Inhibition of Thromboxane Biosynthesis and Platelet Function by Simvastatin in Type IIa Hypercholesterolemia
Abstract Thromboxane A2 (TXA2) biosynthesis is enhanced in the majority of patients with type IIa hypercholesterolemia. Because simvastatin (a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor) was previously shown to reduce platelet aggregation and TXB2 production ex vivo, we investigated TXA2 biosynthesis and platelet function in 24 patients with type IIa hypercholesterolemia randomized to receive in a double-blind fashion simvastatin (20 mg/d) or placebo for 3 months. The urinary excretion of 11-dehydro-TXB2, largely a reflection of platelet TXA2 production in vivo, was measured by a previously validated radioimmunoassay technique. Blood lipid levels and urinary 11-dehydro-TXB2 excretion were significantly (P<.001) reduced by simvastatin. In contrast, placebo-treated patients did not show any statistically significant changes in either blood lipids or 11-dehydro-TXB2 excretion. The reduction in 11-dehydro-TXB2 associated with simvastatin was correlated with the reduction in total cholesterol (r=.81, P<.0001), LDL cholesterol (r=.79, P<.0001), and apolipoprotein B (r=.76, P<.0001) levels. Platelets from patients with type IIa hypercholesterolemia required significantly (P<.01) more collagen and ADP to aggregate and synthesized less TXB2 in response to both agonists after simvastatin therapy. Bleeding time, platelet sensitivity to Iloprost, and blood lipoprotein(a) and HDL cholesterol levels were not significantly affected by either treatment. We conclude that enhanced TXA2 biosynthesis in type IIa hypercholesterolemia is, at least in part, dependent on abnormal cholesterol levels and/or other simvastatin-sensitive mechanisms affecting platelet function.
Presented in part at the 66th Scientific Sessions of the American Heart Association, Atlanta, Ga, November 8-11, 1993, and published in abstract form (Circulation. 1993;88:519).
- Received August 9, 1994.
- Accepted November 15, 1994.
A high incidence of atherosclerosis and thrombotic complications has been associated with type IIa hypercholesterolemia.1 Platelet aggregation and arachidonic acid metabolism via the prostaglandin H (PGH) synthase pathway have been reported to be abnormal in this setting and to contribute, at least in part, to enhanced thrombotic risk.2 3 4 5 6 Altered platelet function has been related to the cholesterol content of platelets, possibly reflecting an exchange between plasma lipoproteins and platelets.7
We recently showed that thromboxane A2 (TXA2) biosynthesis, as reflected by urinary 11-dehydro-TXB2 excretion, is enhanced in the majority of patients with type IIa hypercholesterolemia and suggested that this is partly a consequence of abnormal cholesterol levels.8 Low-dose aspirin largely suppressed increased thromboxane metabolite excretion, thus suggesting that it reflected platelet activation occurring in vivo.8 In a preliminary uncontrolled study, administration of simvastatin, a selective inhibitor of the enzyme 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, to 10 patients with type IIa hypercholesterolemia was associated with statistically significant reductions in both blood cholesterol levels and urinary 11-dehydro-TXB2 excretion.8
In the present study, we sought to verify whether enhanced TXA2 biosynthesis can be reduced through inhibition of cholesterol biosynthesis in type IIa hypercholesterolemia by randomly assigning 24 patients to 3-month treatment with simvastatin or placebo in a double-blind fashion. Moreover, we correlated in vivo and ex vivo indices of platelet function with simvastatin-induced lipid changes.
Twenty-four patients with type IIa hypercholesterolemia (14 women, 10 men; age, 48±15 years; range, 23 to 67 years) were recruited into this study. Type IIa hypercholesterolemia was defined in accordance with World Health Organization criteria,9 based on the determination of plasma total cholesterol and triglyceride levels and lipoprotein fractionation. Patients were selected on the basis of indication to lipid-lowering drug therapy, no contraindication to simvastatin, and willingness to participate in the study. We excluded patients with diabetes mellitus; impaired hepatic function; secondary hypercholesterolemia; history of alcoholism (drug abuse); and concomitant treatment with anticoagulants and antiplatelet drugs, including aspirin, corticosteroids, and theophylline. We also excluded patients with a history of and/or clinical examination positive for evidence of macrovascular complications determined on the basis of clinical symptoms, ECG monitoring during exercise, Holter monitoring, carotid Doppler testing, and Doppler echographic study of the lower limbs. Patients with renal disease (creatinine clearance <80 mL/min, serum creatinine >2 mg/dL, urine albumin excretion >300 mg/d) were also excluded. None of the patients had taken any drugs known to affect lipid metabolism or platelet function for at least 4 weeks before the start of the study. Patients were on a lipid-lowering diet (30% lipids, 52% carbohydrates, 18% proteins) with a polyunsaturated/saturated fat ratio of about 1.3 for at least 8 weeks before the study. Four patients were hypertensive. Eight patients were current smokers, and 16 had never smoked or had stopped smoking at least 2 years before the study.
The patients were randomized in two parallel groups of 12 patients each to receive in a double-blind fashion a fixed dose of simvastatin (20 mg once daily in the evening) or placebo for 3 months. Informed consent was obtained from each patient, and the protocol was approved by the institutional review board.
Twelve-hour urine samples (8 pm to 8 am) were obtained at baseline, on the seventh day, and at the end of the first, second, and third months of treatment. Peripheral venous blood samples were drawn with subjects in the fasting state at the same time points for lipid and safety measurements (creatine phosphokinase, transaminases, alkaline phosphatase, hemoglobin, and hematocrit). Bleeding time was also measured at the corresponding clinical visit.
To verify whether simvastatin or its major metabolite has a direct inhibitory effect on platelet TXA2 production, we incubated 1-mL whole-blood samples from three healthy volunteers (two women, one man; age 27 to 31 years) with simvastatin or its hydroxyacid metabolite L-654,969 (obtained from Merck Sharp & Dohme Research Laboratories through the courtesy of Dr Luigi Carratelli) at 1, 10, and 100 ng/mL for 1 hour at 37°C. Serum TXB2 was measured as a reflection of the platelet biosynthetic capacity in response to endogenously formed thrombin.10 This range of concentrations encompasses the peak plasma levels of simvastatin and its major metabolite after oral dosing in humans.11
Lipid and Platelet Function Measurements
Lipids and apoproteins (apo) were determined immediately after sampling; samples for lipoprotein(a) [Lp(a)] assay were frozen and stored at −70°C until assayed. Total cholesterol (TC) and triglycerides (TG) were measured by enzymatic methods.12 13 HDL cholesterol (HDL-C) was measured after precipitation of lipoproteins containing apoB by phosphotungstic acid/magnesium chloride,14 apoA-I and apoB by the nephelometric method,15 and Lp(a) by radioimmunoassay.16 LDL cholesterol (LDL-C) was calculated by the formula of Friedewald et al17 : LDL-C=TC−(TG/5+HDL-C).17 The quality control for lipids and apoproteins was performed as previously described18 ; Lp(a) radioimmunoassay underwent an interlaboratory standardization program with the Laboratory of Biochemistry of the National Health Institute, Helsinki, Finland. Intra-assay and interassay coefficients of variation for all assays were <3% and <5%, respectively.
For platelet studies, blood was collected into 3.8% sodium citrate (1 mL for 9 mL of blood). Platelet-rich plasma and platelet-poor plasma were prepared as previously described.2 Platelet aggregation was measured in an ELVI 840 aggregometer (Logos) according to the method of Born19 after adjustment for the number of platelets for each individual sample. For each agonist, the threshold aggregating concentration (TAC) was defined as the lowest concentration of the agent that caused a 50% to 60% increase in light transmittance within 3 minutes. The concentrations tested were 0.2 to 10 μmol/L for ADP and 0.1 to 8 μg/mL for collagen. Platelet TXB2 production was measured as previously described.2
Platelet sensitivity to Iloprost (a chemically stable analogue of prostacyclin) was determined as previously described.2
Bleeding time was measured with an automatic template device (Simplate II, General Diagnostics). The incisions were placed in the longitudinal direction on the volar surface of the upper forearm. Blood pressure was regulated according to manufacturer recommendations during bleeding time measurements. The same operator carried out all bleeding time determinations throughout the studies in the same room with a relatively constant temperature.
Serum TXB2 was measured by radioimmunoassay as previously described.10
Urinary 11-Dehydro-TXB2 Assay
Measurement of urinary 11-dehydro-TXB2 was performed by a previously validated radioimmunoassay technique.20 Immunoreactive 11-dehydro-TXB2 was extracted from 20-mL aliquots of each urine collection (pH adjusted to 4.0 to 4.5 with formic acid), run on Sep-Pak C18 cartridges (Waters Associates), and eluted with ethyl acetate. The eluate was subjected to silicic acid column chromatography and eluted with a benzene/ethyl acetate/methanol solution (60:40:30, vol/vol/vol). The overall recovery determined by the addition of 11-dehydro-[3H]TXB2 averaged 71±8%. Immunoreactive 11-dehydro-TXB2 eluted from silicic acid columns was assayed at a final dilution of 1:30 to 1:50 as described elsewhere.21
From the results obtained in a preliminary, open study,8 it was calculated that a sample size of 24 patients would be adequate to detect a 50% difference in 11-dehydro-TXB2 excretion between simvastatin and placebo, with α=.01 and β=.05. The data were analyzed by nonparametric methods to avoid assumptions about the distribution of the measured variables.22 A one-way ANOVA was performed. Subsequent pairwise comparisons were made by the Mann-Whitney U test. Correlations between eicosanoid measurements and other biochemical and functional measurements were assessed by stepwise regression analysis and multiple linear regression. All values are reported as mean±SD. Statistical significance was considered to be indicated by a probability value of <.05.
In Vivo Study
Table 1⇓ details the baseline characteristics of the 24 patients with type IIa hypercholesterolemia. Patients randomized to simvastatin were significantly older and had a lower collagen threshold in platelet aggregation studies than patients on placebo. The plasma lipid pattern is consistent with the characteristic features of this form of hypercholesterolemia. Moreover, the rate of excretion of 11-dehydro-TXB2 averaged 65.8±24.0 ng/h (range, 24 to 126 ng/h), consistent with our previous finding of enhanced metabolite excretion in this disease.8 No statistically significant differences in plasma lipid or urinary 11-dehydro-TXB2 measurements were found between patients randomized to simvastatin and those randomized to placebo.
Placebo treatment for 3 months was not associated with any statistically significant change in blood lipid levels or thromboxane biosynthesis. One patient could not tolerate the prescribed dose at the end of the first month of treatment because of gastric discomfort and withdrew from the study. After the code was broken at the end of the study, treatment identification indicated that he was on placebo. The five measurements obtained in the remaining 11 patients over the 3-month placebo treatment allowed determination of the intrasubject coefficient of variation for each of the blood and urinary indices of lipid metabolism and thromboxane biosynthesis. Table 2⇓ gives these values.
Inhibition of cholesterol biosynthesis by the HMG-CoA reductase inhibitor simvastatin (20 mg/d) was associated with statistically significant, time-dependent reductions in TC by up to 28%, LDL-C by up to 36%, apoB by up to 27% (Fig 1⇓), LDL-C/HDL-C ratio by up to 38%, and 11-dehydro-TXB2 by up to 52% (Fig 2⇓). Plasma TG, HDL-C, apoA-I, and Lp(a) levels were not modified by simvastatin treatment to any statistically significant extent (data not shown).
Percent changes in 11-dehydro-TXB2 excretion associated with simvastatin or placebo were correlated with changes in TC (r=.806, P<.0001; Fig 3⇓), LDL-C (r=.793, P<.0001), and apoB levels (r=.757, P<.0001).
Platelets from simvastatin-treated patients required significantly (P<.01) more ADP to aggregate and synthesized significantly less TXB2 in response to this agonist at 4 weeks of treatment and thereafter (Fig 4⇓). Moreover, simvastatin treatment was associated with a statistically significant (P<.01) increase in the TAC of collagen measured at 1, 2, and 3 months (1.7±0.4, 2.6±1.3, and 2.7±1.9 μg/mL, respectively) of therapy.
No statistically significant changes in these ex vivo measurements of platelet function were detected in placebo-treated patients (collagen TAC, 1.9±1.1, 1.9±1.1, 1.9±1.2, and 1.8±1.0 μg/mL at 1, 4, 8, and 12 weeks of treatment, respectively). Bleeding time and platelet sensitivity to the PGI2 analogue Iloprost were not modified by simvastatin or placebo to any statistically significant extent (data not shown).
In Vitro Study
Whole-blood TXB2 production, a measure of the maximum cyclooxygenase-dependent biosynthetic capacity of blood platelets, averaged 386±105 ng/mL of serum obtained from three healthy volunteers. One-hour incubation with 1, 10, and 100 ng/mL of simvastatin (361±134, 338±132, and 379±131 ng/mL, respectively) or its major metabolite L-654,969 (393±132, 426±83, and 425±85 ng/mL, respectively) did not affect whole-blood TXB2 production to any statistically significant extent.
Several lines of evidence suggest that platelet hyperreactivity might be an important factor contributing to the enhanced risk of thrombotic complications associated with a selective increase in plasma LDL-C (reviewed in References 7 and 237 23 ). Thus, earlier studies in patients with type IIa hypercholesterolemia consistently demonstrated enhanced platelet aggregation in response to a variety of agonists and increased conversion of arachidonate through the platelet PGH synthase pathway to form proaggregatory eicosanoids such as PGH2 and TXA2.2 3 4 5 6 7 8 Our previous study established that TXA2 biosynthesis in vivo is enhanced in the majority of patients with type IIa hypercholesterolemia without macrovascular complications and suggested that this is partly a consequence of abnormal cholesterol levels leading to persistent platelet activation.8
Results of the present study confirm and extend our earlier finding by showing that a 30% to 40% reduction in LDL-C is associated with virtual normalization of altered platelet aggregation ex vivo (Fig 4⇑) and with a 50% reduction in thromboxane metabolite excretion (Fig 2⇑). It has been argued that if the changes in platelet reactivity, sensitivity to PGI2, and eicosanoid biosynthesis reported in hypercholesterolemia are a direct consequence of elevated plasma lipid levels, then one would expect that cholesterol lowering by dietary means, plasmapheresis, or lipid-lowering drugs would result in normalization of platelet function.7 In fact, LDL apheresis, fibric acid derivatives, ion exchange resins, probucol, and HMG-CoA reductase inhibitors have all been evaluated for their effects on platelet function, although on small numbers of patients (reviewed in Reference 2323 ). In these studies, except those carried out with simvastatin, despite LDL-C reductions on the order of 10% to 30%, no change in platelet aggregation could be demonstrated ex vivo.23
Three previous open, uncontrolled studies examined the antiplatelet effects of simvastatin, given in doses ranging between 10 and 40 mg/d for 6 to 8 months, in groups of 10 to 12 patients with type IIa hypercholesterolemia.8 24 25 Reduced platelet aggregation and/or TXA2 biosynthesis were consistently found in these studies in association with 24% to 41% reductions in LDL-C.8 24 25 In comparisons of the effects of simvastatin with those of other lipid-lowering interventions, some consideration should be given to the reduction by simvastatin of the abnormally high LDL-C/HDL-C ratio.26
Because of the uncontrolled nature of previous simvastatin studies, we designed this double-blind, randomized, placebo-controlled study to examine the cause-and-effect relation between the lipid-lowering effect of the drug and its alleged antiplatelet effects. The present results provide unequivocal evidence that administration of simvastatin to patients with type IIa hypercholesterolemia is responsible for changes in platelet function, detectable ex vivo and in vivo. A number of findings argue that these changes in platelet function are a consequence of modifications in the plasma lipid pattern induced by simvastatin rather than a reflection of a direct effect of the drug on platelet biochemistry or function. These include (1) the consistent time-related patterns of lipid-lowering and antiplatelet effects (Figs 1⇑, 2⇑, and 4⇑), (2) the highly significant correlation between the two (Fig 3⇑), and (3) the lack of any direct effect of simvastatin or its main biologically active metabolite on platelet TXA2 production in vitro. However, our studies were not designed to address the issue of the mechanism of action of simvastatin in modulating platelet function, and we cannot exclude the possibility that other properties of the drug might be responsible for the observed changes.
We believe that the present findings have both conceptual and practical implications. First, they suggest that the highly reproducible increase in TXA2 biosynthesis, reflected by thromboxane metabolite excretion in placebo-treated patients, is sustained by simvastatin-sensitive determinants of hypercholesterolemia. Moreover, they suggest that whatever the molecular mechanism of action of simvastatin—and not necessarily other lipid-lowering maneuvers—in modulating platelet TXA2 biosynthesis, this effect may contribute to the overall impact of the drug on thrombotic risk. Finally, because of incomplete inhibition of TXA2 biosynthesis associated with simvastatin treatment, one should further explore the need for concomitant therapy with low-dose aspirin in the subset of hypercholesterolemic patients with persistently elevated levels of 11-dehydro-TXB2 excretion despite optimal dietary and pharmacological control.
This work was supported by a grant from Merck Sharp & Dohme Italia. The expert editorial assistance of Alessandra Migliavacca is gratefully acknowledged.
Ross R. The pathogenesis of atherosclerosis. In: Braunwald E, ed. Heart Disease. Philadelphia, Pa: WB Saunders Co; 1988:1135-1352.
Di Minno G, Silver MJ, Cerbone AM, Rainone A, Postiglione A, Mancini M. Increased fibrinogen binding to platelets from patients with familial hypercholesterolemia. Arteriosclerosis. 1986;6:203-211.
Tremoli E, Colli S, Maderna P, Baldassarre D, Di Minno G. Hypercholesterolemia and platelets. Semin Thromb Hemost. 1993;19: 115-121.
Davı̀ G, Averna M, Catalano I, Barbagallo C, Ganci A, Notarbartolo A, Ciabattoni G, Patrono C. Increased thromboxane biosynthesis in type IIa hypercholesterolemia. Circulation. 1992;85:1792-1798.
Classification of hyperlipidemias and hyperlipoproteinemias. Circulation. 1972;45:501-508. WHO memorandum.
Bologna M, Carlucci G, Biordi L, Mazzeo P, Iannarrelli R. HPLC determination of simvastatin and L 654,969 in plasma after oral administration of simvastatin to hypercholesterolemic patients. Int J Immunopathol Pharmacol. In press.
Roschlau P. Enzymatische bestimmung des gesamtecholesterins in serum. Chem Klin Biochem. 1974;12:226-227.
Wahalefeld AW. Triglyceride determination after enzymatic hydrolysis. In: Bergmeyer HU, ed. Methods of Enzymatic Analysis. New York, NY: Academic Press Inc; 1974:1831-1835.
Lopez-Virella HF, Stone P, Ellis J, Colwell LA. Cholesterol determination in the HDL separated by precipitation with polianions. J Lipid Res. 1970;11:583-595.
Becker W, Schafer W, Ziebenbein W. Standardization of immunochemical quantification of human apolipoprotein A and B. La Ricerca Clin Lab. 1980;10:203-212.
Albers JJ, Adolphson JL, Hazzard WR. Radioimmunoassay of human plasma Lp(a). J Lipid Res. 1977;18:331-338.
Friedewald WT, Levy RJ, Frederickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1980;18:499-502.
Averna MR, Barbagallo CM, Galione A, Carroccio A, Labisi M, Marino G, Montalto G, Notarbartolo A. Serum apolipoprotein profile of hypertriglyceridemic patients with chronic renal failure on hemodialysis: a comparison with type IV hyperlipoproteinemic patients. Metabolism. 1989;38:1-2.
Siegel S. Nonparametric Statistics for the Behavioral Sciences. New York, NY: McGraw-Hill Publishing Co; 1986:184-193.