Atherosclerosis 153 (2000) 181–189

Statins and cardiovascular diseases: the multiple effects oflipid-lowering therapy by statins

U. Rauch a,b, J.I. Osende a,b, J.H. Chesebro b, V. Fuster b, D.A. Vorchheimer b,K. Harris b, P. Harris b, D.A. Sandler b, J.T. Fallon b,c, S. Jayaraman a,

J.J. Badimon a,b,*a The Cardio6ascular Biology Research Laboratory, The Zena and Michael A. Wiener Cardio6ascular Institute,

Mount Sinai School Of Medicine, One Gusta6e L. Le6y Place, Box 1030, New York, NY-10029, USAb The Zena and Michael A. Wiener Cardio6ascular Institute, New York NY-10029, USAc Department of Pathology, Mount Sinai School of Medicine, New York NY-10029, USA

Received 12 October 1999; received in revised form 20 December 1999; accepted 19 January 2000

Abstract

Cholesterol lowering involving different therapies improves the clinical outcome of patients. To define the underlyingpathomechanism, we studied whether treatment with statins was associated with changes in blood thrombogenicity, endothelialdysfunction and soluble adhesion molecule levels. Fifty hypercholesterolemic patients were treated with pravastatin (40 mg/day,n=24) or simvastatin (20 mg/day, n=26). Lipid profile and blood thrombogenicity were assessed in all patients before and after3 months of cholesterol reducing therapy. Blood thrombogenicity was assessed as thrombus formation, perfusing non-anticoagu-lated blood directly from the patients’ vein through the Badimon perfusion chamber (shear rate 1690/s). Endothelial-dependentvasomotor response was tested by laser-Doppler flowmeter. Soluble adhesion molecule level were measured by ELISA. Total andLDL cholesterol were reduced in the two treatment groups by statin therapy. Statin therapy was associated with a significantreduction in blood thrombogenicity and endothelium-dependent vasoresponse. No differences were observed between simvastatinor pravastatin treatment. Lipid lowering by statins had no effect on plasma levels of fibrinogen, sL-selectin, sP-selectin andsICAM-1 antigen. Cholesterol lowering by both statins reduced the increased blood reactivity and endothelial dysfunction presentunder hypercholesterolemia. The multiple effects of lipid lowering therapy by statins may explain the benefits observed in recentepidemiological trials. © 2000 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Statin; Hyperlipidemia; Thrombosis; Endothelium; Atherosclerosis

www.elsevier.com/locate/atherosclerosis

1. Introduction

Hypercholesterolemia — the major risk factors forthe development of atherosclerotic disease [1–4] is asso-ciated with increased deposition of lipids and mono-cytes/macrophages within the arterial wall, endothelialdysfunction and vasoconstriction, enhanced platelet re-activity, and hypercoagulability [5,6]. Reduction ofserum cholesterol levels by different interventions in-cluding diet, exercise, lifestyle changes or pharmacolog-ical approaches is associated with a significant

reduction in cardiovascular mortality and morbidity[3–6]. Recently, the effectiveness of a new class ofpowerful hypolipidemic agents, the statins, has beentested in several large clinical trials. These trials showedsignificant improvement in clinical outcomes of patientswith and without coronary artery disease [7–12]. Theseobservations were applicable to primary and secondaryprevention of coronary events in both, hyper- andnormocholesterolemic populations [4,6–12]. Statintreatment was associated with a consistent reduction incoronary events of :30–40% [13,14].

Angiographic trials assessing the effect of lipid lower-ing treatment on coronary atherosclerosis showed onlya small reduction in coronary stenosis [14,15]. Theclinical benefits observed in these trials significantly

* Corresponding author. Tel.: +1-212-2418484; fax: +1-212-4266962.

E-mail address: jbadimo@smtplink.mssm.edu (J.J. Badimon).

0021-9150/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved.PII: S 0 0 2 1 -9150 (00 )00397 -X

U. Rauch et al. / Atherosclerosis 153 (2000) 181–189182

exceed the expectations based on alterations in vessellumen diameter. This suggested that lipid-reducing ther-apy might have additional effects that would explainthe discrepancy between clinical benefits and lack ofsignificant lesion regression [3–6]. Experimental andclinical studies have indicated a relationship betweenhyperlipidemia and increased blood thrombogenicity[15–21]. It was suggested that correction of hyper-cholesterolemia by pravastatin could normalize bloodthrombogenicity [20]. However, conflicting findings onthe effect of different statins on thrombosis have beenreported by other authors [22,23]. In addition, hyper-lipidemia has been associated with an impaired en-dothelium-dependent vasoresponse [24] and cholesterollowering by the use of specific statins has been reportedto improve endothelial dysfunction [3,24]. The goal ofour study was to delineate whether lipid lowering bystatins, irrespective of the specific statin used, wasassociated with changes in blood thrombogenicity andendothelial dysfunction in the same hyperlipidemic pa-tient population. Furthermore, we studied whetherchanges in thrombogenicity and endothelial functionwere associated with alterations in circulating solubleadhesion molecule levels.

The working hypothesis of this study was that (1) thereduction of plasma lipid levels reduces blood thrombo-genicity and endothelial dysfunction and (2) that thedecreased thrombogenicity of blood and endothelium isassociated with a reduction in circulating soluble adhe-sion molecule levels. Given the number of pathogenicprocesses modulated by plasma lipid levels, we furtherhypothesized that the significant benefits observed inclinical trials with statins might rather be a result of thesignificant reduction of plasma cholesterol level than aspecific effect associated with the administration of onespecific statin. To prove this hypothesis, hyperlipidemicpatients were randomized into two groups receiving oneof two most frequently prescribed statins for lipidreduction.

This study demonstrates multiple effects of choles-terol reducing therapy by the statin class on bloodthrombogenicity and endothelial dysfunction in thesame patient group under hyperlipidemic conditions.The combined reduction of blood thrombogenicity andendothelial dysfunction observed after lipid lowering bystatins may be responsible for the clinical improvementsreported in several large epidemiological trials.

2. Material and methods

2.1. Patient population

The study population included hypercholesterolemicpatients (n=50). Hypercholesterolemia was defined asLDL-cholesterol level \120 mg/dl. Twenty-six of the

patients had documented coronary artery disease and/or peripheral artery disease. The presence of coronaryartery disease was documented by coronary angiogra-phy (stenosis \50%), history of previous myocardialinfarction or typical angina pectoris with typicalchanges in treadmill ECG. Exclusion criteria includedmalignant diseases, hematological disorders, severe hy-pertension, heparin or warfarin therapy and exposureto any lipid lowering therapy within the 8 weeks priorto randomization. The selected patients were instructedto follow a low cholesterol, low fat diet and thenrandomized into two treatment groups. Group I wasassigned to receive pravastatin, 40 mg per day andGroup II simvastatin, 20 mg per day for a period of 3months. All patients were advised to take the medica-tion at evening time. Other drugs the patients werecurrently receiving remained unchanged. The study wasapproved by the Institutional Review Board, and allpatients signed informed consent.

The following routine laboratory parameters weredetermined at baseline and at the follow-up visits: lipidprofile, white blood cell count, platelet count, partialthromboplastin time (aPTT), and fibrinogen level. Inaddition, citrated plasma samples, obtained pre andpost treatment with simvastatin or pravastatin, werealiquoted and frozen for the analysis of sLselectin,sP-selectin and sICAM-1 (R&D systems, Minneapolis,MN). The coefficient of variation for all these parame-ters was B6% in all instances. All assays were per-formed according to the manufacturer’s instructions.

2.2. Assessment of blood thrombogenicity

Blood thrombogenicity was evaluated at baseline and3-months after lipid-lowering therapy in both groups ofpatients. This experimental design allows for each pa-tient to serve as his/her own control. Effect of lipidlowering on blood thrombogenicity was assessed aschanges in the thrombus area formed in an ex-vivoperfusion chamber [25]. The ex-vivo perfusion chambermimics the in vivo local rheologic conditions that de-velop in a mildly stenosed coronary artery. In thisperfusion system, thrombus formation is triggered bythe exposition of components of the subendothelialvessel wall to the flowing blood. Blood from hyperc-holesterolemic patients, prior to and after statin treat-ment, was allowed to circulate through the perfusionsystem. Blood reactivity was assessed by morphometricanalysis of thrombus formed on the thrombogenic sub-strate. The description, validity and reliability of theBadimon perfusion system to study thrombus forma-tion on defined thrombogenic substrates under stablerheological conditions have been previously reported[25,26].

Perfusion chamber experiments and blood samplingwere performed in the mornings, in fasting patients. In

U. Rauch et al. / Atherosclerosis 153 (2000) 181–189 183

brief, a 19-gauge needle was carefully inserted into anantecubital vein. After discarding the first 5 ml ofblood, the needle was connected to three Badimonperfusion chambers placed in series to minimize bloodloss. Native, non-anticoagulated, blood was directlyperfused from the antecubital vein through the perfu-sion chambers at a constant flow maintained by using aperistaltic pump (Masterflex, Cole- Parmer Instru-ments) distal to the chambers. The perfusion chamberconsists of a cylindrical flow channel (1 mm diameterand 2.5 cm length) that allows the blood to flow overthe thrombogenic substrate. All the perfusions wereperformed for a period of 5 minutes at a flow rate of 10ml/min (calculated shear rate of 1690/s; Reynolds num-ber 60; average blood velocity 21.2 cm/s). The selecteddynamic conditions modeled the local rheology devel-oping on mildly stenosed coronary arteries. Our previ-ous work demonstrated that these rheologic conditionsresult in consistent levels of platelet deposition.

To mimic the in vivo situation of severe arterialinjury associated with coronary interventions, porcineaortic tunica media was used as the substrate to triggerthrombus formation. Segments of porcine aorta werecut into segments (2.8×0.8cm) and surgically preparedto expose the deeper components of the arterial wall aspreviously described [25,26].

The histological assessment of thrombus formationwas performed by histomorphometry as previously de-scribed [25]. In brief, the perfused media segments werefixed in 4% paraformaldehyde. Six 2 mm cross sectionswere removed from the perfused media segment: two inthe proximal, middle, and distal thirds of the exposedvessel segments. These tissue sections were then embed-ded in paraffin and processed for histology. Sections (5mm) were cut and stained with a combined Masson’strichrome-elastin to visualize the total thrombus formedon the exposed substrates.

Morphometric analysis of thrombus was conductedat 400-fold magnification. Images were digitized with aSony DKC-5000 camera using Adobe Photoshop 4.0software on a PowerMacIntosh 8500 computer. Throm-bus area on each section was measured by computer-ized planimetry using NIH Image 1.6 software. Theresults are given as the average of the analyzed sectionsper perfused media segments, and the results from thethree perfusion chamber were averaged to obtain theoverall thrombus formation for each patient. The mea-surements were independently done by two of the inves-tigators, who were blinded to the therapy assignment.

2.3. Endothelium-dependent 6asomotor response

Skin blood flow is controlled both by reflexes andlocal factors. Cutaneous vascular response to localwarming is an endothelium-mediated mechanism [27].Therefore, measurements of microcirculatory blood

flow to heat have been used as marker of endotheliumfunction. Vascular responses to local heat wererecorded in a subgroup of 11 hyperlipidemic patients atbaseline and after 3-months on lipid lowering therapy.A laser-Doppler flowmeter with a combined laserDoppler and thermostatic probe (Perimed, model PF4001 Master, Perimed, Jarfalla, Sweden) was used toassess the microcirculatory blood flow [28]. Measure-ments were made on the dorsum of the hand with theprobe fixed to the skin by a double-sided adhesive ring.This technique shows an intraindividual variability ofB2%. Results are expressed as arbitrary perfusionunits (PU) and are an index of the blood flow [29].

All measurements were made in a temperature con-trolled room (19–22°C) with the patients lying insupine position. Skin blood flow was continuouslyrecorded during a baseline period with the probe set to33°C (neutral skin temperature) and during local skinheating to 44°C. After 3 min of monitoring baselineperfusion, blood flow was measured for 10 min over theheating period. The typical response is a maximumvasodilatation response after 3–5 min of skin heatingover 42°C. Comparisons were made between the meanbaseline flow and the area under the curve of the timecourse recorded during 3 min at 33°C (baseline perfu-sion) and 5 min of heating over 42°C (endothelium-de-pendent vasomotor response to heat) after 5 minheating period. Results (mean9SEM) are expressed asperfusion units (PU) and perfusion units x min.

2.4. Data analysis

Differences between the treatment groups were as-sessed using ANOVA. Comparisons pre and post ther-apy within the same group were analyzed usingStudent’s paired t-test. If not otherwise indicated, dataare given as mean9SD for independent measurements.The two tailed significance threshold was set at PB0.05.

3. Results

3.1. Study population

No significant differences between both groups werepresent in the anthrophometric, routine laboratory andother clinical risk factors at baseline (Table 1). Themedication profile pre-treatment was also comparablein the two groups of patients. The serum lipid profilecomprising total cholesterol, LDL cholesterol, HDLcholesterol, and triglycerides was similar in both groupsat baseline. Total cholesterol and LDL cholesterol levelwere significantly reduced in both groups of patientsafter 3-months treatment with either simvastatin orpravastatin. HDL cholesterol and triglyceride levelswere not affected by statins therapy (Table 2).

U. Rauch et al. / Atherosclerosis 153 (2000) 181–189184

Table 1Patient population

Group I Group II

Age (years) 58926292Hypertension 13 14Diabetes mellitus 8 12

1413Smoking12Family history CHD 11

0Cerebral ischemic disease 044Peripheral artery disease

5CAGB 3Angioplasty 7 7

3.2. Blood thrombogenicity

Baseline blood thrombogenicity, evaluated by thethrombus area formed on the perfused substrates, wassimilar in both groups. A representative photomi-crograph of thrombus formation at baseline and 3-months after statin treatment is shown in Fig. 1. Thesimvastatin group showed a significant reduction inthrombus formation (11 0679980 vs 82769728 mm2/mm baseline and post-treatment, respectively; PB0.05). Similar reduction in thrombus formation tookplace in the pravastatin group (12 64891023 vs94119992 mm2/mm at baseline and 3-months posttreatment, respectively; PB0.05). No significant differ-ence in thrombus formation was found between the two

Table 2Plasma lipid profile and plasma levels of fibrinogen, sP-selectin, sL-selectin, CAM-1 of the two patient groups at baseline and after 3-months withstatinsa

Sinvastain group Pravastatin group

Baseline Post treatment Baseline Post treatment

20298*251910Total cholesterol (mg/dl) 17199* 24698151916141911Triglycerides (mg/dl) 123913 165919

4191.9 4192.0 4492.2HDL cholesterol (mg/dl) 4091.917197 12897*11097*LDL cholesterol (mg/dl) 184910

4.890.36.390.44.490.3*Total/HDL cholestrol ratio 6.690.4LDL/HDL cholesterol ratio 4.490.3 3.090.2*2.890.2*4.890.3

263910 243916Fibrinogen (mg/dl) 239919255913105917 115915 130930 129922sP-selectin (ng/ml)

738931sL selectin (ng/ml) 778932834942 7539305493 4893 4795sICAM-1 (ng/ml) 5395

a Data is expressed as mean9SD.* PB0.05 versus baseline values.

Fig. 1. Representative histological sections corresponding to baseline (top) and 3-months post treatment (bottom).

U. Rauch et al. / Atherosclerosis 153 (2000) 181–189 185

Fig. 2. Values of thrombus formation corresponding to the Simvastatin group. A. All patients. B. Subgroup of patients with cardiovascular diseaseand, C. Subgroup of patients without cardiovascular disease. Data expressed as mm2/mm

statin-treated groups. The antithrombotic effect of lipidreduction was still present when the groups were subdi-vided into patients with and without cardiovasculardisease (CVD) for both groups (Figs. 2 and 3). Thegroup of patients with CVD in the Simvastatin group(n=14) showed values of 11 37391087 mm2/mm atbaseline and 836891061 mm2/mm at 3-months, respec-tively, while the group of patients without CVD (n=12) had 10 71091755 mm2/mm at baseline and817091027 mm2/mm at 3-months, respectively. Valuesof thrombus formation within the pravastatin groupwere 13 60791444 and 957391447 mm2/mm at base-line and 3-months, respectively for the CVD group; and11 68991457 mm2/mm at baseline and 925191419mm2/mm at 3-months in the non-CVD group. The lipidlowering effect of the two statins used in our study wasassociated with an :20–25% reduction in thrombusformation as compared with the baseline values.

A correlation between thrombus formation and totalcholesterol/HDL cholesterol ratio was calculated tofurther assess the relation between circulating plasma

cholesterol and blood thrombogenicity. A significantrelationship between amount of thrombus and totalcholesterol/HDL cholesterol ratio was found consider-ing the measurements pre and post lipid lowering treat-ment (correlation coefficient r=0.410, P50.001, Fig.4).

3.3. Endothelium-dependent 6asomotor response (seeTable 3)

Hyperlipidemia has been associated with an impairedendothelium-dependent vasoresponse. Using a laserdoppler flow meter evaluation of blood flow as a indi-cator of endothelium-dependent vasoresponse, we ob-served similar findings.[27] Hyperemic reaction inducedby local warming is an endothelial-dependent mecha-nism and thus, measuring blood flow changes would bean indicator of endothelial function. The effect of lipidlowering was found to be significant when measuringheat-mediated blood flow (49.795.7 PU vs 81.5912.3PU; at baseline and 3-months post-treatment, respec-

U. Rauch et al. / Atherosclerosis 153 (2000) 181–189186

tively; P50.01). These observations suggest an im-provement of the endothelium-dependent vasoresponseby the hypolipidemic treatment. Similar observationswere obtained when analyzing the areas under the curveat the different time/points and conditions.

3.4. Soluble adhesion molecule le6els

Treatment with statins did not induce any significantchanges in the plasma levels of sP-selectin, sL-selectinor sICAM-1 between baseline and post-treatment mea-surements (Table 2).

In addition, the treatment with statins was not asso-ciated with any significant effect on white blood cellcount, platelet count, activated partial thromboplastintime (aPTT) or plasma fibrinogen.

4. Discussion

This study demonstrates that the reduction of plasmacholesterol levels by statins reduces the high bloodthrombogenicity observed in the same group of patientsunder hyperlipidemic conditions. We conclude that thereduction of blood thrombogenicity and endothelialactivity may be mechanisms by which cholesterol lower-ing through statins accounts for the beneficial effects onCHD observed in several large epidemiological trials.

Epidemiological evidence has shown the relationshipbetween plasma total cholesterol and mortality ratefrom CHD [2–12]. A variety of mechanisms have beenproposed to understand the effect of hypercholes-terolemia on atherosclerosis and thrombosis [2–6].Acute thrombus formation depends not only on the

Fig. 3. Values of thrombus formation corresponding to the Pravastatin group. A. All patients. B. Subgroup of patients with cardiovascular diseaseand, C. Subgroup of patients without cardiovascular disease. Data expressed as mm2/mm.

U. Rauch et al. / Atherosclerosis 153 (2000) 181–189 187

Fig. 4. Correlation between total cholesterol/ HDL cholesterol ratioand amount of thrombus formation. Lipid levels and thrombusformation were measured pre and post lipid reducing therapy (n=100 measurements for 50 patients, respectively). R, correlation coeffi-cient

as lipid lowering agent. The effect of placebo treatmenton cholesterol level and thrombus formation has alsobeen tested in the same ex vivo model system and shearstress conditions that we used here [35]. Lipid plasmalevels and thrombus formation were not altered when aplacebo instead of pravastatin was given [35]. In ourstudy, both statins, simvastatin and pravastatin, in-duced a significant hypolipidemic effect that was associ-ated with similar reduction of blood thrombogenicity inthe respective group. Therefore, we conclude that thereduction of blood thrombogenicity could be one of themechanisms by which cholesterol lowering by statinsreduces the incidence of acute thrombotic events.

It has been suggested that endothelial dysfunctionplays a major role, not only in the pathogenesis ofatherosclerosis but also in its ischaemic manifestations.Therefore, normalization of the dysfunctional endothe-lium may restore its antiatherogenic properties. Thereare evidences that endothelial dysfunction is often im-proved by cholesterol lowering [24,36] and it seems tobe mediated by an increased bioavailability of nitricoxide [37]. To define whether the described lipid lower-ing effect of statins was the only mechanism responsiblefor the observed antithrombotic effect, we studied theeffects of lipid reduction by statins on endothelial de-pendent vasoresponse in the same patient population.As described by others, we observed a significantlyimpaired endothelium-dependent vasoresponse in ourhyperlipidemic population after lipid lowering therapycompared to baseline. Furthermore, a marked improve-ment of the endothelial-dependent vasoresponse wasassociated to the lipid lowering effect induced by thestatins.

An additional objective of the study was to assesswhether the observed normalization of blood thrombo-genicity and endothelial function by statin therapywould be associated with changes in plasma levels ofadhesive proteins involved in the interactions betweenblood cells and endothelium. We determined plasmalevels of sP-selectin, sL-selectin and sICAM-1 at base-line and 3-months after statin administration. Treat-ment with statins did not modify levels of adhesiveproteins in the circulating blood. We speculate that thestudy population tested here may have been to small toidentify differences. Since only the circulating solubleantigen levels were determined in this study, it cannotbe excluded that surface-associated selectin and ICAM-1 expression on endothelial cells and/or circulatingblood cells may have been affected by statin therapy.

As indicated by Fig. 5, hyperlipidemia is associatedwith an increased platelet reactivity, endothelial dys-function and plaque progression, altogether leading toan increased risk of acute thrombotic events. Our dataon blood thrombogenicity and endothelial functionwere simultaneously obtained from the same individu-als before and after lipid lowering therapy with twodifferent statins. Our data and the observations of

Table 3Effect of lipid lowering by statins on endothelial activity at baselineand after 3-months with statinsa

Pre-treatment 3-month statins

Hyperlipidemic patients6.590.9Basal values 8.190.9

81.5912.3*,**Heat induced 49.795.7*

* PB0.05 versus basal values.** PB0.05 versus pre-treatment.a Results expressed as PU.

local alterations of the arterial wall, but it is alsomodulated by the thrombogenicity of the circulatingblood [30–33]. Our group and others have recentlyshown that the circulating platelets are activated andhyperactive in hypercholesterolemia [16–20]. In addi-tion, it has been reported that reducing cholesterol levelby pravastatin decreases thrombus formation on deeplyinjured arterial wall [20]. Whether this antithromboticeffect was a result of the cholesterol reduction or aspecific effect of the specific statin used was questioned[21,22]. On the other hand, it has recently been demon-strated that lipid reducing therapy by simvastatinmarkedly depressed thrombin generation and that thiseffect on thrombin generation was not dependent onthe presence of aspirin in patients with hypercholes-terolemia [34].

This study revealed that the blood thrombogenicityat hyperlipidemic baseline conditions is significantlyreduced after reducing cholesterol in the same patientpopulation. The reduction in thrombus formation wasachieved by the use of either simvastatin or pravastatin

U. Rauch et al. / Atherosclerosis 153 (2000) 181–189188

Fig. 5. Multiple effects of hyperlipidemia on blood and arterial wall. Implications for acute coronary events.

others suggest that multiple biological systems may beaffected by lipid lowering therapy with statins. Thisstudy strongly supports the hypothesis that thepleiotropic effects of lipid lowering therapy by statinsmay explain the significant clinical benefits of statintherapy observed in several large epidemiological trials[4–6,38–41]. Thus, the observed improvement in clini-cal outcome after statin therapy seems rather to be aresult of interfering with multiple pathogenic parame-ters present under hypercholesterolemic conditions(Fig. 5) than the intrinsic effect of one specific statin.The major limitation of our study is that we can notrule out whether all statins as a drug class have a directeffect on thrombogenicity and endothelial function orwhether the observed changes are mediated by the lipidreduction only.

In summary, we conclude that the reduction of bloodthrombogenicity and endothelial activity may be mech-anisms by which cholesterol-reducing therapy withstatins improves the clinical outcome of patients withhypercholesterolemia.

Acknowledgements

This work was supported in part by grants from theSpanish Society of Cardiology (to Dr. J Osende) andthe followings grants from the National Institutes ofHealth (NIH P50 HL54469 to Dr J.J. Badimon and DrJ.T. Fallon) and (NIH 5 MO1 RR 00071) to the Mount

Sinai General Clinical Research Center. The authorswould like to express their appreciation to the excellentwork, care and dedication of the nurse coordinatorsStella Palencia and Ida Guzman.

References

[1] National Heart, Lung, and Blood Institute Fact Book, FiscalYear 1997. Bethesda, Md: National Heart, Lung, and BloodInstitute, National Institute of Health; 1998.

[2] Tracy RP, Tracy PB. New views on the relationship of plasmalipids to cardiovascular disease. Circulation 1997;95:1347–8.

[3] Expert Panel on Detection, Evaluation, and Treatment of HighBlood Cholesterol in Adults. National cholesterol educationprogram: second report of the expert panel on detection, evalua-tion, and treatment of high blood cholesterol in adults (adulttreatment panel II). Circulation 1994;89:1329–445.

[4] Cleeman JI, Lenfant C. The national cholesterol education pro-gram: progress and prospects. J Am Med Acad 1998;280:2099–104.

[5] Archbold RA, Timmis AD. Cholesterol lowering and coronaryartery disease: mechanisms of risk reduction. Heart1998;80:543–7.

[6] Gould AL, Rossouw JE, Santanello NC, Heyse JF, Furberg CD.Cholesterol reduction yields clinical benefit. Impact of statintrials. Circulation 1998;97:946–52.

[7] Randomized trial of cholesterol lowering in 4444 patients withcoronary heart disease: the Scandinavian Simvastatin SurvivalStudy (4S). Lancet 1994;344:1383–9.

[8] Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronaryheart disease with pravastatin in men with hypercholesterolemia.N Engl J Med 1995;333:1301–13077.

U. Rauch et al. / Atherosclerosis 153 (2000) 181–189 189

[9] Sacks FM, Pfeffer MA, Moye LA, et al. The effect of pravas-tatin on coronary events after myocardial infarction in patientswith average cholesterol levels. N Engl J Med 1996;335:1001–9.

[10] Downs JR, Clearfield M, Weis S, et al. AM for the AFCAPS/TexCAPS Research Group. Primary prevention of acute coro-nary events with lovastatin in men and women with averagecholesterol levels: results of AFCAPS/TexCAPS. J Am Med Am1998;279:1615–22.

[11] The Long-Term Intervention with Pravastatin in Ischaemic Dis-ease (LIPID) Study Group. Prevention of cardiovascular eventsand death with pravastatin in patients with coronary heartdisease and a broad range of initial cholesterol levels. N Engl JMed 1998;1995(333):1301–7.

[12] Jukema JW, Bruschke AV, vanBowen AJ, et al. Effect of lipidlowering by pravastatin on progression and regression of coro-nary disease in symptomatic men with normal to moderatelyelevated cholesterol levels. The Regression Growth EvaluationStatin Study. (REGRESS). Circulation 1995;91:2528–40.

[13] Herbert PR, Gaziano JM, Chan KS, Hennekens CH. Choles-terol lowering with statin drugs, risk of stroke, and total mortal-ity: an overview of randomized trials. J Am Med Acad1997;278:313–21.

[14] Gotto AM. Cholesterol management in theory and practice.Circulation 1997;96:4424–30.

[15] Fuster V, Badimon JJ. Regression or stabilization of atheroscle-rosis means regression or stabilization of what we don’t see inthe arteriogram. Eur Heart J 1995;16:6–12.

[16] Carvalho AC, Colman RW, Lees RS. Platelet function in hyper-lipoproteinemia. New Engl J Med 1974;290:434–8.

[17] Badimon JJ, Badimon L, Turitto VT, Fuster V. Platelet deposi-tion at high shear rates is enhanced by high plasma cholesterollevels: in vivo study in the rabbit model. Arterioscler Thromb1991;11:395–402.

[18] Cadroy Y, Lemozy S, Diquelou A, et al., Human type IIahyperlipoproteinemia enhances platelet-collagen adhesion inflowing nonanticoagulated blood, Arterioscler Thromb1993;13:1650–3.

[19] Davi G, Averna M, Catalano I, et al. Increased thromboxanebiosynthesis in type IIa hypercholesterolemia. Circulation1992;85:1792–8.

[20] Lacoste L, Lam JYT, Hung J, Letchacovski G, Solymoss CB,Waters D. Hyperlipidemia and coronary disease: correction ofthe increased thrombogenic potential with cholesterol reduction.Circulation 1995;92:3172–7.

[21] Broijersen A, Hamsten A, Eriksson M, Angelin B, Hjemdahl P.Platelet activity in vivo in hyperlipoproteinemia — importanceof combined hyperlipidemia. Thromb Haemost 1998;79:268–75.

[22] Aviram M, Hussein O, Rosenblat M, Schlezinger S, Hayek T,Keidar S. Interactions of platelets, macrophages, and lipo-proteins in hypercholesterolemia: antiatherogenic effects ofHMG-CoA reductase inhibitor therapy. Cardiovasc Pharmacol1998;31:39–45.

[23] Broijersen A, Eriksson M, Leijd B, Angelin B, Hjemdahl P. Noinfluence of simvastatin treatment on platelet function in vivo inpatients with hypercholesterolemia. Arterioscler Thromb VascBiol 1997;17:273–8.

[24] Anderson TJ, Meredith IT, Yeung AC, Frei B, Selwyn AP, GanzP. The effect of cholesterol-lowering and antioxidant therapy on

endothelium-dependent coronary vasomotion. N Engl J Med1995;332:488–93.

[25] Dangas G, Badimon JJ, Coller BS, et al. Administration ofabciximab during percutaneous coronary intervention reducesboth ex vivo platelet thrombus formation and fibrin deposition.Implications for a potential anticoagulant effect of abciximab.Arterioscler Thromb Vasc Biol 1998;18:1342–9.

[26] Badimon L, Turitto V, Rosemark J, Badimon JJ, Fuster V.Characterization of a tubular flow chamber for studying plateletinteraction with biological and prosthetic materials: depositionof indium-labeled platelets on collagen, subendothelium, andgoretex. J Lab Clin Med 1987;110:706–18.

[27] Pergola PE, Kellogg DL, Jr, Johnson JM, Kosiba WA, SolomonDE. Role of sympathetic nerves in the vascular effects of localtemperature in human forearm skin. Am J Physiol1993;265:H785–92.

[28] Nilsson GE, Tenland T, Oberg P. Evaluation of a laser Dopplerflow meter for measurement of tissue blood flow. EEE TransBiomed Engl 1980;27:597–604.

[29] Stern MD. In vivo evaluation of microcirculation by coherentlight scattering. Nature 1975;254:56–8.

[30] Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogen-esis of coronary artery disease, parts 1 and 2. N Engl J Med1992;326:310–8.

[31] Badimon JJ, Fuster V, Chesebro J, Badimon L. Coronaryatherosclerosis: a multifactorial disease. Circulation 1993;87:3–16.

[32] Falk E, Shah PK, Fuster V. Coronary Plaque disruption. Circu-lation 1995;92:657–71.

[33] Burke AP, Farb A, Malcom GT, Liang YH, Smialek J, VirmaniR. Coronary risk factors and plaque morphology in men withcoronary disease who died suddenly. N Engl J Med1997;336:1276–82.

[34] Szczeklik A, Musial J, Undas A, et al. Jankowski M. Inhibitionof thrombin generation by simvastatin and lack of additiveeffects of aspirin in patients with marked hypercholesterolemia. JAm Coll Cardiol 1999;33:1286–93.

[35] Dangas G, Badimon JJ, Smith DA, Unger AH, Levine D, ShaoJH, Meraj P, Fier C, Fallon JT, Ambrose JA. Pravastatintherapy in hyperlipidemia: effects on thrombus formation andthe systemic hemostatic profile. J. Am. Coll. Cardiol.1999;33:1294–304.

[36] O’Driscoll G, Green D, Taylor RR. Simvastatin, an HMG-coen-zyme A reductase inhibitor improves endothelial function within1 month. Circulation 1997;95:1126–31.

[37] John S, Schlaich M, Langelfeld M, et al. Increased bioavailabil-ity of nitric oxide after lipid-lowering therapy in hypercholes-terolemic patients: a randomized, placebo-controlled,double-blind study. Circulation 1998;98:211–6.

[38] Vaughan CJ, Murphy MB, Buckley BM. Statins do more thanjust lower cholesterol. Lancet 1996;348:1079–82.

[39] Rosenson RS, Tangney CC. Antiatherothrombotic properties ofstatins. Implications for cardiovascular event reduction. J AmMed Acad 1998;279:1643–50.

[40] Libby P, Aikawa M. New Insights into plaque stabilisation bylipid lowering. Drugs 1998;56:9–13.

[41] Bellosta S, Via D, Canavesi M, et al. HMG-CoA reductaseinhibitors reduce MMP-9 secretion by macrophages. ArteriosclThromb Vasc Biol 1998;18:1671–8.

.