Investigation of fluid mechanical influences on the clotting of a blood analogue fluid

Abstract

Whilst advances in both cardiac surgery and artificial heart valve design have greatly enhanced survival expectations following cardiac valve replacement, an Ideal valve has yet to be found: Tissue valves are still liable to lipid deposition and calcification, and mechanical prostheses cause thrombosis unless accompanied by a sufficient level of anticoagulation therapy.

Thrombosis in flowing blood around mechanical prostheses can arise from both the materials of construction of the valve and the local hydrodynamics. Incidence of thrombus due to the former can be eliminated by the use of such materials as vitreous and pyrolytic carbon. From a haemodynamic perspective, stasis, stagnation, shear stress and mass transfer have all been proposed as correlates of clotting; but, as yet, no one has determined the relative Importance of each.

The present method of assessing thrombogenicity of valves and similar devices is by Implantation in animals. This, apart from being distasteful, is expensive, requires lengthy experimentation and leads to results that may not necessarily apply to the human vascular system. The use of blood in vitro is unsatisfactory for heart valve evaluation, principally because in order to avoid recirculation of clotted material a single-pass system using blood at a rate of about 30 gallons/hour Is required. Also, methods of storage of blood may affect the blood chemistry causing blood in vitro to be at best an analogue for blood in viva In this thesis a blood analogue fluid for thrombogenicity assessment is investigated, both to determine its suitability as such and to identify the hydrodynamic conditions for deposition of clot from an enzyme-triggered coagulation.

Lewis proposed a milk mixture as being a suitable analogue fluid for the evaluation of thrombogenic potential of cardiovascular Implants. In Hladovec's net experiment and Petschek's stagnation point flow chamber experiment the milk clot deposition patterns were found to be remarkably similar, both macroscopically and microscopically, to thrombus deposition from blood. When Bjork-Shiley and Starr-Edwards valves were tested with this milk mixture In an experimental heart pump, clot formed in the same regions of the valves as does thrombus when the valves are implanted in the human heart. Lewis, however, had difficulty controlling the temperature of the milk mixture and therefore could not, with certainty, differentiate quantitatively between the clot forming tendencies of the different valves.

To overcome this problem of temperature control, a new test rig was designed (and built commercially), capable of supplying the prepared milk mixture at a constant temperature, in steady or pulsatile flow, to the object under test. An Investigation of the deposition of coagulum on a variety of test objects, from both steady-and pulsatile milk flows, has been conducted using this apparatus. The objects, solids of revolution of shapes used by Vorhauer in viva were housed in a cylindrical test chamber to provide axisymmetrical flow. Clot found on the upstream face of the objects Is thought to be adventitious being caused by vapour bubbles adhering to the object, by - surface Irregularities, or by Impinging particles of clot dislodged from the test chamber wall further upstream. Deposited coagulum downstream is generally smoother and more uniformly spread around the object, indicating a more ordered deposition pattern. In steady flow, clot is found to deposit in the wake of the test-bodies, as in Vorhauer's experiments, but in'pulsatile flow a rather strange clotting pattern is observed: An azimuthal band of clot adheres to the downstream side of both the tear-drop and the sphere whilst -virtually no clot deposits downstream of the cone.

Residence time experiments In the same apparatus around the, same test objects have revealed that the unexpected deposition in pulsatile flow occurs only in regions of stasis. Further analysis shows that whilst stasis is necessary for milk clot depostion, It Is not a sufficient condition.

A modified Lee White Test, devised to determine the effects, if any, of agitation on the milk clotting reaction, Indicates that whilst agitation has no apparent influence on the length of the induction period, its effect on the adhesivity of the final clot to a solid surface is profound when agitation occurs at the end of the induction phase. Thus stasis, and some aspect of agitation (most probably high mass transfer rates) in the vicinity of a surface are proposed as the concomitant conditions for clotting on that surface.