The transmission of sound waves in liquid-filled catheter tubes used for intravascular blood-pressure recording.

Abstract

The blood pressure wave-form recorded by a transducer of finite compliance at the end of a liquid-filled tube, is often distorted by acoustic resonance. In most recent publications, e.g.Leraand (1962),Yanov (1963), the theoretical treatment of this problem is based on the dubious asumption that the catheter compliance and the compressibility of the liquid are negligible, and that the system behaves as a simple resonant circuit with one degree of freedom, in which the mass of the fluid, or more correctly its inertance, resonates with the compliance (springiness) of the transducer. In a few cases the published theory does not rest on this assumption. In one article, the catheter has been considered as the π-section of a low-pass filter, in which half the compliance appears across the transducer,Vierhout (1965). This is shown to be quantitatively incorrect, as the effective fraction is one-third and not one-half. In any case the implication that there is a definite cut-off frequency is misleading.A. T. Hansen (1949) has made a more rigorous analysis in which the result is expressed in terms of complex hyperbolic functions, from which it seems difficult to draw conclusions except by numerical evaluation. It has, however, been found possible to simplify Hansen's results. It is shown that the catheter itself may be made incapable of causing resonance. This can be accomplished by a form of impedance matching at the input; this technique only differs from normal telecommunication practice in respect of the ability of the system to operate down to zero frequency and in the deliberate mismatching of the output. In principle the catheter input includes a constriction whose acoustic resistance is capable of absorbing half the pressure of a steepfronted pressure wave at the moment of its application. This loss of signal strength is exactly compensated by the response of the transducer both to the incident and reflected waves at the output, the latter being absorbed on its return to the constriction. This plan not only removes any sharply defined upper limit to the frequency which can be recorded, but also considerably simplifies the equalization problem. Except for minor adjustments, the equalization is independent of catheter length.