Beneficial Effects of Pravastatin(6Colestyramine/Niacin) Initiated

Immediately After a Coronary Event(The Randomized Lipid-Coronary Artery

Disease [L-CAD] Study)Hans-Richard Arntz, MD, Rahul Agrawal, MD, Werner Wunderlich, MSc,

Luise Schnitzer, MD, Richard Stern, MD, Frank Fischer, MSc, andHeinz-Peter Schultheiss, MD

Secondary prevention of coronary heart disease by an-tilipidemic therapy beginning at >3 months after anacute coronary syndrome is well documented. The im-pact, however, of immediate initiation of antilipidemictherapy on coronary stenoses and clinical outcome inpatients with acute coronary syndrome is unknown. Inour study, patients were randomized, on average, 6days after an acute myocardial infarction and/or per-cutaneous transluminal coronary angioplasty secondaryto unstable angina, to pravastatin (combined, when nec-essary, with cholestyramine and/or nicotinic acid) toachieve low-density lipoprotein cholesterol levels of<130 mg/dl (group A, n 5 70). In controls (group B,n 5 56), antilipidemic therapy was determined by fam-ily physicians. Quantitative coronary angiography wasperformed at inclusion, and at 6- and 24-month follow-up. The combined clinical end points were total mortal-

ity, cardiovascular death, nonfatal myocardial infarc-tion, need for coronary intervention, stroke, and newonset of peripheral vascular disease. Minimal lumendiameter in group A increased by 0.05 6 0.20 mm after6 months and 0.13 6 0.29 mm after 24 months,whereas it decreased by 0.08 6 0.20 mm and 0.18 60.27 mm, respectively, in group B (p 5 0.004 at 6months and p <0.001 at 24 months). After 2 years, 29patients of 56 patients in group B, but only 16 of 70patients in group A, experienced a clinical end point(p 5 0.005; odds ratio 0.28, confidence intervals 0.13 to0.6). We conclude that pravastatin-based therapy initi-ated immediately after an acute coronary syndrome iswell tolerated and safe, lessens coronary atherosclero-sis, and has a pronounced clinical benefit. Q2000 byExcerpta Medica, Inc.

(Am J Cardiol 2000;86:1293–1298)

Secondary prevention studies have shown thattreatment with 3-hydroxymethyl-3-glutaryl coen-

zyme A reductase inhibitors may prevent recurrentcoronary events by reducing elevated cholesterol lev-els.1–3 In addition, angiographic studies have indicatedthat this therapy may reduce progression and induceregression of coronary lesions.4–9 These studies, how-ever, have primarily included patients several monthsafter a qualifying event. Moreover, there is a lack ofserial studies where the angiographic course is docu-mented early (6 months) after initiation of therapyfollowed by a longer period of follow-up. The objec-tives of this prospective open-label, randomized, sin-gle-center study were to compare the effects of cho-lesterol-lowering therapy initiated immediately afteran event with the effects of usual outpatient treatmenton the course of coronary atherosclerosis and clinicaloutcome.

METHODSPatients and therapy: Patients with total cholesterol

of .200 to ,400 mg/dl and low-density lipoprotein(LDL) cholesterol of .130 to ,300 mg/dl (afterexclusion of secondary forms of hyperlipidemia) withan acute myocardial infarction (defined by new Qwaves and increase of enzymes of.3-fold the normalvalue) and/or who underwent emergency percutane-ous transluminal coronary balloon angioplasty due tosevere or unstable angina pectoris (defined by clinicalsymptoms and electrocardiographic [ECG] alterationsduring rest and/or exercise) were included in thisstudy. Randomization was stratified for gender andreason for inclusion (infarction and/or coronary angio-plasty). Patients with an acute infarction were in-cluded only if their lipids could be determined within8 hours after onset of symptoms; those in the balloonangioplasty group with a history of infarction wereincluded only when the infarction was$3 monthsearlier. Patients.75 years old, diabetics (fastingblood glucose.125 mg/dl), patients with postcoro-nary artery bypass graft, known malignant disease,serious kidney or liver dysfunction (creatimine.1.5mg/dl, alanine aminotransferase and aspartate amino-transferase.2 times normal value), or women ofchild-bearing age not using a reliable form of contra-

From the Medical Clinic II, Cardiology and Pulmology, Klinikum Ben-jamin Franklin, Free University of Berlin, Berlin, Germany. This studywas supported in part by a grant from Bristol-Myers Squibb Company,Munich, Germany. Manuscript received February 14, 2000; revisedmanuscript received and accepted June 20, 2000.

Address for reprints: Hans-Richard Arntz, MD, Medical Clinic II,Cardiology and Pulmology, Klinikum Benjamin Franklin, Hindenburg-damm 30, D-12200 Berlin, Germany.

1293©2000 by Excerpta Medica, Inc. All rights reserved. 0002-9149/00/$–see front matterThe American Journal of Cardiology Vol. 86 December 15, 2000 PII S0002-9149(00)01230-3

ception were excluded. The protocol was approved bythe Ethics Committee of our institution. All patientssigned an informed consent form.

Patients were blindly assigned to intensive antili-pidemic therapy (group A) or standard care (group B).All patients received dietary counseling per the stan-dards of the European Atherosclerosis Society.10 Anaverage 66 5 days after the qualifying event, patientsin group A with a baseline LDL cholesterol level of,160 mg/dl initially received pravastatin 20 mg/day;those with an LDL cholesterol of$160 mg/dl re-ceived 40 mg/day. To achieve LDL cholesterol levelsof #130 mg/dl, the pravastatin dosage was increasedto 40 mg/day, and nicotinic acid (1.5 to 6 g/day)and/or cholestyramine (4 to 32 g/day) were added ifnecessary. In group B, antilipidemic therapy was de-termined by the family physician. Clinic visits andlaboratory studies were scheduled for both groups at4-week intervals for the first 6 months, at 2-monthintervals for another 6 months, and at 3-month inter-vals in the second year of the study. Lipid values wereprovided to the family physicians without recommen-dations.

Lipid determinations: Triglycerides and cholesterollevels were manually determined by routine methods(triglycerides, “GPO-PAP”; cholesterol, “CHOD-PAP” test kits; Boehringer Mannheim, Mannheim,Germany). LDL cholesterol was determined by se-quential preparatory ultracentrifugation (d5 1.006kg/m3, Beckmann L7-65 ultracentrifuge, Rotor 50 Ti,40,000 rpm, 4°C, 20 hours) and consecutive LDLcholesterol precipitation utilizing the BoehringerMannheim high-density lipoprotein (HDL) cholesteroltest kit. LDL cholesterol was recalculated from thefractions. The recovery rate was 976 7%. Lipopro-tein(a) (Lp(a)) was measured with a sandwich enzymeimmunoassay (Immunozym Lp(a), Immuno, Heidel-berg, Germany).

Quantitative coronary angiography: Angiogramswere taken after administration of 0.6 mg sublingualnitroglycerin on a biplane isocentric Philips Polydiag-nost C-arm and L-ARC Optimus (Best, The Nether-lands) system and recorded on cine film. Rotation andangulations as well as tube-to-image intensifier dis-tances were automatically recorded. Selection of end-diastolic frames was done in an off-line analysis. Twocorresponding biplane cine frames were 1.56-foldmagnified, videodigitized (CAP-35B; Elmo, Tokyo,Japan), and stored (5123 512 matrix size) for com-puter analysis in a cardiac work station (CARDIO500, Kontron, Munich, Germany). Quantitative coro-nary angiography by automatic contour detection ofthe repeated twofold magnified coronary segment wasperformed by a geometric edge differentiation tech-nique and a minimal cost algorithm.11 A radiographicmagnification factor derived from the 3-dimensionallocation of the vessel segment was used to scale theimage data to absolute dimensions.11–13The diameterswere corrected for distortions due to point-spreadfunction and first derivative-based edge detection.11,14

A 0.02 6 0.08 mm and 0.26 1.1% accuracy andprecision of the edge detection and calibration tech-

niques, respectively, were used, which have been es-tablished in validation studies.15,16 The angiographicview with the best visualization of the stenotic seg-ment was selected for minimal lumen diameter mea-surement using calculated segment foreshortening.17

Evaluation of angiograms was blinded with respect totherapy.

Clinical events: Nonfatal myocardial infarction, cor-onary balloon angioplasty or bypass grafting, stroke,new onset of occlusive peripheral vascular disease,cardiovascular death, and all-cause mortality wereprospectively defined as secondary end points.

Statistical analysis: An intention-to-treat analysiswas performed. For group comparisons, we used thechi-square test with continuity correction for dichoto-mous variables. For comparisons of age, blood glu-cose, and lipid values between groups we utilized theMann-Whitney U test. For comparisons to baselinewithin the groups, the Wilcoxon signed-rank test wasapplied. Differences in quantitative angiography weretested by the 2-sidedt test. The time to event wascalculated according to the Kaplan-Meier method withthe log-rank test. A stepwise regression model wasused for multivariate analysis. The significance levelwas set at p,0.05.

A sample size of 80 patients per group was esti-mated to show a 30% difference of progression and a14% difference in regression with 80% power and a pvalue of,0.05 in a 1-sided test. This calculation was

FIGURE 1. Trial profiles of the study.

1294 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 86 DECEMBER 15, 2000

performed under the expectation of a smaller differ-ence in cholesterol levels between the groups thanobserved and was derived from studies using muchless sensitive methods for quantification of angio-grams.

RESULTSOf the 870 patients examined, 135 were eligible

and randomized to either intensive treatment (groupA) or controls (group B) (Figure 1). Nine patients ingroup B withdrew their consent after learning towhich group they were included. Thus, 70 patients(group A) and 56 patients (group B) entered the study.Baseline characteristics in both groups were well

matched (Table 1). Fifty-five patients(44%) had an acute infarction, 37(67%) of those also had undergoneangioplasty, and the remaining 71patients (56%) were enrolled afterangioplasty only.

Total cholesterol was signifi-cantly reduced in the intensive ther-apy group by an average of 20% andLDL cholesterol by 28% (p,0.001)over the whole study period, whereasin group B, both values remainedclose to baseline during the wholestudy. Triglycerides, HDL-choles-terol, and Lp(a) levels remained un-changed in both groups (Table 2).LDL cholesterol of#130 mg/dl wasachieved in 40 of 70 patients ingroup A with pravastatin therapyonly. In 8 of 70 patients, LDL cho-lesterol of#130 mg/dl was achievedby combining pravastatin with cho-lestyramine and/or nicotinic acid.The other patients did not tolerate thecombination therapy because of sideeffects from cholestyramine or nico-tinic acid. Pravastatin was well tol-erated and there were no adverseevents in the early phase of treat-ment. The compliance was.90%according to pill count. In group B, 8of 56 patients achieved a target valueof an LDL cholesterol of#130 mg/dl, but only for a short period. Only13 patients in group B received anyantilipidemic drugs: 8 were on statintherapy, and 5 patients were on mis-cellaneous medications but also notfor more than a few months. Furthermedication was equally prescribed inboth groups (Table 1).

After 6 months, 47 of 56 patientsstill observed in group A had anevaluable control angiogram and 127non-angioplasty stenoses were mea-surable. In group B, there were com-parable angiograms for 38 of 51 pa-tients with a total of 113 measurable

non-angioplasty stenoses. After 6 months the minimallumen diameter increased by 0.056 0.20 mm ingroup A and decreased by 0.086 0.20 mm in groupB (p 5 0.004). After 24 months, a complete series of3 evaluable angiograms (at start, and at follow-upafter 6 and 24 months) could be measured for 31 of 51patients in group A (with a total of 77 non-angioplastystenoses) and in 27 of 46 patients in group B (with atotal of 72 stenoses). The minimal lumen diameterincreased with continued intensive treatment by0.13 6 0.29 mm in group A, whereas in group B itdecreased further by 0.186 0.27 mm from baselinevalues (p,0.001). The increase in difference between6 and 24 months was also significant (p5 0.021). For

TABLE 1 Baseline Characteristics of Patients, Reason for Inclusion, and AdditionalTherapy During Study (intensive care group A; standard care group B)

Group A(n 5 70)

Group B(n 5 56)

p Value(Odds Ratio; 90%

Confidence Intervals)

Baseline characteristics: median (minimum–maximum)Age (yrs) 55 (35–75) 59 (33–75) 0.16Men (%) 81% 79% 0.86 (1.5 [0.7–3.0])Total cholesterol* (mg/dl) 237 223 0.12

(204–319) (283–310)LDL cholesterol* (mg/dl) 176 172 0.66

(131–240) (132–239)HDL cholesterol* (mg/dl) 31 32 0.79

(10–63) (17–59)Triglycerides† (mg/dl) 146 137 0.052

(63–400) (42–423)Lp(a) (mg/dl) 6 13 0.26

(1–98) (1–82)Blood glucose (mg/dl)‡ 91.9 95.5 0.23

(68.5–125) (54.1–125)History

Myocardial infarction 45 39 0.66 (1.3 [0.6–2.7])(64%) (70%)

Coronary angioplasty/stenting 8 6 1.0 (1.1 [0.4–3.3])(12%) (11%)

Peripheral vascular disease 3 2 1.0 (1.2 [0.2–7.5])(4%) (4%)

Systemic hypertension 22 18 1.0 (1.0 [0.4–2.2])(31%) (32%)

Hyperlipemia 10 2 0.08 (4.5 [0.9–21.5])(14%) (4%)

Smoking 49 36 0.48 (1.4 [0.7–2.9])(70%) (62%)

TIA/PRIND/stroke 3 1 0.8 (1.2 [0.3–9.1])(4%) (2%)

Reason for inclusionAcute myocardial infarction 17% 11% 0.44 (1.7 [0.6–4.9])Myocardial infarction 1 angioplasty 29% 30% 0.98 (1.1 [0.5–2.4])Coronary angioplasty only 54% 59% 0.73 (1.2 [0.6–2.5])

Additional therapy (only patients who received medication for $ 6 mos)Antiplatelets 61/61 44/44 —

(100%) (100%)b blockers 23/61 23/44 0.2 (1.8 [0.8–4.0])

(38%) (52%)ACE inhibitors 11/61 11/44 0.33 (2.2 [0.8–5.9])

(18%) (25%)Calcium antagonists 15/61 17/44 0.19 (1.5 [0.8–4.5])

(25%) (39%)

*To convert into mMol/L multiply by 0.0259.†To convert into mMol/L multiply by 0.0113.‡To convert into mMol/L multiply by 0.0555.ACE 5 angiotensin-converting enzyme; PRIND 5 prolonged reversible ischemic neurological defict;

TIA 5 transient ischemic attack.

CORONARY ARTERY DISEASE/L–CAD STUDY 1295

angioplasty stenoses, however, treatment had no sig-nificant effect.

To illustrate the significant changes in coronarystenoses, we looked for increases or decreases in di-ameters of$0.3 mm (.10-fold outside the measure-ment accuracy). In Group A, in 23 of 77 stenoses, thelumen diameter increased, and in 6 of 77 stenoses itdecreased. In group B, there were increases by$0.3mm in 7 of 72 stenoses and decreases by$0.3 mm in20 of 72 stenoses (p,0.001) (Figure 2).

There were 18 prespecified clinical events in 16 of70 patients of group A and 34 events in 29 of 56patients of group B during the study (p5 0.005, oddsratio 3.6, confidence intervals 1.6 to 7.8). A total of 4patients died. In group A, a stroke in 1 patient wasdetermined to be cause of death by the family physi-cian, but no autopsy was performed. Another patientdied from primary ventricular fibrillation before tak-ing the first dose of pravastatin. In group B, 1 patientcommitted suicide and 1 died of colon cancer, whichwas unknown at the time of randomization. Two pa-tients of group B had primary ventricular fibrillation,but had successful out-of-hospital resuscitation. Onenonfatal myocardial infarction occurred in group B. Ineach group, 1 patient developed a new peripheralarterial occlusion and 1 suffered an ischemic stroke.There were 27 coronary interventions in 24 of 56patients in group B (3 coronary bypass grafts, 21 hadcoronary angioplasty). In group A, there were 12coronary interventions in 11 of 70 patients, including1 coronary bypass grafting. All but 2 interventions in2 patients in group A were due to clinical anginapectoris and/or a positive stress test. Figure 3 showsthe Kaplan-Meier curves for all major cardiovascularevents and subdivided for the subgroups of patientsincluded after an acute myocardial infarction or an-gioplasty only. In a multivariate stepwise regressionanalysis that included age (#65 and.65 years), gen-der, history of infarction, hypertension, smoking sta-tus, reason for inclusion, and therapy (group A or B),

only group A treatment was significant for clinicaloutcome (p5 0.006; odds ratio 3.0, confidence inter-vals 1.4 to 6.5).

DISCUSSIONPravastatin was effective in reducing cardiovascu-

lar events and in slowing the progression of athero-sclerosis in primary and secondary prevention popu-lations.2,3,9,18,19Compared with previous studies,4,5,9

this study is the first to show that cholesterol loweringwith pravastatin (in combination with cholestyramineand/or nicotinic acid when necessary) in patientstreated immediately after an acute coronary syndromecan prevent progression and achieve regression ofcoronary lesions within 6 months. This study, likeothers, however, could not find an influence of inten-sive antilipidemic treatment on the development ofrestenoses after angioplasty.20,21

By including the patients an average of 6 days afteran acute coronary syndrome, we selected a group athigh risk for new events. In a relatively short timeframe, we found marked differences in clinical coursewith a surprising consistency in patients included inthe study after an acute myocardial infarction as wellas after angioplasty alone. These findings correspondwith the results of the Long-term Intervention withPravastatin in Ischemic Disease study,3 and may becaused by properties of pravastatin beyond its lipid-lowering action, such as its anti-atherothrombotic orhematorheologic effects.22,23 Even if therapy was notblinded, the physicians who performed coronary an-giography and decided on interventions were not in-volved in the trial and were unaware of treatmentgroup allocation. Thus, an investigator bias with re-spect to interventions can be excluded. This presump-tion is underlined by the observation that in group Aonly 18 of 70 patients had a total of 24 unscheduledhospitalizations, whereas in group B, 33 unscheduledhospitalizations were necessary in 28 of 56 patientsdue to cardiovascular events (p5 0.009; odds ratio

TABLE 2 Lipid Values (median, minimum to maximum values) at Baseline, and at Day 30, 180, and 720, and Averaged Over the24 Months of the Study Period (intensive care group A, standard care group B)

Group A Group B

Day of Study Day of Study

0 30 180 720 0 30 180 720

Total cholesterol‡ 237 188* 182* 188* 223 236 224 226(204–319) (123–281) (124–305) (131–300) (183–310) (165–312) (176–368) (182–325)

LDL cholesterol‡ 176 123* 122* 125* 172 179 161 167(131–240) (79–237) (91–245) (59–221) (132–239) (102–238) (120–308) (115–256)

HDL cholesterol‡ 31 31 31 31 32 34 35 35(10–63) (22–50) (20–66) (19–65) (17–59) (22–68) (20–52) (22–52)

Lp(a) 6 — 8 8 13 — 14 17(1–98) (1–98) (1–101) (1–82) (1–99) (1–70)

Triglycerides§ 146 119 146† 139 137 119 102 131(63–400) (52–303) (58–1,483) (48–840) (42–423) (56–400) (37–393) (46–699)

*p ,0.001 compared with baseline and group B values.†p 5 0.008 compared with group B values at day 180.‡To convert into mMol/L multiply by 0.0259.§To convert into mMol/L multiply by 0.0113.There were no differences in baseline values between the groups. In group B there was no significant change in lipid levels during the study.

1296 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 86 DECEMBER 15, 2000

3.0, confidence intervals 1.4 to 6.3). Antilipidemictherapy was determined by the family physician in thecontrol group because placebo treatment was consid-ered unethical when the study was initiated. Corre-lated with other studies, such as the European Actionon Secondary Prevention through Intervention to Re-duce Events study,24 standard care was unsatisfactoryin our investigation. In group B, only 13 of 56 patientsreceived any antilipidemic medication, but only for ashort time and with inadequate effects.

There are limitations to our study: most notablydue to ethical considerations, it was an open study.The evaluation, however, of angiographic data wereperfomed in a blinded manner, as were the decisionsfor intervention during the study. The withdrawal ofconsent of 9 patients in the control group was notanticipated. Thus, a selection bias cannot be totallyexcluded, even if the risk profile between the groupswas equally distributed. By request we know from theBureau of Vital Statistics that these 9 patients werealive at the end of the 2-year investigational period.Even under the “worst-case assumption” that none ofthe 9 patients had had a severe adverse cardiovascular

event, the result would still be significant (p,0.02,odds ratio 0.4, confidence intervals 0.2 to 0.8) in favorof intensive therapy. A further limitation is the smallnumber of evaluable follow-up angiograms, partiallydue to poor film quality. In addition, some patientsrefused to agree to control angiography (1 of group Aand 2 patients of group B after 6 months and 7 patientsof group A and 8 patients of group B after 24 months,respectively). A selection bias with respect to angiog-raphy, however, seems to be improbable, because theproportion of patients within a valuable series of an-giograms was comparable in both groups at 6 months

FIGURE 2. Minimal lumen diameter (MLD) change in millimeters(y axis) and the number of occurrences in absolute numbers (xaxis) for individual nonpercutaneous transluminal coronary an-gioplasty stenoses within 24 months of study (group A, 77 mea-surable stenoses; group B, 72 stenoses). The intensive treatmentgroup is shown on the left side (grid bars); the standard caregroup is shown on the right side (striped bars). The occurrenceand the extent of increases in MLD with iintensive therapy (Tx)(A) are higher than with standard care (B). In contrast, the occur-rence and extent of decreases in MLD are lower with intensivetherapy (C) compared with standard care (D).

FIGURE 3. Kaplan-Meier curves for major cardiovascular events.(A) All patients. (B) Patients included after an acute myocardialinfarction. (C) Patients included after percutaneous transluminalcoronary angioplasty due to severe and/or unstable angina pec-toris alone.

CORONARY ARTERY DISEASE/L–CAD STUDY 1297

and at 24 months. Finally, because only a small num-ber of patients was included in this single-centerstudy, further larger studies are needed to confirm ourfindings.25

1. Scandinavian Simvastatin Survival Study Group. Randomised trial of choles-terol lowering in 4444 patients with coronary heart disease: the ScandinavianSimvastatin Survival Study (4S).Lancet1994;344:1383–1389.2. Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG,Brown L, Warnica JW, Arnold JM, Wun CC, Davis BR, Braunwald E. The effectof pravastatin on coronary events after myocardial infarction in patients withaverage cholesterol levels.N Engl J Med1996;335:1001–1009.3. LIPID Study Group. Prevention of cardiovascular events and death withpravastatin in patients with coronary heart disease and a broad range of initialcholesterol levels.N Engl J Med1998;339:1349–1357.4. Brown G, Albers JJ, Fisher LD, Schaefer SM, Lin JT, Kaplan C, Zhao XQ,Bisson BD, Fitzpatrick VF, Dodge HT. Regression of coronary artery disease asa result of intensive lipid-lowering therapy in men with high levels of apolipopro-tein B. N Engl J Med1990;323:1289–1295.5. MAAS Investigators. Effect of simvastatin on coronary atheroma: the Mul-ticentre Anti-Atheroma-Study (MAAS).Lancet1994;344:633–638.6. Schuler G, Hambrecht R, Schlierf G, Grunze M, Methfessel S, Hauer K,Kubler W. Myocardial perfusion and regression of coronary artery disease inpatients on a regimen of intensive physical exercise and low fat diet.J Am CollCardiol 1992;19:34–42.7. The Post Coronary Artery Bypass Graft Trial Investigators. The effect ofaggressive lowering of low-density lipoprotein cholesterol levels and low-doseanticoagulation on obstructive changes in saphenous-vein coronary-artery bypassgrafts.N Engl J Med1997;336:153–163.8. Watts GF, Lewis B, Brunt JN, Lewis ES, Coltart DJ, Smith LD, Mann JI, SwanAV. Effects on coronary artery disease of lipid-lowering diet, or diet pluscholestyramine, in the St Thomas Atherosclerosis Regression Study (STARS).Lancet1992;339:563–569.9. Jukema JW, Bruschke AV, van Boven AJ, Reiber JH, Bal ET, ZwindermanAH, Jansen H, Boerma GJ, van Rappard FM, Lie KI. Effects of lipid lowering bypravastatin on progression and regression of coronary artery disease in symp-tomatic men with normal to moderately elevated serum cholesterol levels: theRegression Growth Evaluation Statin Study (REGRESS).Circulation 1995;91:2528–2540.10. Study Group of European Atherosclerosis Society. Strategies for the preven-tion of coronary heart disease.Eur Heart J1987;8:77–88.11. Buchi M, Hess OM, Kirkeeide RL, Suter T, Muser M, Osenberg HP, NiedererP, Anliker M, Gould KL, Krayenbuhl HP. Validation of a new automatic systemfor biplane quantitative coronary arteriography.Int J Card Imaging1990;5:93–103.

12. Wunderlich W, Fischer F, Linderer T, Kirkeeide RL. Analytic isocentercalibration: a new approach for accurate x-ray gantries.Angiology1995;46:577–582.13. Wollschlager H, Lee P, Zeiher A, Solzbach U, Bonzel T, Just H. Improve-ment of quantitative angiography by exact calculation of radiological magnifica-tion factors.Comp Cardiol1985;12:359–362.14. Wunderlich W, Linderer T, Backs B, Fischer F, No¨hring R, Schro¨der R.Optimizing edge detection in quantitative coronary arteriography: problems andproposals.Comp Cardiol1993;20:583–586.15. Kristiansen G, Wunderlich W, Fischer F, Ro¨hrig B, Arntz HR, Horstkotte D,Schultheiss HP. Accuracy and precision of the analytic calibration method inquantitative coronary arteriography.Comput Cardiol1996;23:553–556.16. Hausleiter J, Jost S, Nolte CW, Dirschinger J, Kastrati A, Stiel GM, Wunder-lich W, Fischer F, Linderer T, Hausmann D, Schomig A. Comparative in-vitrovalidation of eight first- and second-generation quantitative coronary angiographysystems.Coron Artery Dis1997;8:83–90.17. Wunderlich W, Fischer F, Morguet AJ, Horstkotte D, Schultheiss HP.Foreshortening display: an on-line method for spatial interpretation and projec-tion assessment of the target coronary segment.Comp Cardiol1998;25:437–440.18. Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW,McKillop JH, Packard CJ. Prevention of coronary heart disease with pravastatinin men with hypercholesterolemia.N Engl J Med1995;33:1301–1307.19. Salonen R, Nyyssonen K, Prokkala E, Rummukaien J, Belder R. KuopioAtherosclerosis Prevention Study (KAPS): a population-based primary preven-tive trial of the effect of LDL lowering on atherosclerotic progression in carotidand femoral arteries.Circulation 1995;92:1758–1764.20. Bertrand ME, McFadden EP, Fruchart JC, Van Belle E, Commeau P, GrollierG, Bassand JP, Machecourt J, Cassagnes J, Mossard JM, et al. Effect of prava-statin on angiographic restenosis after coronary balloon angioplasty.J Am CollCardiol 1997;30:863–869.21. Weintraub WS, Boccuzzi SJ, Klein L. Lack of effect of lovastatin onrestenosis after coronary angioplasty.N Engl J Med1994;331:1331–1337.22. Rosenson RS, Tangney CC. Antiatherothrombotic properties of statins:implications for cardiovascular event reduction.JAMA 1998;279:1643–1650.23. Dangas G, Badimon JJ, Smith DA, Unger AH, Levine D, Shao JH, Meraj P,Fier C, Fallon JT, Ambrose JA. Pravastatin therapy in hyperlipidemia: effects onthrombus formation and the systemic hemostatic profile.J Am Coll Cardiol1999;33:1294–1304.24. EUROASPIRE Study Group. A European Society of Cardiology survey ofsecondary prevention of coronary heart disease: principal results.Eur Heart J1997;18:1569–1582.25. Schwartz GG, Oliver MF, Ezekowitz MD, Ganz P, Waters D, Kane JP, TexterM, Pressler ML, Black D, Chaitman BR, Olsson AG. Rationale and design of theMyocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIR-ACL) study that evaluates atorvastatin in unstable angina pectoris and in non–Q-wave acute myocardial infarction.Am J Cardiol1998;81:578–581.

1298 THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 86 DECEMBER 15, 2000