An assessment of solute kinetics and the application of mathematical modelling in the haemodialysis process

• Main Author: Spalding, Elaine MacGregor
• Published: University of Edinburgh 2008
• Subjects: 616.6
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.662300

Mathematical modelling is utilised to provide potential explanations for the clearance characteristics of phosphate and beta2-microglobulin during chronic dialysis. The phosphate model is explored further with studies in acute renal failure and the effect of dialysis on intra-erythrocytic phosphate concentrations is assessed. Diurnal variation in phosphate concentration is explored. High haematocrit is not found to have a significant effect on the clearance of solutes across a wide range of molecule size. A four-pool model that can be applied in both acute and chronic renal failure is proposed to explain the observed kinetic behaviour of phosphate. Studies of intracellular phosphate concentrations fail to demonstrate release of phosphate from erythrocytic stores during dialysis. Diurnal variation in phosphate concentration is demonstrated in subjects with normal renal function and also in advanced chronic kidney disease. A multi-pool model explains the kinetic behaviour of beta2-microglobulin during dialysis. Beta-2-microglobulin deposition is assumed to be a staged process with some deposits easily accessible during dialysis and some more resistant to depuration. Patients receiving high-flux dialysis or haemodiafiltration are shown to have lower circulating beta2-microglobulin levels and less symptomatic dialysis related amyloid, but evidence of amyloid deposition is still found when assessed by a scintigraphic imaging technique. Age and duration of dialysis are shown to be the best predictors of symptomatic amyloid deposition. The results of the studies in this thesis indicate that, for solutes such as phosphate and beta2-microglobulin which have complex intra-dialytic kinetics, current dialysis techniques are insufficient to achieve adequate solute removal. It will be necessary to deliver longer or perhaps more frequent dialysis therapy in order to achieve this goal. Mathematical modelling facilitates understanding of the pathophysiology of the dialysis process and provides a platform for the development and monitoring of improved dialysis strategies.