An enzymatic and a novel capillary gas-liquid chromatographic assay have been developed and compared for the analysis of red blood cell sorbitol concentrations. Factors affecting the performance of both assays were investigated and the analytical conditions optimised. The enzyme assay was found to be accurate and precise but lacked specificity with at least xylitol also acting as a substrate. The capillary GLC assay was accurate and precise with the advantage of increased specificity for polyol determinations. The enzyme assay was used to determine the response of erythrocyte sorbitol content in a clinical trial of diabetics treated with oral Sorbinil a novel aldose reductase inhibitor. The sorbitol content was reduced and this trend was reversed upon cessation of Sorbinil treatment. The capillary GLC method was used to determine the daily effect of oral Sorbinil upon erythrocyte sorbitol and myo-inositol concentrations in a clinical trial with diabetic subjects. It was found that Sorbinil therapy reduced the raised erythrocyte sorbitol by 60% within the first 24 hours of treatment at a lower dose than used in previous studies. The erythrocyte myo-inositol content of diabetic subjects was not altered by Sorbinil therapy. However, it was found, for the first time, that the myo-inositol level in diabetics was found to be significantly different to healthy, normal non-diabetic subjects. The daily variation in erythrocyte sorbitol and myo-inositol in normal non-diabetic subjects was investigated. The levels were found to be very consistent both for individuals and as a group and were significantly different to the levels found in diabetic subjects. The activity of erythrocyte aldose reductase was found to be greater in diabetics compared to normals. Capillary GLC was used to confirm the erythrocyte accumulation of sorbitol upon incubation with glucose. The concentration of other components in the erythrocyte were seen to change upon glucose incubation. The increase of one component was initially identified by its retention as glucosamine. An attempt to correlate the erythrocyte aldose reductase activity with the degree of severity of subjects with diabetic neuropathy was unsuccessful. Further work is required to optimise the assay conditions. It is concluded that the erythrocyte sorbitol levels can be used in conjunction with other tests and clinical observations to manage the long term complications associated with diabetes. In particular, the enzyme assay may be of use as a screening test. The capillary GLC assay will be of most use in further studies to determine the polyol and carbohydrate changes and their role in the onset of neuropathy.

Thesis (M. Phil.)--University of Hong Kong, 2009. / Includes bibliographical references (leaves 89-111) Also available in print.