sprtm

spMe

mSm

YbPaJsw

0d

Comparison of Efficacy and Safety of Atorvastatin (80 mg) toSimvastatin (20 to 40 mg) in Patients Aged <65 Versus

>65 Years With Coronary Heart Disease (from theIncremental DEcrease through Aggressive

Lipid Lowering [IDEAL] Study)Matti J. Tikkanen, MD, PhDa,*, Ingar Holme, PhDb, Nilo B. Cater, MDc, Michael Szarek, PhDc,

Ole Faergeman, MD, DMScd, John J.P. Kastelein, MD, PhDe, Anders G. Olsson, MD, PhDf,Mogens Lytken Larsen, MD, DMScd, Christina Lindahl, MDg, and

Terje R. Pedersen, MD, PhDb, for the IDEAL Investigators

The efficacy and safety of atorvastatin (80 mg/day) versus simvastatin (20 to 40 mg/day) inolder (age >65 years) versus younger (<65 years) patients were assessed in a prespecifiedsecondary analysis of the 8,888 patients with myocardial infarction in the IDEAL trial, arandomized open-label study. Several cardiovascular end points were evaluated, includingthe occurrence of a first major coronary event (MCE; nonfatal myocardial infarction,coronary heart disease death, or resuscitated cardiac arrest), the primary end point of thetrial, and occurrence of any cardiovascular event (MCE, stroke, revascularization, unstableangina, congestive heart failure, and peripheral artery disease). Although there were nosignificant interactions between age and treatment, the magnitude of effect in favor ofatorvastatin was higher in younger versus older patients (occurrence of first MCE, hazardratio [HR] 0.80, 95% confidence interval [CI] 0.66 to 0.98; and HR 0.95, 95% CI 0.80 to1.15, respectively; occurrence of any cardiovascular (CV) event, HR 0.80, 95% CI 0.71 to0.89; and HR 0.88, 95% CI 0.79 to 0.99, respectively). These results were likely influencedby adherence, which was lower in older patients and those receiving atorvastatin comparedwith those receiving simvastatin. Rates of any reported serious adverse event were higherin older patients, but did not differ between the 2 statin groups. In conclusion, except forany CV events in the older group, significant reductions in primary and secondary endpoints were observed only in patients <65 years of age. The safety of atorvastatin (80 mg)and simvastatin (20 to 40 mg) was similar in patients aged <65 and >65 years with stable

coronary disease. © 2009 Elsevier Inc. (Am J Cardiol 2009;103:577–582)

ctow

DPKtHPMaeP

PAPS

4

Starting in 1999, the Incremental DEcrease through Aggres-ive Lipid lowering (IDEAL) trial1 randomly assigned 8,888atients aged �80 years with previous myocardial infarction toeceive either intensive therapy with atorvastatin 80 mg/day orhe standard statin therapy at the time (simvastatin 20 to 40g/day) to compare the ability of these regimens to lower

aDepartment of Medicine, Helsinki University Central Hospital, Hel-inki, Finland; bCentre for Preventative Medicine, Ullevål University Hos-ital, Oslo, Norway; cPfizer Inc., New York, New York; dDepartment ofedicine Cardiology A, Århus University Hospital, Århus, Denmark;

Academic Hospital Amsterdam, Amsterdam, The Netherlands; fDepart-ent of Internal Medicine, University Hospital, Linköping; and gPfizerweden, Täby, Sweden. Manuscript received August 15, 2008; revisedanuscript received and accepted October 24, 2008.

This investigator-initiated study was supported by Pfizer Inc., Nework, New York. No assistance with content development was provided,ut editorial support was provided by Chris Cadman, PhD, at Envisionharma, funded by Pfizer Inc. Dr. Tikkanen was supported by honorariand consultancy fees from Pfizer and Merck,Whitehouse Station, Newersey, and a research grant from Takeda, Deerfield, Illinois; Dr. Olsson wasupported by a consulting agreements with AstraZeneca, Wilmington, Dela-

are, KaroBio, Huddinge, Sweden, Pfizer, and Roche, Basel, Switzerland;

002-9149/09/$ – see front matter © 2009 Elsevier Inc.oi:10.1016/j.amjcard.2008.10.029

ardiovascular risk. This trial differed from other major statinrials in several respects. It used a prospective, randomized,pen-label, blinded end point classification design.2 Patientsere given prescriptions for the study drugs and had their

r. Larsen was supported by consultancy fees and research grants fromfizer, MSD, Whitehouse Station, New Jersey, and AstraZeneca; Dr.astelein was supported by consulting and lecture fees from Pfizer, As-

raZeneca, Merck, and Schering-Plough, Kenilworth, New Jersey; Dr.olme was supported by Steering Committee honoraria from Pfizer; Dr.edersen was supported by consultation fees and honoraria from Pfizer,erck, Merck AG, Zug, Switzerland, and AstraZeneca, and research grants

nd Steering Committee fees from Pfizer and Merck; Dr. Cater is anmployee of Pfizer Inc., and Drs. Szarek and Lindahl were employees offizer at the time of the data analysis and drafting of the report.

The IDEAL Steering Committee consists of: Terje R. Pedersen, MD,hD (Chair); Ole Faergeman, MD, DMSc; John J.P. Kastelein, MD, PhD;nders G. Olsson, MD, PhD; Matti J. Tikkanen, MD, PhD; Ingar Holme,hD; Mogens Lytken Larsen, MD, DMSc; Christina Lindahl, MD; andonal Bhatia, MD.

*Corresponding author: Tel: �358-9-4717-2210; Fax: �358-9-4717-013.

E-mail address: matti.j.tikkanen@helsinki.fi (M.J. Tikkanen).

www.AJConline.org

pwdTettcots

M

eowl8aoFatpm

weise(nrelctd

mpCddg

R

(B

Fb

TB

V

MMP

578 The American Journal of Cardiology (www.AJConline.org)

rescriptions filled at their designated pharmacy. Most patientsere already on statin therapy and were switched to the studyrug without a wash-out or run-in period at randomization.hus, in some respects, the IDEAL study design resembledveryday clinical practice more closely than other major statinrials. In view of the reported decrease in adherence to statinreatment in elderly patients in routine clinical settings,3,4 wearried out a prespecified secondary analysis to more thor-ughly compare treatment efficacy in younger (�65 years atime of randomization) and older (65 to 80 years) patients, withpecial emphasis on adherence.

ethods

The open-label prospective, randomized, open-label, blindednd point classification design; primary features; and findingsf the IDEAL trial have been reported previously.1,5 IDEALas carried out in Denmark, Finland, Iceland, The Nether-

ands, Norway, and Sweden. Patients aged �80 years (n �,888) with a previous myocardial infarction were randomlyssigned to receive either intensive (atorvastatin 80 mg/day)r moderate statin therapy (simvastatin 20 to 40 mg/day;igure 1). The latter regimen was the standard statin therapyt the time in participating countries. Patients were switchedo study drugs at randomization without wash-out or run-ineriods and had their prescriptions filled at designated phar-

igure 1. IDEAL study design. *Except for the 6% of patients in Finlanlinded end point evaluation; TC � total cholesterol.

able 1aseline characteristics of younger and older patients

ariable

�65

Simvastatin(n � 2,583)

ean age (yrs) 55.1 � 6.4en 2,166 (83.9%)

rerandomization statin therapy 2,022 (78.3%)Simvastatin 1,318 (51.0%)Atorvastatin 335 (13.0%)Pravastatin 262 (10.1%)Other statins 108 (4.2%)

Values expressed as mean � SD or number (percent).

acies, except for the 6% of patients enrolled in Finland, l

ho received their drugs from study centers. Adherence forach subject was defined as total study medication exposure,n other words, the percentage of time using the prescribedtudy drug of total follow-up time. The primary compositend point in IDEAL was time to first major coronary eventMCE), defined as coronary heart disease (CHD) death,onfatal myocardial infarction, or resuscitated cardiac ar-est. Secondary end points included major cardiovascularvents (MCE and stroke), any CHD event (MCE, revascu-arization, and hospitalization for unstable angina), and anyardiovascular event (any CHD event, stroke, hospitaliza-ion for congestive heart failure, and peripheral arterialisease).

Descriptive statistics with mean � SE were given foreasurements of lipid variables. Differences between end

oint rates within each age group were analyzed using aox proportional regression model with time to event as theependent time variable and treatment as the only indepen-ent variable. Hazard ratios (HRs) between treatmentroups with 95% confidence limits were provided.

esults

There were 5,129 patients (58%) aged �65 years and 3,75942%) aged 65 to 80 years at the time of randomization.aseline characteristics of the 2 age groups by treatment are

myocardial infarction; PROBE � prospective, randomized, open-label

Age (yrs)

65–80

torvastatin� 2,546)

Simvastatin(n � 1,866)

Atorvastatin(n � 1,893)

5.2 � 6.4 70.7 � 3.9 70.8 � 4.051 (84.5%) 1,431 (76.7%) 1,439 (76.0%)83 (77.9%) 1,353 (72.5%) 1,355 (71.6%)00 (51.1%) 912 (48.9%) 933 (49.3%)16 (12.4%) 177 (9.5%) 183 (9.7%)71 (10.6%) 169 (9.1%) 148 (7.8%)00 (3.9%) 96 (5.1%) 95 (5.0%)

d. MI �

A(n

52,11,91,3

321

isted in Table 1. Mean age of younger patients was approx-

imypytpllg

pl8at

pityywt

gy(sfdltaecvw

f(ivtawM

FA

TB

L

T

L

T

H

tl

579Coronary Artery Disease/Intensive Statin Therapy in Older Patients

mately 55 years, and that of older patients was approxi-ately 71 years. There were more women in the older than

ounger group, and slightly more younger patients were onrerandomization statin therapy. Approximately 51% ofounger and 49% of older patients were on simvastatinherapy, whereas only 13% of younger and 10% of olderatients were on atorvastatin therapy. Baseline low-densityipoprotein (LDL), high-density lipoprotein, and total cho-esterol and triglycerides were similar in the 2 treatmentroups (Table 2).

As previously reported, overall adherence, defined asercentage of time using the prescribed drug of total fol-ow-up time, was high, although 11% of patients (979 of,888) were �80% adherent. Regardless of age, overalldherence was lower on atorvastatin (89%) than simvastatin

igure 2. Overall adherence in younger and older patients by treatment.dherence in favor of simvastatin in both age groups (p �0.001).

able 2aseline and follow-up levels of lipids and lipoproteins

ipids and Lipoproteins Baseline

otal cholesterol (mg/dl)Simvastatin, �65 yrs 196 � 0.8Atorvastatin, �65 yrs 197 � 0.8Simvastatin, �65 yrs 195 � 0.8Atorvastatin, �65 yrs 197 � 0.8DL cholesterol (mg/dl)Simvastatin, �65 yrs 121 � 0.8Atorvastatin, �65 yrs 121 � 0.8Simvastatin, �65 yrs 121 � 0.8Atorvastatin, �65 yrs 122 � 0.8riglycerides (mg/dl)Simvastatin, �65 yrs 156 � 1.8Atorvastatin, �65 yrs 160 � 1.8Simvastatin, �65 yrs 135 � 1.8Atorvastatin, �65 yrs 138 � 1.8igh density lipoprotein-cholesterol (mg/dl)Simvastatin, �65 yrs 45 � 0.4Atorvastatin, �65 yrs 45 � 0.4Simvastatin, �65 yrs 48 � 0.4Atorvastatin, �65 yrs 48 � 0.4

Values expressed as mean � SE or mean difference (95% CI). To convriglycerides to mmol/L, multiply by 0.0113. The p values for absolute eipoprotein cholesterol in patients aged �65 years, for which p � 0.002.

* Last obtained measurement.

herapy (95%; p �0.0001).1 Adherence was lower for older fi

atients in both treatment groups, but the overall differencen adherence between age groups was smaller in simvasta-in-treated patients (95% [�65 years] vs 94% [65 to 80ears]) than in atorvastatin-treated patients (91% [�65ears] vs 84% [65 to 80 years]; Figure 2). These differencesere apparent after 1 year of follow-up and persisted

hroughout the course of the trial.Differences in plasma lipids between the 2 treatment

roups are listed in Table 2 according to age group. Inounger patients, 1-year LDL cholesterol was 24.6 mg/dl0.64 mmol/L) lower in the atorvastatin group than in theimvastatin group. In older patients, the corresponding dif-erence was 20.4 mg/dl (0.53 mmol/L). The incrementalecreases associated with atorvastatin therapy in total cho-esterol and triglycerides were also higher in younger pa-ients. Regardless of age, the incremental decreases associ-ted with atorvastatin therapy became smaller toward thend of the trial. High-density lipoprotein cholesterol in-reased with both statins, although slightly more with sim-astatin, and again, the difference between treatment groupsas smaller in older patients (Table 2).There was no significant heterogeneity of treatment ef-

ect by age for any composite end point in the IDEAL studyFigure 3), indicating that the benefit of atorvastatin was sim-lar for younger and older patients. Nevertheless, the cardio-ascular risk reductions associated with atorvastatin therapyended to be numerically lower in the older than youngerge group. Thus, although atorvastatin was associatedith a 20% decrease in risk of the primary end point ofCE in the group aged �65 years (HR 0.80, 95% con-

Year Final Value* Absolute Difference BetweenAtorvastatin And Simvastatin

at 1 Year

� 0.8 180 � 0.8 �30.9 (�32.5 to �29.4)� 0.8 154 � 0.8� 0.8 173 � 0.8 �25.5 (�27.5 to �23.6)� 0.8 152 � 0.8

� 0.4 104 � 0.8 �24.6 (�25.9 to �23.2)� 0.4 83 � 0.8� 0.8 99 � 0.8 �20.4 (�22.0 to �18.6)� 0.8 83 � 0.8

� 1.8 150 � 1.8 �27.4 (�31.0 to �23.9)� 1.8 129 � 1.8� 1.8 127 � 1.8 �23.0 (�26.5 to �19.5)� 0.9 112 � 1.8

� 0.4 48 � 0.4 �1.5 (�1.9 to �1.2)� 0.4 46 � 0.4� 0.4 49 � 0.4 �0.8 (�1.2 to �0.4)� 0.4 48 � 0.4

es for cholesterol to mmol/L, multiply by 0.02586; to convert values forere �0.001 for all between-group comparisons, except for high-density

1

178147172147

104799979

148123127107

46444948

ert valuffect w

dence interval [CI] 0.66 to 0.98), with similarly signif-

it�weF

gewg

Fh periphe

TR

V

AD

M

LN �

TM

A

OU —

580 The American Journal of Cardiology (www.AJConline.org)

cant reductions in the secondary composite end points,he corresponding reductions in risk in the group aged

65 years were 4% to 12%, and statistical significanceas achieved for only the end point of any cardiovascular

vents in older patients (HR 0.88; 95% CI 0.79 to 0.99;

igure 3. Intensive atorvastatin therapy reduced event rates in both youngeart failure; CV � cardiovascular; MI � myocardial infarction; PAD �

able 3ates of adverse events and treatment discontinuation

ariable

ny serious adverse eventose reductions or discontinuationsSubjects with dose reduction or temporary discontinuation due to

adverse eventsSubjects who permanently discontinued treatment due to adverse eventsuscle and liver abnormalities

Myopathy reported by investigatorRhabdomyolysis reported by investigatorCreatine kinase �10 � ULN at 2 consecutive measures*AST �3 � ULN at 2 consecutive measures*ALT �3 � ULN at 2 consecutive measures*

* Time range is 4 to 10 days apart.ALT � alanine aminotransferase; AST � aspartate aminotransferase; U

able 4ortality rates

�65

Simvastatin(n � 2,583)

Atorvastatin(n � 2,546)

HR

ll-cause mortality 124 (4.8%) 120 (4.7%) 0.99Cardiovascular 71 (2.7%) 72 (2.8%) 1.04Noncardiovascular 53 (2.1%) 48 (1.9%) 0.92Malignant disease 39 (1.5%) 38 (1.5%) 0.99Suicide/violence/accidental death 6 (0.2%) 3 (0.1%)ther 6 (0.2%) 6 (0.2%)nclassified 2 (0.1%) 1 (0.0%)

igure 3). a

There were no differences between the 2 treatmentroups in number of patients who reported serious adversevents (Table 3), but the percentage reporting such eventsas higher in patients �65 years in both treatmentroups. In the younger group, 40.1% of patients receiving

older patients. *p values are for test of heterogeneity. CHF � congestiveral artery disease.

Age (yrs)

�65 65–80

Simvastatinn � 2,583)

Atorvastatin(n � 2,546)

Simvastatin(n � 1,866)

Atorvastatin(n � 1,893)

,058 (41.0%) 1,021 (40.1%) 1,050 (56.3%) 1,043 (55.1%)

226 (8.7%) 564 (22.2%) 206 (11.0%) 535 (28.3%)

109 (4.2%) 202 (7.9%) 77 (4.1%) 224 (11.8%)

6 2 5 41 1 2 10 0 0 0

1 (0.04%) 4 (0.16%) 1 (0.05%) 14 (0.74%)2 (0.08%) 14 (0.55%) 3 (0.16%) 29 (1.53%)

upper limit of normal.

Age (yrs)

65–80

I) pValue

Simvastatin(n � 1,866)

Atorvastatin(n � 1,893)

HR (95% CI) pValue

.27) 0.92 250 (13.4%) 246 (13.0%) 0.97 (0.81–1.15) 0.69

.44) 0.83 147 (7.9%) 151 (8.0%) 1.01 (0.8–1.26) 0.95

.36) 0.69 103 (5.5%) 95 (5.0%) 0.9 (0.68–1.2) 0.48

.55) 0.98 73 (3.9%) 61 (3.2%) 0.82 (0.58–1.15) 0.26— 3 (0.2%) 2 (0.1%) — —— 24 (1.3%) 26 (1.4%) — —— 3 (0.2%) 6 (0.3%) — —

er and

(

1

(95% C

(0.77–1(0.75–1(0.62–1(0.64–1

——

torvastatin and 41.0% receiving simvastatin reported

stbtccs(auwy

miropg

dcpoctfng

D

tatoep(wrrmdmottsawvapqwdf

e

edIwprtsppaTlvlstoa

tawtppLMpfescagi

rttaphpatespwcg

581Coronary Artery Disease/Intensive Statin Therapy in Older Patients

erious adverse events. The corresponding incidences inhe older group were 55.1% and 56.3%, respectively. Inoth age groups, more patients receiving atorvastatinhan simvastatin had dose reductions or temporary dis-ontinuations or permanently discontinued treatment be-ause of adverse effects, and the difference favoringimvastatin was numerically greater in the older groupTable 3). In patients �65 years, 11.8% of patients ontorvastatin and 4.1% on simvastatin therapy discontin-ed treatment permanently because of adverse events,hereas the corresponding proportions in patients �65ears were 7.9% and 4.2%, respectively.

Although there were several reports by investigators ofyopathy and rhabdomyolysis, there were no significant

ncreases in creatine kinase, measured by the central labo-atory (Table 3). Rates of significant increases in aspartater alanine aminotransferase were low. However, as ex-ected, they were numerically higher in the atorvastatinroup (Table 3).

As listed in Table 4, in each age group, there were noifferences between treatment groups in rates of cardiovas-ular, noncardiovascular, or all-cause mortality. As ex-ected, mortality rates in all categories were higher in thelder than younger age group. However, within each ageategory, there were no significant differences betweenreatment groups in mortality rates. Of note, rates of deathsrom malignant disease were similar in younger patients andumerically lower in the atorvastatin versus simvastatinroup in older patients (HR 0.82, 95% CI 0.58 to 1.15).

iscussion

Several large placebo-controlled trials have clearly shownhat the preventive benefits of LDL cholesterol reductionssociated with statin therapy seen in middle-aged popula-ions were also seen in patients �65 years.6,7 Conversely,nly 1 major clinical outcomes statin trial compared thefficacy of intensive with moderate statin therapy in elderlyatients with stable CHD.8 The Treating to New TargetsTNT) trial included men and women aged 35 to 75 yearsith established CHD who entered an 8-week open-label

un-in period with atorvastatin 10 mg. At the end of theun-in phase, 10,001 patients with LDL cholesterol �130g/dL (�3.4 mmol/L) were randomly allocated to receive

ouble-blind therapy with atorvastatin 10 or 80 mg for aedian of 4.9 years.9 The prespecified secondary analysis

f the group aged �65 years showed no heterogeneity ofreatment effect for primary or secondary outcomes. Al-hough the relative risk reduction with intensive therapy waslightly lower in older compared with younger patients, thebsolute benefit was the same in both age groups.8 Thereas also a small increase in rates of treatment-related ad-erse events and withdrawals in older patients who receivedtorvastatin 80 mg/day compared with those of youngeratients. Although these differences between age groupsualitatively resembled those observed in IDEAL, theyere much less pronounced in TNT, and we believe thisifference between the 2 trials can be ascribed to the designeatures analyzed next.

In routine clinical care settings, it is a common experi-

nce that adherence to statin therapy is unsatisfactory in

lderly patients, and a significant proportion of patientsiscontinue therapy as early as 6 months after initiation.3,4

n our open-label trial, adherence was relatively good, buthy was adherence lower in patients receiving atorvastatin,articularly in older patients? The proportion of patientseporting serious adverse events was virtually the same inhe 2 treatment groups, but more atorvastatin than simva-tatin patients discontinued treatment prematurely. Threeotential explanations should be considered. First, it wasossible in principle that older patients did not toleratetorvastatin as well as simvastatin. However, data listed inable 3 suggested that this possibility was extremely un-

ikely because rates of reported serious adverse events wereirtually identical in the 2 treatment groups. Instead, dataisted in Table 3 indicated that it was the reaction of phy-icians and patients to adverse events that differed betweenreatment groups. This finding therefore suggested thatther explanations were responsible for higher discontinu-tion rates in atorvastatin-treated patients.

Second, IDEAL was an open-label study.2 Neither pa-ients nor physicians were blinded to treatment allocationnd were in general more familiar with simvastatin thanith the more potent atorvastatin. At the time of the study,

here were significant concerns about the risks of decreasinglasma LDL cholesterol to very low levels. The IDEALrotocol specified that statin dosage should be reduced ifDL cholesterol decreased to �39 mg/dl (�1 mmol/L).oreover, these concerns were heightened early in the trial

eriod when another statin (cerivastatin) was withdrawnrom the market in 2001 because of severe adverse effects,10

specially in older patients, and this withdrawal was theubject of extensive media coverage. Patients and physi-ians in the IDEAL trial were therefore more likely toscribe an adverse event to treatment in the atorvastatinroup than in the simvastatin group, especially if it occurredn an older patient.

A third potential explanation was the influence of pre-andomization statin therapy. Approximately 76% of pa-ients were treated with a statin before randomization, andhey were switched directly to 1 of the study drugs withoutwash-out or run-in period. Before randomization, 50% ofatients had been on simvastatin therapy, whereas only 12%ad been on atorvastatin therapy. A larger proportion ofatients were therefore accustomed to simvastatin than totorvastatin. Thus, these potential explanations suggestedhere may have been greater inclination to regard adversevents as drug related if the patient was receiving atorva-tatin than simvastatin, although the same proportion ofatients in both groups reported adverse events. Our studyas limited to patients not �80 years at baseline. Women

onstituted a larger proportion of the older than youngerroup, but adjusting for gender did not alter results.

1. Pedersen TR, Faergeman O, Kastelein JJ, Olsson AG, Tikkanen MJ,Holme I, Larsen ML, Bendiksen FS, Lindahl C, Szarek M, Tsai J.High-dose atorvastatin vs usual-dose simvastatin for secondary pre-vention after myocardial infarction: the IDEAL study: a randomizedcontrolled trial. JAMA 2005;294:2437–2445.

2. Hansson L, Hedner T, Dahlof B. Prospective randomized open blindedend-point (PROBE) study. A novel design for intervention trials.

Prospective Randomized Open Blinded End-Point. Blood Press 1992;1:113–119.

1

582 The American Journal of Cardiology (www.AJConline.org)

3. Jackevicius CA, Mamdani M, Tu JV. Adherence with statin therapy inelderly patients with and without acute coronary syndromes. JAMA2002;288:462–467.

4. Benner JS, Glynn RJ, Mogun H, Neumann PJ, Weinstein MC, AvornJ. Long-term persistence in use of statin therapy in elderly patients.JAMA 2002;288:455–461.

5. Pedersen TR, Faergeman O, Kastelein JJ, Olsson AG, Tikkanen MJ,Holme I, Larsen ML, Bendiksen FS, Lindahl C, Palmer G. Design andbaseline characteristics of the Incremental Decrease in End Points throughAggressive Lipid Lowering Study. Am J Cardiol 2004;94:720–724.

6. LaRosa JC, He J, Vupputuri S. Effect of statins on risk of coronarydisease: a meta-analysis of randomized controlled trials. JAMA 1999;282:2340–2346.

7. Colhoun HM, Betteridge DJ, Durrington PN, Hitman GA, Neil

HAW, Livingstone SJ, Thomason MJ, Mackness MI, Charlton-

Menys V, Fuller JH; CARDS Investigators. Primary preventionof cardiovascular disease with atorvastatin in type 2 diabetes inthe Collaborative Atorvastatin Diabetes Study (CARDS): a multi-centre randomized placebo-controlled trial. Lancet 2004;364:685– 696.

8. Wenger NK, Lewis SJ, Herrington DM, Bittner V, Welty FK. Out-comes of using high- or low-dose atorvastatin in patients 65 years ofage or older with stable coronary heart disease. Ann Intern Med2007;147:1–9.

9. LaRosa JC, Grundy SM, Waters DD, Shear C, Barter P, Fruchart JC,Gotto AM, Greten H, Kastelein JJ, Shepherd J, Wenger NK. Intensivelipid lowering with atorvastatin in patients with stable coronary dis-ease. N Engl J Med 2005;352:1425–1435.

0. Furberg CD, Pitt B. Withdrawal of cerivastatin from the world market.

Curr Control Trials Cardiovasc Med 2001;2:205–207.