Abstract
Thermal dilution as a method for measuring peripheral blood flow has been the basis of at least two major pieces of work in the past, (Clark 1966, Richards 1970). It is questionable whether anybody to date has been able to make a thermal dilution probe which, when inserted into a blood vessel of a living subject, could measure with accuracy the blood flow rate minute by minute over a long period. The possible reasons for the failures in vivo are examined in this thesis. The presence of blood clot on and around the probes is shown to be a major problem caused by the use of a centralising structure on the probe tip. Methods for making the centralising cage less thrombogenic are examined in the light of an enquiry into blood clotting mechanisms. Available techniques for making plastic surfaces less thrombogenic are reviewed, and their suitability for use on the thermal dilution probe examined. An anti-thrombogenic catheter is developed and tested in vivo using radioactive labelling techniques to monitor clot formation. The outcome of subsequent tests result in the abandonment of the centralising cage technique, the ramifications of which have necessitated the redesign of the flow probe and restatement of the problem of blood flow measurement by thermal dilution. A new method (based on that of Fronek and Ganz 1960), is investigated using a single hole for injecting saline. The effects of the injection on flow and pressure are considered both in vivo and in vitro, for various angles of injection, rates of injection (i. e. various kinetic energies of jet), flow rates and vessel diameters. The effect of the distance between the injection site and the measuring site on the accuracy of the method, is investigated. A venous velocity trace and an arterial velocity trace are examined using a mathematical model in order to demonstrate the unpredictability of "distance-distortion". A conflict of interests arises between the need to reduce changes in flow during injection to a minimum, whilst obtaining a suitably homogeneous mixture for measurement to be made in a short enough distance to reduce distance distortion to an acceptable level.