to the established view, the presence of reflected waves in the central arterial system is a potential pathological phenomenon and not a characteristic of the healthy system. This may have important theoretical implications for the way hemodynamic findings are interpreted. Finally we suggest that arterial area ratio can be considered as another criterion of normality of arterial junctions in the interpretation of angiograms.

Hydraulic input impedance measurements in physical models of the arterial wall.

A white noise method was used to measure the hydraulic input impedance and transmission characteristics in physical models of an arterial system made of single, unbranched latex tubes. The experimentally obtained impedance curves show a rise in modulus and a positive phase at high frequencies in the absence of wave reflections. Using the impedance moduli in the presence of wave reflections, wave velocity and attenuation were calculated. The influence of wall nonlinearity on hydraulic impedance was also examined. It is concluded that, in the model used neither wave reflections nor wall nonlinearity can account for the deviations of the experimental impedance curves from the theoretically predicted ones. Impedance moduli in the presence of reflections may be used to study transmission characteristics (wave velocity and attenuation) of the model.

Frequency response of the arterial wall.

The high frequency response of human common iliac arterial segments in vitro was investigated. It was found that at those high frequencies the response resembles that of a second order underdamped system. However, to stimulate the arterial response throughout the frequency range, a higher order model is required. A fifth order system appears to describe the observed behaviour in a satisfactory way between 0.02 and 200 Hz.

Circumferential and longitudinal viscoelasticity of human iliac arterial segments in vitro.

A random noise technique was used to measure the circumferential and longitudinal dynamic elasticity of human common iliac arteries in vitro. For circumferential measurements the frequency ranged from 0.016 to 20 Hz; the phase lag of diameter behind pressure was found to be almost constant (about 5 degrees) and the Young's modulus of elasticity to increase rapidly at first and then more gradually beyond 1-2 Hz. Somewhat similar results were obtained for longitudinal elasticity. Arterial segments were found to be anisotropic when kept at in vivo length and under normal distending pressure.

Arterial system configuration and wave reflection.

On the assumption that the arterial segments between branching points are cylindrical, the relationship between total arterial cross-sectional area and the distance from the heart for the entire length of the arterial system is derived. Based on this relationship, a model of the whole arterial system is proposed; it is composed of three tubes of common origin and of unequal lengths and cross-sectional areas. This model may be useful for simulation purposes, with particular reference to pulse wave reflections.

Physical modelling of the arterial wall. Part 2: Simulation of the non-linear elasticity of the arterial wall.

In part 1 it was established that the elasticity of tubes made from natural rubber is, at low pressures, close to that of the arterial wall; in this part a method is described for reinforcing the walls of rubber tubes so that they may be used for simulating arterial elasticity at higher pressures. The technique is versatile in that, within reasonable limits, wall non-linearity can be modified; tubes with non-linear wall elasticity were produced and tested. A modification of the method enabled us to produce geometrically non-uniform rubber tubes which may be useful in the construction of physical models of the arterial system.

Physical modelling of the arterial wall. Part 1: Testing of tubes of various materials.

Tubes of various elastic materials were tested using a purpose-built apparatus to select those most appropriate for physical simulation of the arterial wall. The influences of temperature and longitudinal stress were measured in selected tubes. It was found that the static elasticity of latex tubes is close to that of the arterial wall for intraluminal pressures corresponding to the lower range of intra-arterial pressures.

Photoelectric transducer for measuring the length and diameter of elastic vessels.

This photoelectric transducer was developed to measure diameter and length variations in arterial segments in vitro and in elastic tubes. Its construction is simple, its nonlinearity small and its sensitivity adequate. The cut-off frequency is 300 Hz and its small phase lag changes linearly with frequency.