ALBERT

Notes: Summary We have studied the effects of clofibrate treatment on glucose tolerance and plasma insulin, plasma triglyceride, cholesterol and non-esterified fatty acid (NEFA) levels, and on various haematological variables (including plasma fibrinogen level, red cell flexibility, whole blood viscosity, and plasma β-thromboglobulin level) in patients with mature-onset diabetes. Twenty-two patients (11 men and 11 women) were randomly allotted to treatment with clofibrate, 1 g twice daily, or a corn-oil placebo for 12 weeks, and then changed to the alternate medication for another 12 weeks. Half the patients took clofibrate in the first 12 weeks of the study, and half took the placebo. The patients stayed on their usual diet, and 13 also took tolbutamide before and during the trial. The trial was double-blind. At the beginning, middle and end of the trial fasting measurements were made, and plasma glucose, insulin, triglyceride, and NEFA concentrations were then measured repeatedly during the next 8 h (from 8.00 a. m. to 4 p. m.), to allow calculation of the mean 8-h concentration of these substances. In general, plasma concentrations of glucose, triglyceride, cholesterol, NEFA and fibrinogen were lower when the patients were taking clofibrate then when they were taking the corn-oil placebo, but higher when taking the placebo than at entry to the trial. We favour the explanation that clofibrate has lowered these concentrations, when compared with the placebo. The alternative interpretation, that 2 g per day of the placebo increases plasma concentrations of glucose, triglyceride, cholesterol, NEFA and fibrinogen, and that clofibrate has little effect, seems unlikely. The first interpretation, that clofibrate has a positive effect when compared with an inert placebo, has been adopted when interpreting the results. Clofibrate treatment led to a 15% lower fasting blood glucose level, and 11% lower mean 8-h glucose concentration than did placebo (p〈0.01) but it did not significantly change plasma insulin concentration. The fasting and mean 8-hour concentrations of plasma triglyceride and fasting plasma cholesterol concentrations were reduced by clofibrate (by 44%, 33% and 10% respectively, p〈0.05). Clofibrate decreased the fasting plasma NEFA level by 27% (p〈0.01), and the mean 8-h plasma NEFA concentration by 23% (p〈0.05). A weak relationship between the mean 8-h levels of plasma NEFA and plasma glucose (r=0.49, p〈0.05) was consistent with the suggestion that the change in plasma glucose could, in part, be due to a change in NEFA concentration. The mean plasma fibrinogen concentration was decreased 23% by clofibrate (p〈0.01). There was a positive correlation between the observed decrease during treatment and the baseline fibrinogen concentration (r=0.80, p〈0.001), i. e. the greatest decrease occurred in those subjects with the highest plasma fibrinogen concentrations. Whole blood viscosity fell slightly, but erythrocyte flexibility was not significantly changed by clofibrate. The mean haemoglobin concentration and leucocyte count fell slightly during clofibrate treatment and the platelet count rose. β-thromboglobulin was not affected. Clofibrate treatment was associated with rises in plasma albumin, urea, creatine kinase and aspartate aminotransferase, and falls in plasma bilirubin, γ-glutamyl-transpeptidase and alkaline phosphatase. Most of these changes occurred within the reference range.