Effects of gemfibrozil and ciprofibrate on plasma levels of tissue-type plasminogen activator, plasminogen activator inhibitor-1 and fibrinogen in hyperlipidaemic patients

Abstract

Evaluation of fibrate treatment in humans has focused primarily on its anti-lipidaemic effects. A potentially favourable haemostasis-modulating activity of fibrates has also been recognized but the data are not consistent. We sought to learn more about this variability by examining the effects of gemfibrozil and ciprofibrate on plasma levels of tissue-type plasminogen activator (t-PA), plasminogen activator inhibitor-1 (PAI-1) and fibrinogen in primary hyperlipidaemic patients after six and twelve weeks of treatment using different assay systems for PAI-1 and fibrinogen. Although both fibrates effectively lowered triglyceride and cholesterol levels, no effect on the elevated baseline antigen levels of t-PA and PAI-1 was observed after fibrate treatment. However, both fibrates influenced plasma fibrinogen levels, albeit in a different way. Fibrinogen antigen levels were elevated by 17.6% (p <0.05) and 24.3% (p <0.001) with gemfibrozil after six and twelve weeks, respectively, whereas with ciprofibrate there was no effect. Using a Clauss functional assay with either a mechanical end point or a turbidity-based end point, no significant change in fibrinogen levels was seen after six weeks of gemfibrozil treatment. However, after twelve weeks, gemfibrozil enhanced functional fibrinogen levels by 7.2% (p <0.05) as assessed by the Clauss mechanical assay, but decreased functional fibrinogen levels by 12.5% (p <0.0001) when a Clauss assay based on turbidity was used. After six or twelve weeks of ciprofibrate treatment, functional fibrinogen levels were decreased by 10.1% (p <0.001) and 10.5% (p <0.0001), respectively on the basis of Clauss mechanical and by 14.2% (p <0.001) and 28.2% (p <0.0001), respectively with the Clauss turbidimetric assay. A remarkable and consistent finding with both fibrates was the decrease in functionality of fibrinogen as assessed by the ratio of functional fibrinogen (determined by either of the two Clauss assays) to fibrinogen antigen. Taken together, our results indicate that at least part of the variability in the effects of fibrates on haemostatic parameters can be explained by intrinsic differences between various fibrates, by differences in treatment period and/or by the different outcomes of various assay systems. Interestingly, the two fibrates tested both reduced the functionality of fibrinogen.