Eur J Clin Pharmacol(1992) 43:219-223

@Springer-Verlag] 992

Bezafibrate lowers plasma lipids, fibrinogen and platelet aggregability in hypertriglyceridaemia E Pazzucconi, L. Mannucci, L. Mussoni, G. Gianfranceschi, P. Maderna, P. Werba, G. Franceschini, C. R. Sirtori, and E.Tremoli E. Grossi Paoletti Center,Institute of PharmacologicalSciences,Universityof Milan, Milan, Italy Received:August 1, 1991/Acceptedin revisedform:March2, 1992 Sunnnary. The effects of bezafibrate 400 mg/day (slow release formulation) on plasma lipids/lipoproteins and on selected haemostatic parameters were evaluated in a double blind cross-over study in patients with Type IIb and IV hyperlipoproteinaemia. Placebo treatment did not influence any of those parameters, but the drug significantly reduced plasma triglycerides ( - 45 %) and VLDL cholesterol, as well as causing a 12% fail in total cholesterol and a 20% increase in HDL cholesterol. Apo AI levels were significantly increased following bezafibrate and Apo B was reduced by about 20%. In addition to changes in the plasma lipid profile, bezafibrate reduced the sensitivity of platelets to the aggregatory effect of collagen, with no effect on TXB2 production. Fibrinogen levels after bezafibrate treatment were significantly lowered, the effect being more marked in patients with hyperfibrinogenaemia. Bezafibrate did not influence the plasma fibrinolytic profile. It is concluded that bezafibrate, besides its beneficial effects on the plasma lipoprotein profile, can exert beneficial changes on specific haemostatic parameters. Key words: Bezafibrate, Hyperlipoproteinaemia; fibrinogen, fibrinolysis, platelet aggregation, plasma lipids

Alterations in plasma cholesterol and triglycerides (TG) are associated with changes in several specific haemostatic variables. Elevated cholesterol levels may cause functional and biochemical changes in platelets, resulting in increased aggregability and enhanced generation of thromboxane A2 [1, 2]. An effect of elevated TG levels on the same parameters is less uniformly described, possibly because of the technical difficulty of monitoring aggregability in turbid plasma samples. More recently, attention has been directed to other haemostatic factors besides platelets that might be assodated with the enhanced thrombotic risk characteristic of hyperlipidaemia [3]. Fibrinogen, proposed as an inde-

pendent risk factor for stroke and myocardial infarction [4-6], may affect platelet aggregation by interacting with cellular integrins [7-9], and it may also affect blood viscosity and endothelial cell function [10, 11]. The fibrinogen level in plasma is, however, very largely determined by genetic factors [12] and is only weakly related to the cholesterol level [13]. Cigarette smoking, pregnancy and oral contraceptives may elevate plasma fibrinogen, whereas a reduction can be effected by a small number of pharmaceutical agents, including certain fibric acid derivatives, particularly bezafibrate [14]. While plasma fibrinogen levels do not seem to be clearly related to elevated plasma lipids, alterations in fibrinolysis typically occur in hypertriglyceridaemia [1518], both related to tissue type plasminogen activator (t-PA) and to plasminogen activator inhibitor (PAI-1) [19]. Elevated plasma PAL1 levels are associated with an increased incidence of cardiovascular disease [20, 21] and a direct correlation between plasma TG and PAI-1 levels has been found in young survivors from myocardial infarction [21]. Significant elevations both of t-PA and PAI-1 levels were recently discribed in Type IV patients [18]. Indirect impairment of the fibrinolytic system may also result from elevated plasma lipoprotein(a) (Lp(a)), which is strongly associated with an increased inddence of coronary [22, 23] and cerebrovascular [24] diseases. Plasma Lp(a) may displace plasminogen from its endothelial binding sites [25] and increase PAI-1 expression [26]. Bezafibrate, a lipid lowering drug belonging to the fibric acid series, has been shown in open studies to reduce plasma fibrinogen, particularly in patients with hyperfibrinogenaemia, as well as to diminish platelet aggregability [27]. In a controlIed study, a reduction of fibrinogen was also described in Type I diabetics [28]. Since this property is not shared by all fibrates, particularly not by gemfibrozil [29], it seemed of interest to evaluate a series of haemostatic variables in hypertrigtyceridaemic patients in a double blind placebo controlled trial.

220

TabLe1. Effects of Bezafibrate treatment on lipid and lipoprotein parameters Bezafibrate Total cholesterol (mmol-1 1) Triglycerides ( m m o l - 1 ~) LDL-C (mmol. 1-~) HDL-C (mmol. 1-~) VLDL-C (mmol.1 ~) Apo A - I (rag. dl-~) Apo A - I I (mg-d1-1) Apo B (mg. dl-~)

Placebo

TO

T1

TO

T1

7.1 (0.3) 4.0 (0.8) 4.2 (0.3) 1.1 (0.1) 1.8 (0.3) 120 (3.1) 26.3 (0.5) 156 (9.7)

6.2 (0.3)** 2.2 (0.4)*'*** 3.9 (0.3) 1.3 (0.1)* 1.0 (0.2)**** I32 (4.6)* 26.9 (0.6) 126 (7.4)***.**

7.0 (0.3) 4.3 (1.0) 3.9 (0.5) 1.1 (0.1) 2.0 (0.4) 125 (5.5) 25.8 (0.4) 155 (15.4)

7.1 (0.4) 3.5 (0.5) 4.1 (0.5) t.2 (0.1) 1.8 (0.3) 120 (3.4) 26.2 (0.5) 157 (10.5)

Values are the means with (SEM). * P < 0.05 and ** P < 0.01 vs initial value of bezafibrate treatment. *** P < 0.05 vs the final value of placebo treatment. TO = baseline; T1 after 8 week treatment

Materials and methods

Platelet studies

Patients

Blood was anticoagulated with 0.13 mol.1-~ trisodium citrate (9:1 v/v) and platelet poor plasma (PPP) was obtained as previously described [2]. Platelet aggregation was determined by the Born turbb dimetric technique [37] in an Elvi 840 aggregometer (Elvi Logos, Milan, Italy), using collagen (Mascia Brunelli, Milan, Italy) as the aggregating agent. For each subject, the concentration of collagen giving a 50 % decrease in optical density (ACs0) was defined on the basis of a linear regression obtained by plotting the concentrations of collagen used versus the decrease in optical density determined 5 rain after the addition of collagen. Thromboxane Ba (TXB2) formation was determined in PRP incubated for 7.5 min at 37 °C (stirring 1000 rpm) with collagen (2.5 and 5 gg/ml). The reaction was stopped by the addition of 2.5 ml methanol and TXB2 formation determined by a specific radioimmunoassay [2]. [?H]TXB2 was purchased from New England Nuclear (Boston, MA, USA), synthetic TXB2 was from Sigma Chemical Co. (St. Louis, MO, USA).

Twelve hypertriglyceridaemic patients, 6 with a diagnosis of Type IV and six with Type IIb hyperlipoproteinaemia, were enrolled [30]. The subjects, 8 m and 4 f, age range 24-68 y, were not on any concomitant therapy and they were asked to avoid any intake of aspirin or related drugs. Patients with hypothyroidism, gastrointestinal or liver disorders, diabetes mellitus, alcoholism, history of coagulation disorders and hypertension were excluded. To allow "in vitro" platelet aggregation studies, patients with T G levels > 400 rag. dl-~ were not included. All patients underwent a complete physical examination and a standard battery of biochemical and haematologic tests before and at the end of the study.

Protocol After selection for the study, two baseline fasting blood samples were collected from all patients at 4 week intervals. Variation exceeding 8 % and 20 % in total cholesterol and TG levels was considered an exclusion criterion. Immediately after the second sample, patients were allocated at random to a period of 8 weeks on either treatment A (Bezafibrate 400 mg once a day, as a slow release formulation) or B (placebo). This was followed by a 4 week wash-out period and the alternate treatment was then given for a further 8 weeks. A complete plasma lipid/lipoprotein profile, including total cholesterol (TC), lipoprotein cholesterol distribution (in low density, LDL-C; very low density, VLDL-C; and high density, HDL-C) and total TG, was performed at certain times ( - 2 , 0, 8, 12, and 20 weeks). At those times apolipoprotein (apo) AI, AII, B, ptatelet aggregation, fibrinolytic parameters and fibrinogen levels were also determined. Patients were also seen at intermediate intervals (4 and 16 weeks) in order to assure compliance with diet and drug/placebo administration. Standard lipid/lipoprotein levels, determined at these visits (TC, TG, HDL-C), are not reported here.

Laboratory methods Lipids/lipoproteins. ~Ibtal cholesterol (TC) and triglycerides (TG) were measured by enzymatic methods [31, 32]; HDL-C was measured after precipitation of VLDL and LDL with dextran sulphate-MgClz [33]; apolipoproteins according to Rifai et al. [34]. VLDL (d < 1.006 g.ml -~) and LDL (d 1.006-.1.063 g. ml -t) were separated by preparative ultracentrifugation [35]. Lp(a) was determined with a commercially available ELISA (Therumo Medical Corporation Macra Lp(a) Enzyme Immunoassay Kit), which uses a monoclonal anti-Lp(a) coated on microtitre plates to extract Lp(a) from undiluted plasma [36].

Fibrinogen Fibrinogen levels were measured on citrated plasma by the Clauss thrombin clotting method [38] using an automated coagulometer (Schnitger Gross, Mascia Brunelli, Milan, Italy).

Fibrinolysis Blood for measurements of fibrinolytic parameters was drawn from an antecubital vein of the forearm of fasting subjects with minimal stasis (between 9:00 and 10:00 h) after 15 min supine rest and 10 min venous stasis, the latter produced in the controlateral arm by a sphygmomanometer cuff inflated to the mean blood pressure of the patient. Blood was anticoagulated with 0.13 mol. 1-I trisodium citrate (9:1, v/v), in precooled plastic tubes and kept on crushed ice until cen trifuged. Platelet poor plasma, obtained after 20 min centrifugation at 800 x g at 4 °C, was stored in small aliquots at - 70 °C until analysed. The euglobulin Iysis area (ELA) was measured using the euglobulin fractions obtained after acidification of diluted plasma (1:10) at pH 5.9 with acetic acid (0.25%) at 4°C. 30 ~tl of resuspended euglobulins were placed on the surface of two different fibrin plates and the area, measured after incubation at 37°C for 18 h, was expressed as diameter (ram) of lysis. t-PA antigen levels were determined by an ELISA IMUBIND-5 (American Diagnostica, USA), following the manufacturer's recommendations. This assay detects both free t-PA and t-PA complexed with inhibitors and includes the use of quenching and irrelevant antibodies to exclude false positive results. Values are the means of duplicate measurements, expressed as ng. ml 1. Functional PAL1 activity was assayed by a two-stage indirect enzymatic test (Ortho Diagnostica System, Milan, Italy). Briefly,

221 mg/dl 4O0

300

i

200

100

TO T1 Bezafibrate

TO

T1 Placebo

Fig. L Effect of bezafibrate on fibrinogen levels; TO = baseline, T1 after 8 week treatment; * = P < 0.05 vs the initial values of the bezafibrate treatment period (TO) t-PA was added in excess (40 and 30 IU. ml-~) to undiluted and diluted (1:2) plasmas and after 10 rain incubation at room temperature, the residual t-PA activity was measured using the chromogenic substrate S-2251 in the presence of plasminogen and CNBr fibrin(ogen) fragments. PlasmaPAIantigen levels were determined using a commercially available ELISA kit (IMULYSE, Biopool AB, UmeS, Sweden). It includes the use of quenching and irrelevant antibodies to exclude false positive results. This assay detects both active and inactive molecules, Values are means of duplicate measurements, expressed as ng. ml- ~.

Statistical analysis Statistical analysis of the data was performed by two-way analysis of variance (ANOVA). The carry over effect was evaluated by the Grizzle procedure. Results

All subjects concluded the study without undue side effects. While placebo treatment did not influence the evaluated lipid and lipoprotein parameters, bezafibrate induced a significant fall in plasma T G levels ( - 45 % versus basal and post-placebo values). V L D L cholesterol levels were also reduced to a similar extent ( - 43 %). Small but significant reductions in mean total ( - 12 %) and L D L ( - 7 %) cholesterol levels were also observed following bezafibrate, concomitant with an almost - 20 % reduction in apo B levels. Mean H D L - C levels were significantly increased following bezafibrate treatment ( + 20%). Similarly apo A I was increased ( + 10%), whereas Apo AII levels remained unchanged throughout the study (Table 1). Basal Lp(a) levels showed a wide range of variation (range 2.1-52.7 mg. dl- 1), treatment with bezafibrate did not modify plasma levels of this atherogenic tipoprotein fraction (20.6 (10.3) and 22.8 (11.5) mg.dl -~ before and after 8 weeks of treatment). Interesting changes in selected haemostatic parameters were noted together with the changes in serum lipid profile. Fibrinogen levels showed an average reduction after bezafibrate of 21% (Fig. 1).

While platelet aggregabitity was not influenced by placebo, drug treatment reduced platelet sensitivity to collagen, as shown by the increases in the concentrations of collagen required to elicit 50 % aggregation (Table 2). In Fig. 2 is shown an example of aggregatory tracings at baseline and following bezafibrate treatment. The drug reduced the response of PRP to the aggregatory effect of collagen, the effect being overcome at increasing collagen concentrations. No effect of either placebo or bezafibrate on TXB2 synthesis by collagen stimulated PRP was observed (Table 2). Unlike the case for plasma fibrinogen and platelet aggregation, fibrinolytic activity, both resting and after venous occlusion, was not modified by the two treatments (placebo or bezafibrate). Similarly, no effects on t-PA levels, both resting and after venous occlusion, and on PAI-1 antigen and activity were obselwed (Table 3).

Discussion

Bezafibrate, administered for 8 weeks to patients with elevated T G levels, with or without abnormal levels of total and LDL-cholesterol, can improve not only plasma lipid levels but also favorably influence specific haemostatic parameters. This study investigated these latter additional properties of the drug within a controlled protocol. The effects of bezafibrate on triglyceridaemia confirmed previous data in the literature, consistent with the effects of the drug on V L D L catabolism through activation of both lipoprotein and possibly hepatic lipase [39]. Reductions in T G and V L D L cholesterol were not accompanied by changes in the opposite direction of L D L - C levels [40, 41]. On the contrary, L D L - C and apo B were reduced at the end of the treatment, both in type IV and IIB patients, thus indicating that bezafibrate may improve abnormal L D L receptor binding properties of L D L and/or correct an elevated L D L biosynthesis [42]. H D L - C and apo AI levels were also raised after treatment, without changes in apo AII. In this short term study, plasma Lp(a) levels were apparently not modified. A n effect of bezafibrate on Lp(a) was previously described by others following longer term treatment [43]. It is possible that this study was too short to allow a consistent effect of the drug on Lp(a); on the other hand, since mean Lp(a) levels of treated patients were mostly within normal limits (8 out 12 had levels below 15 mg. dl- 1), a further reduction in the levels of this potentially atherogenic lipoprotein was not likeb: Table 2. Effects ofbezafibrate treatment on platelets

Bezafibrate TO T1

Placebo TO

T1 AC.s0for collagen (gg.mt-l) 0.50 (0.08) 0.85 (0.15)* 0.60 (0.08) 0.74 (0.15) TXB2a(ng-ml -~) 56.1(6.0) 77.4(15.0) 66.7(7.4) 61.4(26.2) Values are the means with (SEM). * P < 0.05 vs initial value of bezafibrate treatment. "TXB2 synthesis stimulated by 2.5 btg-ml- ~collagen. TO = baseline; T1 after 8 week treatment

222 COLLAGEN

COLLAGEN

0.5 gg/ml

1 ~g/ml

bezafibrate

co

bezafibrate Baseline Baseline

Table 3. Effects of bezafibrate treatment on fibrinolyticparameters

Bezafibrate

Placebo

TO

TO

T1

T1

ELA(mm)

b.o. 10.0(0.6) 10.2(0.6) 10.6 (1.0) 10.7 (1.0) a.o. 18.8 (1.7) 18.6 (1.3) 15.2 (1.9) 17.6 (2.3)

t-PA:ag(ng/ml)

b.o. 10.2 (1.2) 10.1 (1.0) 11.2 (1.0) 11.1 (1.0) a.o. 36.5 (7.7) 26.0 (2.7) 25.6 (3.9) 31.7 (5.l)

PAI activity (U/ml) b.o. 16.7 (1.6) 27.3 (3.2) 22.4 (3.9) 24.0 (4.2) PAI-l:ag (ng/ml)

b.o. 15.2 (2.6) 24.0 (2.6) 23.1 (5.2) 20.3 (2.7)

Values are the means with (SEM). b. o. before venous occlusion; a. o. after venous occlusion. TO = baseline; TI after 8 week treatment

Selected haemostatic parameters were clearly improved by bezafibrate. Fibrinogen levels have been previously shown to be reduced in open studies on hyperfibrinogenaemic patients [27], as well as in cross over studies on hypertriglyceridaemics and diabetics [44]. As already pointed out, not all fibrates reduce fibrinogen and most notably, there is general agreement about a lack of activity of gemfibrozil [29]. Since this latter compound has a powerful preventive effect against cardiovascular disease in mildly hyperlipidaemic patients [44], it is possible that this additional property of bezafibrate might play a significant role in prevention of cardiovascular disease. Consistently with earlier data [27], this reduction in fbrinogenaemia occurred to a larger extent in hyperfibrinogenaemic patients, thus offering a very safe and effective treatment for such patients, without exposing normal fibrinogenaemic patients to a significant risk. Interestingly levels of fibrinogen after bezafibrate tended to be within a range, associated with a lower risk of cardiovascular disease [45]. Following treatment with bezafibrate, the sensitivity of platelets to the effects of collagen was significantly reduced. That bezafibrate, similarly to other fibrates, could influence platelet aggregation induced by ADP was already reported by other authors [27]. Since bezafibrate influenced the aggregation induced by collagen with no effect on TXA; synthesis, its antiaggregatory effect is likely to take place by mechanism(s) other than the inhibition of platelet cyclooxygenase. Taken together these data suggest that bezafibrate may exert an antithrombotic activity, eventually additive to that of cyclooxygenase inhibitors.

Fig,2. Aggregatory tracings of asubjectbefore (baseline) and after bezafibrate treatment. The inhibitory activity of the drug is evident when the threshold aggregatory concentration of collagen is used to stimulate PRP

Since patients with elevated TG levels with or without concomitant CHD have been previously reported to have impaired fibrinolytic activity [21], it seemed of interest to evaluate parameters of fibrinolysis in this study. The different fibrinolytic parameters were, however, unchanged after bezafibrate treatment in spite of a reduction of plasma TG. At present these data are of difficult explanation. However, although most studies [15-17] have indicated a direct correlation between TG levels and PAI-1, the evidence that TG reduction is accompanied by normalization of PAI-1 levels is still lacking. A predominant genetic regulation of PAI-1 levels has recently been suggested [46]. In conclusion, this controlled investigation clearly indicates that bezafibrate, besides its beneficial effects on the plasma lipoprotein profile, can positively influence specific haemostatic parameters involved in the thrombotic and atherosclerotic processes, in particular platelet aggregability and fibrinogenaemia. In this latter case, the effect is more pronounced under conditions in which levels tend to be elevated. These observations suggest that bezafibrate may differ from other commonly used fibrates, in particular gemfibrozil, that besides being ineffective in plasma fibrinogen [29], in similar patients in our hands, also did not modify platelet aggregability [47]. This activity profile is particularly attractive in evaluating the long term therapeutic potential of bezafibrate in patients with a high atherosclerotic risk.

References Tremoli E, Maderna R Sirtori M, Sirtori CR 1979) Platelet aggregation and malondialdehyde formation in type II a hypercholesterolemic patients. Haemostasis 8:4%53 2. Tremoli E, Maderna R Colli S, Morazzoni G, Sirtori M, Sirtori CR (1984) Increased platelet sensitivityand thromboxane B2formarion in type-II hyperlipoproteinaemicpatients. Eur J Clin Invest 14:329-333 3. Stuart J, George A J, Davies A J, Auktand A, Hurlow RA (1981) Hematological stress syndrome in atherosclerosis. J Clin Pathol 34:464-467 4. Balleisen L, Bailey J, Epping PH, Shnlte H, van de Loo J (1985) Epidemiological study on factor VII, factor VIII and fibrinogen in an industrial population, t. Baseline data on the relation to the age, gender, body weight, smoking, alcohol, pill using, and menopause. Thromb Haemost 54:475-479

223 5. Meade TW, Brozovic M, Chakrabarti RR, et al (1986) Haemostatic function and ischemic heart disease: Principal results of the Northwick Park Heart Study. Lancet 2:533-537 6. Welin L, Svardsudd K, Wilhelmsen L, Larsson B, Tibblin G (1987) Analysis for risk factors of stroke in a cohort of men born in 1913. N Engl J Med 317:521-526 7. Marguerie GA, Thomas-Maison N, Larrieu MJ, Plow EF (1982) The interaction of fibrinogen with human platelets in a plasma milieu. Blood 59:91-95 8. Nurden AT, Caen JP (1974) An abnormal platelet glycoprotein pattern in three cases of Glanzmann's thrombasthenia. Br J Haemato128:253-260 9. Bennet JS, Vilaire G (1979) Exposure of platelet fibrinogen receptors by ADP and epinephrine. J Clin Invest 64:1393-1401 10. Schweiger B, Pohl I, Driessen H, Heidtmann H, SchmidSchoembein H (1981) Influence of fibrinogen concentration on flow velocity of red blood cells in rat mesentery. Bibl Anat 20: 230-233 1i. Dejana E, Languino LR, Polentarutti N e t al. (1985) Interaction between fibrinogen and cultured endothelial cells. Induction of migration and specific binding. J Clin Invest 75:11-18 12. Humphries SE, Cook M, Dubowitz M, Stirling Y, Meade TW (1987) Role of genetic variations at the fibrhlogen locus in determination of plasma fibrinogen concentrations. Lancet 1: 14521455 13. Wilhelmsen K, Svardsudd K, Korsan-Bergsten K, Larsson B, Welin L, Tibblin G (1984) Fibrinogen as a risk factor for stroke and myocardial infarction. N Engl J Med 311: 501-505 14. Di Minno G, Mancini M (1990) Measuring plasma fbrinogen to predict stroke and myocardial infarction. Arteriosclerosis 10:1-7 15. Epstein SE, Rosing DR, Brakman R Redwood DR, Astrup T (1970) Impaired fibrinolyticresponse to exercise in patients with type-IV hyperlipoproteinaemia. Lancet II: 631-633 16. Ahmad M, Saku K, Glas-Greenwalt E Kashyap ML (1985) Inhibitor and activator of plasminogen in patients with hyperlipoproteinemias. Thromb Haemost 54:266 17. Lowe GDO, McArdle BM, Stromberg R Lorimer AR, Forbes CD, Prentice CRM (1982) Increased blood viscosity and fibrinolytic inhibitor in type II hyperlipoproteinaemia. Lancet I: 472475 18. Mussoni L, Mannucci L, Sirtori M, Camera M, Maderna R Sironi L, Tremoli E (in press) Hypertriglyceridemia and regulation of fibrinolytic activity. Arteriosclerosis and Thrombosis 19. Bachmann F (1987) Fibrinolysis. In: Verstraete M, Vermylen J, Lijnen HR, Amout J, eds. Thrombosis and Haemostasis. Leuven University Press, Leuven, pp 227-265 20. Mehta J, Mehta R Lawson D, Saldeen T (1987) Plasma tissue plasminogen activator inhibitor levels in coronary artery disease: correlation with age and serum triglyceride concentrations. J Am Coll CardioI 9:263-268 21. Hamsten A, Wiman B, de Faire U, Blombfick M (1985) Increased plasma levels of a rapid inhibitor of tissue plasminogen activator in young survivors of myocardial infarction. N Engl J Med 313:1557-1563 22. Dahlen G, Berg T, Gillnas T, Ericson C (1975) Lp(a) lipoprotein/pre-beta 1-1ipoprotein in Swedish middle aged males and in patients with coronary disease. Clin Genet 7:334-341 23. Frick MH, Dahlen G, Berg K, Valle M, Hekali P (1978) Serum lipids in angiographically assessed coronary atherosclerosis. Chest 73:62-65 24. Murai A, Miyahra T, Fujimoto N, Matsuda M, Kameyama M (1986) Lp(a) lipoprotein as a risk factor for coronary heart disease and cerebral infarction. Atherosclerosis 59:199--204 25. Miles LA, Plow EF (1990) Lp(a): An interloper into the fibrinolytic system? Thromb Haemost 63:331-335 26. Etingin OR, Hajjar DR Hajjar KA, Harpel PC, Nachman RC (I991) Lipoprotein(a) regulates plasminogen activator inhibitor i expression in endothelial cells. JBC 266:2459-2465 27. Niort G, Bulgal-el!i A, Cassader M, Pagano G (1988) Effect of short-term treatment with bezafibrate on plasma fibrinogen. A

platelet activation and blood flterability in atherosclerotic hyperfibrinogenemic patients. Atherosclerosis 71: 1t 3-119 28. Durrington PN, Winocour PH, Bhatnagar D (1990) Bezafibrate retard in patients with fi~sulin-dependent diabetes: effect on serum lipoproteins, fibrinogen and glycemic control. J Cardiovasc Pharmaco116 [Suppl 9]: $30-$34 29. Andersen R Smith R Seljeflot I, Brataker S, Arnesen H (1990) Effects of gemfibrozil on lipids and haemostasis after myocardial infarction. Thromb Haemost 63:174-177 30. WHO Memorandum (1972) Classification of hyperlipoproteinemias. Circulation 45:501-508 31. ROschlau R Bernt E, Gruber W (1974) Enzymafische Bestimmung des Gesamt-Cholesterins im Serum. Z Klin Chem Klin Biochem 12:403-407 32. Bucolo G, David H (1973) Quantitative determination of serum triglycerides by the use of enzymes. Clin Chem 19:476-482 33. Warnick GR, Benderson J, Albers JJ (1982) Dextran sulphate precipitation procedure for quantitation of high density lipoprotein. Clin Chem 28:1379-1388 34. Rifai N, King ME (1986) Immunoturbidimetric assay of lipoproteins AI, A I I and B in serum. Clin Chem 32:95%961 35. Havel RJ, Eder HA, Bragdon JH (1955) The distribution and chemical composition of ultracentrifugally separated lipoproreins in human serum. J Clin Invest 34:1345-1353 36. Giacherio D, Matz K (1990) Evaluation of an enzyme immunoassay kit for the determination of lipoprotein(a). Clin Chem 36 (1990) 955 37. Born GVR (1963) The aggregation of blood platelets. J Physiol (Loud) 168:178-195 38. Clauss A (1957) Gerinnungsphysiologische Schnellmethode zur Bestimmung des Nbrinogens. Acta Haemato117: 237-245 39. Cardonnet LJ, Eberhardt D (1982) Hypolipaemic effect of bezafibrate at doses of 400 mg and 600 mg daily in different types of primary hyperlipoproteinaemias: comparison with other treatments. In: Crepaldi et al. (eds) Lipoprotein metabolism and therapy of lipid disorders. Excerpta Medica, Amsterdam, pp 138-145 40. Hutt V, Wechsler JG, Kl6r H-U, Ditschuneit H (1983) Changes in the concentration and composition of lipids lipoproteins in primary hyperlipoproteinaemia during treatment with bezafibrate. Arzneimittelforschung 33:1185-1190 41. Oster P, Schlierf G, Lang PD, Mordasini R, Vollmar J (1985) Diurnal lipid and lipoprotein profiles with bezafibrate and clorofibrate in healthy volunteers. Pharmatherapeutica 4:267-277 42. Monk JR Todd PA (1987) Bezafibrate: a review. Drugs 33: 539576 43. Schwartzkopff W, Bimmermann A (1988) ttyperlipoprotein~tmie Typ II a, IIb und IV. Mtinch Med Wschr 130:422 44. Almdr LO, KjellstrOm T (1986) The fibrinolytic system and coagulation during bezafibrate treatment and hypertriglyceridemia. Atheroscterosis 61:81-85 45. Kannel WB, Wolf PA, Castelli WP, D'Agostino RP (1987) Fibrinogen and risk of cardiovascular disease. The Framingham Study. JAMA 258:1183-1186 46. Dawson S, Hamsten A, Wiman B, Henney A, Humphries S (1991) Genetic variation at the plasminogen activator inhibitor-1 locus is associated with altered levels of ptasma pIasminogen activator inhibitor-1 activity. Arteriosclerosis Thrombosis 11: 183190 47. Sirtori CR, Franceschini G, Oianfranceschi G, Sirtori M, Montanari G, Tremoli E, Maderna R Colli S, Zoppi F (1987) Effects of gemfibrozil on plasma lipoprotein-apolipoprotein distribution and platelet reactivity in patients with hypertxiglyceridemia. J Lab Clin Med I10:279-286 Prof. Elena Tremoli Institute of Pharmacological Sciences University of Milan Via Balzaretti 9 1-20133 Milano - MI Italy