statins for primary prevention

7/31/2019 Statins for Primary Prevention

1/14

A lthough statins are known to improvesurvival and relevant clinical outcomesin high-risk populations, 1 evidence of their clinical bene t in lower risk populations is

more equivocal. Initially, low-risk populationswere de ned by the absence of known coronaryartery disease (and their treatment was termedprimary prevention). However, it was sub -sequently recognized that these populationsincluded both patients at very high risk of coro-nary artery disease (e.g., those with severeperipheral vascular disease) and those at verylow risk (e.g., those aged < 40 years who haveno diabetes or hypertension and have low- density lipoprotein cholesterol level of less than1.8 mmol/L). Accordingly, current guidelinesfor the use of statins are based on the projected

risk of an atherosclerotic event rather thansolely on the presence or absence of knowncoronary artery disease. 2,3

Results of the recent JUPITER study (Justi -

cation for the Use of Statins in Prevention: anIntervention Trial Evaluating Rosuvastatin) 4 haverenewed enthusiasm for the use of statins in peo-ple without a history of coronary artery diseaseand have generated further controversy as towhether high-potency statins such as rosuva statinand atorvastatin lead to better clinical outcomesthan low-potency statins such as prava statin, sim-vastatin, uvastatin and lovastatin. We did a sys-tematic review of randomized trials to assess theef cacy and harms of statins in people at low car-diovascular risk, including indirect comparisonsof high-potency and low-potency statins.

Ef cacy of statins for primary prevention in peopleat low cardiovascular risk: a meta-analysis

Marcello Tonelli MD SM, Anita Lloyd MSc, Fiona Clement PhD, Jon Conly BA, Don Husereau BScPharm MSc,Brenda Hemmelgarn PhD MD, Scott Klarenbach MD MSc, Finlay A. McAlister MD MSc,Natasha Wiebe MMath PStat, Braden Manns MD MSc; for the Alberta Kidney Disease Network

Competing interests:Marcello Tonelli hasreceived research supportfrom P zer; in 2011 heserved on an advisory boardfor Merck, providing expertopinion on global chronickidney disease andcardiovascular disease.Brenda Hemmelgarn hasreceived research supportfrom Merck and Amgen.Braden Manns has receivedresearch support fromMerck Canada. Nocompeting interests declaredby the other authors.

This article has been peerreviewed.

Correspondence to:Dr. Marcello Tonelli, mtonelli-admin @med.ualberta.ca

CMAJ 2011. DOI:10.1503 /cmaj.101280

ResearchCMAJ

Background: Statins were initially used to im -

prove cardiovascular outcomes in people withestablished coronary artery disease, but re -cently their use has become more common inpeople at low cardiovascular risk. We did a sys-tematic review of randomized trials to assessthe ef cacy and harms of statins in these individuals.

Methods: We searched MEDLINE and EMBASE(to Jan. 28, 2011), registries of health technol-ogy assessments and clinical trials, and refer-ence lists of relevant reviews. We included tri-als that randomly assigned participants at lowcardiovascular risk to receive a statin versus aplacebo or no statin. We de ned low risk asan observed 10-year risk of less than 20% forcardiovascular-related death or nonfatal myo -cardial infarction, but we explored other de -nitions in sensitivity analyses.

Results: We identi ed 29 eligible trials involv-ing a total of 80 711 participants. All-cause

mortality was signi cantly lower among pa -

tients receiving a statin than among controls(relative risk [RR] 0.90, 95% con dence inter-val [CI] 0.840.97) for trials with a 10-year riskof cardiovascular disease < 20% [primaryanalysis] and 0.83, 95% CI 0.730.94, for trialswith 10-year risk < 10% [sensitivity analysis]).Patients in the statin group were also signi -cantly less likely than controls to have nonfa-tal myocardial infarction (RR 0.64, 95% CI0.490.84) and nonfatal stroke (RR 0.81, 95%CI 0.680.96). Neither metaregression norstrati ed analyses suggested statistically sig-ni cant differences in ef cacy between high-and low-potency statins, or larger reductions

in cholesterol.Interpretation: Statins were found to be ef -cacious in preventing death and cardiovascu-lar morbidity in people at low cardiovascularrisk. Reductions in relative risk were similar tothose seen in patients with a history of coro-nary artery disease.

Abstract

2011 Canadian Medical Association or its licensors CMAJ, November 8, 2011, 183(16) E1189

See related research article by Conly and colleagues at www.cmaj.ca/lookup/doi/10.1503/cmaj.101281 and related commentary by Gupta atwww.cmaj.ca/lookup/doi/10.1503/cmaj.111674


2/14

Methods

We performed a systematic review of publishedand unpublished randomized controlled trialsthat compared statins with no statin or placebo.We used accepted methods for literaturesearches, article selection, data extraction and

risk-of-bias assessment. The review was reportedaccording to accepted guidelines. 5,6

Literature searchWe searched MEDLINE (1950 to Jan. 28, 2011)and EMBASE (1950 to Jan. 28, 2011). Details of the search strategy appear in Appendix 1 (avail-able at www.cmaj.ca/lookup /suppl/doi:10.1503

/cmaj.101280/-/DC1). Because of high searchyields, an update of a modi ed version of theNational Institute for Health and Clinical Excel-lence search 7 was performed, with elimination of studies published before 2003. Searches were

not restricted by language (when a translatorcould be found).

We also searched registries of health technol-ogy assessments and clinical trials, conductedmanual searches of reference lists of relevantreviews, and contacted Canadian manufacturersof statins (Astra Zeneca, Merck Frosst, P zer,Bristol Myers Squibb, Novartis) for additionalstudies and unpublished reports of trials.

Study selection and validity assessmentWe included parallel-group randomized con-trolled trials if they included people 16 years orolder who were at low cardiovascular risk (asde ned in the next paragraph), the follow-upperiod was at least six months, and an eligiblestatin (atorvastatin, uvastatin, lovastatin, prava -statin, rosuvastatin or simvastatin) was comparedwith no statin (placebo or standard care). We cat-egorized statins according to their pharmaco-logic effect on lowering cholesterol as either lowpotency ( uvastatin, lovastatin, pravastatin andsimva statin) or high potency (atorvastatin androsuva statin). 8 To be eligible, studies also had toreport one or more of the following outcomes:

all-cause mortality, unstable angina, acute myo -cardial infarction (fatal or nonfatal), stroke ortransient ischemic attack (fatal or nonfatal), sur-gical or percutaneous revascularization, length of stay, quality of life, persistence on statin therapy,and adverse events. Trials with fewer than 30participants per study arm were excluded toimprove the ef ciency of the work without anappreciable loss of power. 9

Trials were considered to have enrolled par-ticipants at low cardiovascular risk if the 10-yearrisk of cardiovascular-related death or nonfatalmyocardial infarction among participants was

less than 20%, 10 as assessed by extrapolation of observed risk in the control group of each trial.In general, this corresponded to participants whowere free from cardiovascular disease (i.e., noprior acute coronary syndrome or coronary re -vascularization, no prior ischemic stroke and noprior revascularization or loss of limb owing to

peripheral arterial disease) and diabetes. Datafrom trials in which some, but not all, partici-pants had known cardiovascular disease wereincluded if the control group had a low cardio-vascular risk (as de ned above).

In sensitivity analyses, we calculated the esti-mated 10-year risk of cardiovascular-related deathor nonfatal myocardial infarction for the averageparticipant in each trial, using mean baseline char-acteristics from the control group and two com-monly used formulas from DAgostino and col-leagues 11 and the Third Adult Treatment Panel. 12

We also used a number of other de nitions for

low risk in sensitivity analyses.Two reviewers screened each citation. Trials

considered to be relevant by one or both review-ers were retrieved, and the full text was indepen-dently assessed by two reviewers for inclusion.Disagreements were resolved by a third partythrough consensus.

Two reviewers independently assessed eachstudys risk of bias using a condensed version of the Chalmers Index 13 as well as other characteris-tics that in uence the risk of biased estimates of effectiveness in meta-analyses. 1416 Disagreementswere resolved by a third party through consensus.

Data extractionOne reviewer extracted data from selected trials,and a second reviewer checked the data for accu-racy. Results of intention-to-treat analyses werecollected if reported.

We classi ed adverse events as serious if theywere de ned as such by the primary authors or if their severity was unspeci ed but they led towithdrawal from therapy or study. We also col-lected data on the incidence of new diabetes, can-cer and rhabdomyolysis. Myocardial infarctions

were classi ed as fatal, nonfatal or fatal/ nonfatal;those in the fatal/ nonfatal category were fromstudies that did not specify the type of myocardialinfarction or separate totals for fatal and nonfatalmyocardial infarctions. Strokes were classi ed ina similar manner. Angina was classi ed as un -stable if de ned as such by the primary authors, if it required admission to hospital or if it necessi-tated revascularization. Angina that did not meetthese criteria was classi ed as unspeci edangina. Types of revascularization in cluded coro-nary artery by pass graft surgery, percutaneouscoronary intervention, percutaneous transluminal

Research

E1190 CMAJ, November 8, 2011, 183(16)


3/14

coronary angioplasty, interventional vascular pro-cedure, coronary revascularization, arterial revas-cularization and coronary angioplasty. For studiesin which the number of cardiovascular-relateddeaths or nonfatal myocardial infarctions wasunclear, we contacted the original authors forinformation so that we could calculate the 10-

year risk.

Data synthesis and analysisWe pooled the results using the random-effectsmodel described by DerSimonian and Laird. 17

We used relative risks (RRs) to summarize di -chotomous results. We quanti ed statistical het-erogeneity using the I 2 statistic and used small,moderate and large to describe values of 25%, 50% and 75%. 18,19 A weighted regressiontest 20 was used to test for publication bias.

We used univariable and bivariable meta -regression analyses 18 to examine whether the fol-

lowing variables in uenced the association be -tween statin use and all-cause mortality: durationof study, year of study, baseline cholesterol lev-els, observed annual cardiovascular risk in thecontrol arm, the absolute and percent change inlow-density lipoprotein cholesterol from baselinein the statin arm, the point at which low-densitylipoprotein cholesterol was measured, daily ini-tial dose of statin, potency of statin (high v. low),mean age of participants, proportion of partici-pants who were male, proportion who had dia-betes, proportion who had hypertension, andrisk-of-bias items.

The number needed to treat and the absoluterisk reduction were calculated for outcomes withstatistically signi cant relative risks. The numberneeded to treat indicates the number of patientswho need to be given statin treatment to preventone event and is the reciprocal of the absoluterisk reduction (difference in probabilities of anevent between treatment and control groups). 21

In subgroup analyses, we examined all-causemortality and other clinical outcomes for trialsstrati ed by statin potency (high v. low). We alsoexplored the relative effects of statins using a

method of indirect comparison of treatments de -scribed by Bucher and colleagues, 22 a techniquethat facilitates the comparison of any two statinsnot directly compared in any one study. Here, themagnitude of treatment effects (e.g., relativerisk) for the direct evidence of treatment A ver-sus B and treatment B versus C were comparedwith acquired indirect evidence of treatment Aversus C.

Analyses were conducted using RevMan Ver-sion 4.2 (The Cochrane Collaboration, 2002) andStata/MP 11.2 software (Stata Corp, 2011).

Results

Literature searchOf the 15 250 articles identi ed, 29 ( n = 80 711)met our inclusion criteria (Figure 1). 4,2350 Atorva -statin was the study drug in six trials ( n = 11 894),

uvastatin in four trials ( n = 2107), lovastatin in

three ( n = 15 769), pravastatin in nine ( n =30 974), rosuva statin in four ( n = 19 129) andsimvastatin in three ( n = 838). Twenty-two of thetrials were placebo-controlled, and the remainingseven reported usual care, conventional treatment,standard treatment, no treatment or diet as thecomparators. The median year of publication was2004 (range 19912010). The median duration of follow-up was two years (range 0.55.3 years).Details of the studies are summarized in Table 1(at the end of the article).

Risk-of-bias assessmentThe 29 trials generally exhibited moderate risk of bias (Table 2, at the end of the article). Aweighted regression test 20 using all-cause mortal-ity results detected no statistical evidence of pub-lication bias (bias = 0.03, p = 0.92).

Characteristics of participantsThe median age of the participants was 58 years(range 5176), and the proportion who were

Research

CMAJ, November 8, 2011, 183(16) E1191

Excluded n = 14 426 Not relevant n = 14 388 Not retrievable n = 6 No translator n = 32

Excluded n = 795 Not original research n = 292 Not low-risk population n = 187 No relevant outcomes n = 142

Sample size < 30n = 43

No placebo or usual-carearm n = 32

Follow-up < 6 mo n = 32 Multiple publication n = 30 No statin arm n = 24 Not an RCT n = 7 No useable data n = 5 Not human n = 1

Trials included in meta-analysisn = 29

Articles retrieved for detailed evaluationn = 824

Potentially relevant records identified

and screened for retrievaln = 15 250

Figure 1: Selection of studies for inclusion in the meta-analysis. RCT = random-ized controlled trial.


4/14

male ranged from 0% to 100% (median 62%).The median proportion of participants who haddiabetes was 7% (range 0%35%), and hyper-tension 47% (range 16%100%) (Table 1).

The baseline lipid levels in the in cluded trialswere: total cholesterol, median 6.0 (range 4.87.6) mmol/L; low-density lipoprotein cholesterol,

median 4.0 (range 2.85.2) mmol/L; high-densitylipoprotein cholesterol, median 1.3 (range 1.01.9) mmol/L; and triglycerides, median 1.7(range 1.23.6) mmol/L. Seven studies reportedbaseline C-reactive protein levels: median 0.17(range 0.157.7) mg/dL (16.6 [range 13.8732.4]nmol/L). The mean 10-year risk of cardiovascu-lar-related death or nonfatal myocardial infarctionwas 6% (range 0%18%).

OutcomesTable 3 presents a summary of the relative risks(overall and strati ed by high- and low-potency

statins), number needed to treat and absolute risk reduction for all outcomes.

All-cause mortalityTwenty-three trials ( n = 79 495) reported all-causemortality (Figure 2). The trial-level relative risk could not be estimated for four trials that reportedno events in either group. 51 From the remaining 19trials ( n = 78 321), the pooled relative risk of death was signi cantly lower among statin recipi-ents than among controls (RR 0.90, 95% con -dence interval [CI] 0.840.97; I 2 = 2.0%).

In the metaregression analyses, none of thecharacteristics listed in the methods section(including baseline cholesterol levels or changein lipid levels during the trials) signi cantlymodi ed the association between statin use andall-cause mortality at the level of p < 0.05(Appendix 2, available at www.cmaj.ca/lookup

/suppl/doi:10.1503/cmaj.101280/-/DC1).

Myocardial infarctionThirteen trials ( n = 48 023) reported the numberof participants with fatal, nonfatal or unspeci edmyocardial infarctions. The pooled relative risk was signi cantly lower among statin recipientsthan among controls (RR 0.63, 95% CI 0.500.79; I 2 = 13%].

From eight trials ( n = 31 424) that provideddata on fatal myocardial infarctions, we foundthat the pooled relative risk did not differ signi -cantly between treatment groups (RR 0.96, 95%

CI 0.51.85; I 2 = 0%).Ten trials ( n = 49 222) reported nonfatal

myocardial infarctions. The pooled relative risk was signi cantly lower among statin recipientsthan among controls (RR 0.64, 95% CI 0.490.84; I 2 = 44%) (Figure 3). After removingstudies that enrolled people with prior myocar-dial infarction, we found that the point estimatewas similar to the estimate from the primaryanalysis (eight trials, n = 48 595; RR 0.64,95% CI 0.480.86; moderate heterogeneity[ I 2 = 54%]).

Research

E1192 CMAJ, November 8, 2011, 183(16)

Table 3: Relative risk of serious outcomes associated with the use of statins in patients at low cardiovascular risk*

OutcomeOverall

RR (95% CI)

RR for high-potency statins

(95% CI)

RR for low-potency statins

(95% CI)

Numberneeded to treat

(95% CI)

Absolute riskreduction, %

(95% CI)

Death from any cause 0.90 (0.840.97) 0.85 (0.740.96) 0.90 (0.791.03) 239 (149796) 0.42 (0.130.67)

Myocardial infarction 0.63 (0.500.79) 0.47 (0.310.71) 0.68 (0.530.87) 216 (160381) 0.46 (0.260.63)

Fatal myocardial infarction 0.96 (0.501.85) 1.54 (0.613.89) 0.59 (0.231.51)

Nonfatal myocardial infarction 0.64 (0.490.84) 0.47 (0.340.67) 0.77 (0.591.00) 153 (108343) 0.66 (0.290.93)

Stroke 0.83 (0.740.93) 0.70 (0.550.87) 0.89 (0.771.03) 291 (190707) 0.34 (0.140.53)

Fatal stroke 0.91 (0.651.29) 0.50 (0.132.0) 0.95 (0.671.35)

Nonfatal stroke 0.81 (0.680.96) 0.51 (0.330.79) 0.88 (0.731.06) 335 (1991592) 0.30 (0.060.50)Unstable angina 0.71 (0.550.92) 0.73 (0.481.11) 0.70 (0.510.97) 431 (2781563) 0.23 (0.060.36)

Angina, type unspecified 0.83 (0.571.22) 1.05 (0.129.23) 0.83 (0.561.22)

Revascularization 0.66 (0.570.77) 0.74 (0.391.40) 0.66 (0.550.78) 131 (103193) 0.77 (0.520.97)

Serious adverse event 1.01 (0.961.07) 0.96 (0.861.09) 1.03 (0.981.10)

Rhabdomyolysis 1.29 (0.256.68) 2.99 (0.3128.75) 0.50 (0.055.51)

Cancer 1.00 (0.931.08) 0.95 (0.811.11) 1.02 (0.931.11)

New diabetes 1.05 (0.841.32) 1.17 (1.041.32) 0.76 (0.561.02)

Note: CI = confidence interval, RR = relative risk.*Low cardiovascular risk was defined as an observed 10-year risk of l ess than 20% for cardiovascular-related death or nonfatal myocardial infarction.Number needed to treat and absolute risk reduction for each outcome were calculated on the basis of the pooled risk in the control arm from all included trials.Number needed to treat and absolute risk reduction were calculated only for outcomes with a statistically significant estimate of effect.


5/14

StrokeFourteen trials ( n = 60 841) reported the num-ber of participants with fatal, nonfatal or unde-

ned strokes. The pooled relative risk of strokewas signi cantly lower among statin recipientsthan among controls (RR 0.83, 95% CI 0.740.93; I 2 = 0%).

Five trials ( n = 36 118) reported on the risk of fatal stroke; the pooled relative risk did not differsigni cantly be tween treatment groups (RR 0.91,95% CI 0.651.29; I 2 = 0%). Nine trials ( n =37 333) reported the number of participants withnonfatal stroke; the pooled relative risk was sig-ni cantly lower among statin recipients (relativerisk 0.81 [0.680.96]; I 2 = 0%) (Figure 4).

Coronary revascularizationEight trials ( n = 43 708) reported the number of participants who underwent percutaneous or sur-

gical coronary revascularization. The incidenceof revascularization was signi cantly loweramong statin recipients than among controls (RR0.66, 95% CI 0.570.77; I 2 = 7%).

Other outcomesFour trials ( n = 35 017) reported on the risk of

unstable angina: the pooled relative risk was sig-ni cantly lower among statin recipients thanamong controls (RR 0.71, 95% CI 0.550.92;

I 2 = 0%). Three trials ( n = 9082) re ported on therisk of un speci ed angina: the pooled relativerisk did not differ signi cantly between treat-ment groups (RR 0.83, 95% CI 0.571.22; I 2 =0%). Thirteen trials reported on the proportion of participants who adhered to statin therapy (range63%98%). Length of hospital stay, persistenceon statin therapy and quality-of-life measureswere not reported in any of the studies.

Research

CMAJ, November 8, 2011, 183(16) E1193

0.1 0.2 0.5 1 2 5 10

Favoursstatin

Favours nostatin

High-potency statin

RR (95% CI)

Statin No statin RR (95% CI)Study

Low-potency statin*

No. of events, n / N

33/6 5821/4603/224

106/3 3021/151

80/3 30487/2 219

631/5 1701/6070/2061/254

13/4330/414/283

55/3 8660/87

186/5 1685/700/4851/702

198/8 901

9/3144/58

403/15 698

1 419/42 887

3/1 6638/4594/223

135/3 2930/154

77/3 30178/1 634

641/5 1851/6220/1020/254

12/4310/415/285

79/3 9660/93

212/5 1375/730/1190/282

247/8 901

6/3255/65

475/14 902

1 518/36 608

1 016/27 189 1 043/21 706

EXCEL45ACAPS38KAPS41WOSCOPS43CAIUS42AFCAPS/TexCAPS39KLIS37ALLHATLLT40Bruckert et al. 33Muldoon et al. 29

PHYLLIS30PREVEND-IT31Rejnmark et al. 32HYRIM28MEGA24

ASCOTLLA36Holmberg et al. 26Bone et al. 23METEOR34JUPITER4

LEADe46LORD48

OverallI = 2.0%

SubtotalI = 0%

SubtotalI = 18.8%

ESPLANADE49

2.78 (0.859.05)0.12 (0.020.99)0.75 (0.173.30)0.78 (0.611.01)3.06 (0.1374.51)1.04 (0.761.41)0.82 (0.611.11)0.99 (0.891.09)1.02 (0.0616.35)

Not estimable3.00 (0.1273.30)1.08 (0.502.34)

Not estimable0.81 (0.222.97)0.71 (0.511.00)

Not estimable

0.90 (0.791.03)

0.87 (0.721.06)1.04 (0.323.45)

Not estimable1.21 (0.0529.56)0.80 (0.670.96)

1.55 (0.564.31)0.90 (0.253.18)

0.85 (0.740.96)

0.90 (0.840.97)

Figure 2: Risk of death from any cause associated with the use of statins (versus no statins) in pa tients atlow cardiovascular risk (observed 10-year risk of cardiovascular-related death or nonfatal myocardial infarc-tion < 20%). A relative risk (RR) of less than 1.0 indicates fewer deaths with the use of statins. CI = con -dence interval. *Lovastatin, uvastatin, pravastatin and simvastatin. Rosuva statin and atorvastatin. Forcomplete study names, see Box 1.


6/14

Research

E1194 CMAJ, November 8, 2011, 183(16)


Low-potency statin*ACAPS38 )24.392.0(00.1064 / 5KAPS41 )79.131.0(05.0422 / 3WOSCOPS43 )68.075.0(07.02033 / 341

CAIUS42 )65.550.0(15.0151 / 1KLIS37 )82.277.0(33.19122 / 63Muldoon et al. 29 elbamitsetoN602 / 0Rejnmark et al. 32 elbamitsetoN14 / 0MEGA26 )00.103.0(55.06683 / 61ESPLANADE49 )36.7731.0(02.378 / 1

)00.195.0(77.065501 / 502latotbuSI = 15.8%

High-potency statinASCOTLLA36 )67.004.0(55.08615 / 06METEOR34 )65.9250.0(12.1207 / 1JUPITER4 )85.022.0(53.01098 / 22

)76.043.0(74.017741 / 38latotbuSI

= 22.1%)48.094.0(46.072352 / 882llarevO

I = 44.3%


5/459

267/9 965

2/15420/1 634

0/1020/41

30/3 9660/93

204/3 2936/223

RR (95% CI)0.1 0.2 0.5 1 2 5 10

108/5 1370/282

62/8 901

170/14 320

437/24 285

Favoursstatin

Favours nostatin

Figure 3: Risk of nonfatal myocardial infarction associated with the use of statins (versus no statins) inpatients at low cardiovascular risk (observed 10-year risk of cardiovascular-related death or nonfatalmyocardial infarction < 20%). A relative risk (RR) of less than 1.0 indicates fewer events with the use ofstatins. CI = con dence interval. *Lovastatin, uvastatin, pravastatin and simvastatin. Rosuvastatin andatorvastatin. For complete study names, see Box 1.


Low-potency statin*ACAPS38 )57.210.0(41.0064 / 0KAPS41 )39.311.0(66.0422 / 2WOSCOPS43 )92.165.0(58.02033 / 04ALLHATLLT40 )11.127.0(98.00715 / 651Muldoon et al. 29 1/206ESPLANADE49 )36.7731.0(02.378 / 1

)60.137.0(88.09449 / 002latotbuSI = 0%

High-potency statinBone et al. 23 )70.8130.0(47.0584 / 1

0.22 (0.014.57)JUPITER4

)08.033.0(25.01098 / 03

)97.033.0(15.04449 / 13latotbuSI = 0%

)69.086.0(18.039881 / 132llarevOI = 0%


3/459

228/9 355

175/5 1850/1020/93

47/3 2933/223

RR (95% CI)0.1 0.2 0.5 1 2 5 10

0/119

58/8 901

60/9 085

288/18 440

Favoursstatin Favours nostatin

1.49 (0.0636.32)

LORD48 0/58 2/65

Figure 4: Risk of nonfatal stroke associated with the use of statins (versus no statins) in patients at low car-diovascular risk (observed 10-year risk of cardiovascular-related death or nonfatal myocardial infarction< 20%). A relative risk (RR) of less than 1.0 indicates fewer events with the use of statins. CI = con denceinterval. *Lovastatin, uvastatin, pravastatin and simvastatin. Rosuvastatin and atorvastatin. For com-plete study names, see Box 1.


7/14

Adverse eventsTwenty-one trials ( n = 47 589) reported the num-ber of participants who had serious adverseevents. The trial-level relative risk could not beestimated for four trials that reported no events ineither group. From the remaining 17 trials ( n =47 021), the pooled risk of serious ad verse events

did not differ signi cantly be tween treatmentgroups (RR 1.01, 95% CI 0.961.07; I 2 = 8%).Ten trials ( n = 45 557) reported the number of

participants who experienced rhabdomyolysis.From the three trials in which this outcomeoccurred ( n = 34 712), the pooled risk of rhab-domyolysis did not differ signi cantly betweengroups (RR 1.29, 95% CI 0.256.68; I 2 = 0%).

Ten trials ( n = 62 547) reported the number of participants in whom cancer was diagnosed, andfour trials ( n = 31 818) reported the proportionwith new diabetes. The pooled risk of cancer didnot differ signi cantly between groups (RR 1.00,

95% CI 0.931.08; I 2 = 0%), nor did the pooledrisk of diabetes (RR 1.05, 95% CI 0.841.32),although heterogeneity between studies in thelatter analysis was large ( I 2 = 64%).

Subgroup and sensitivity analysesIndirect comparisons between high- and low-potency statins did not show a signi cant differ-ence in the risk of all-cause mortality (RR forhigh- v. low-potency statins: 0.94, 95% CI 0.791.13), fatal or nonfatal myocardial infarction(RR 0.69, 95% CI 0.431.12) or stroke (RR0.79, 95% CI 0.611.02).

The sensitivity analyses, many of which usedalternative de nitions of low cardiovascular risk,showed ndings that were consistent with theresults of the primary analysis across a wide varietyof assumptions and conditions (Appendices 3 and4, available at www .cmaj .ca /lookup /suppl/doi:10.1503 /cmaj .101280 / -/DC1). In particular, use of estimates rather than observed data to classify trialswith respect to 10-year cardiovascular risk led tosimilar conclusions regarding the ef cacy of statinsin re ducing all-cause mortality. In addition, whenlow cardiovascular risk was de ned as a 10-year

risk of cardiovascular-related death or nonfatalmyocardial infarction of less than 10%, the pooledrelative risk was similar to that of the primaryanalysis (RR 0.83, 95% CI 0.730.94). The pooledrelative risk of all-cause mortality among statinrecipients was statistically signi cant in virtually allof the sensitivity analyses (28 of 33). In addition,the point estimate for the pooled relative risk wasrelatively stable in these analyses (range 0.780.93,as compared with the point estimate of 0.90 in theprimary analysis), except for one analysis thatincluded trials with a follow-up period shorter thanthe median of two years (pooled RR 0.99).

InterpretationWe found that the use of statins by people at lowcardiovascular risk reduced the relative risk of death from any cause by 10%. Treatment withstatins also reduced the risk of stroke, myocardialinfarction, unstable angina and coronary revascu-larization to an extent similar to that seen in trials

involving patients with coronary artery disease.52

In a meta-analysis of the ef cacy and safetyof statins among patients with coronary arterydisease, the number needed to treat was 86 toprevent a single death from any cause and 62 toprevent a single nonfatal myocardial infarction. 52

The corresponding numbers needed to treatamong people at low cardiovascular risk by ourprimary de nition were 239 and 153, whichre ect the generally low rates of vascular eventsin this population. Because statins might be usedinde nitely to reduce cardiovascular risk, theabsolute bene t may in crease (accompanied by

corresponding decreases in the number neededto treat) with longer durations of treatment, al -though this remains speculative.

Although we sought information from all tri-als on outcomes associated with serious harmfrom statins, information on serious adverseevents was available for only 59% (47 589/ 80 711) of the trial participants. This low propor-tion is consistent with previously documentedde ciencies in re porting harm among participantsin randomized controlled trials. 53,54 Our experi-ence with soliciting this information from investi-gators revealed a lack of clarity among trialauthors regarding how serious adverse eventswere re ported in published reports. Be cause therisk of serious morbidity would be expected todecline among statin users, our nding of a lack of signi cant difference in the rate of seriousadverse events between treatment groups and ournonsigni cant estimate of elevated risk amongstatin users should be interpreted with caution. Acomplete analysis of published and unpublisheddata from individual patients on the effect of statins among low-risk patients would be a usefuladdition to the literature, but it would require

cooperation among the various stakeholders.Our metaregression and indirect comparisons

did not show statistically signi cant differencesin the ef cacy of high- and low-potency statins.However, metaregression has relatively low sta-tistical power. The as-yet unproven potential forhigh-potency statins to prevent cardiovascularoutcomes more effectively must be balancedagainst their higher costs compared with low-potency statins. The recent availability of generic(lower cost) atorvastatin makes this issue of par-ticular importance for decision-makers and third-party payers.

Research

CMAJ, November 8, 2011, 183(16) E1195


8/14

Research

Unlike a recent systematic review that in -cluded statin trials involving both low- and high-risk people, 55 we did not nd a signi cantly in -creased risk of new diabetes among low-risk statin users, perhaps because of differences instudy populations or statistical power. If such aneffect exists, this would be expected to further

reduce the absolute bene t of statin treatment inlow-risk populations over the long term.Our ndings also differ from those of another

recently published systematic review of 11 trialsinvolving 65 229 people without cardiovasculardisease that found no evidence of a bene t of statin use on all-cause mortality. 56 The authorsreported a relative risk reduction for all-causemortality that was similar to ours (9% v. 10%),but their estimate was less precise, leading to a95% CI that crossed unity (0.831.01). Eight tri-als in that review were included in ourstudy. 4,24,28,31,36,39,40,43 However, we included an addi-

tional 11 trials that reported on all-cause mortal-ity, and we excluded 2 trials 57,58 that exclusivelyenrolled participants with a diagnosis of diabetesat baseline. A third study included in the priorreviews was excluded from our study becausethe 10-year risk for cardiovascular-related deathor nonfatal myocardial infarction was greaterthan 20%. 59

Strengths and limitationsThis is an up-to-date, comprehensive systematicreview of the clinical implications of statin useamong people at low cardiovascular risk. We in -cluded studies in which participants were consid-ered to be at low risk according to an acceptedde nition a 10-year incidence of less than 20%for cardiovascular-related death or nonfatalmyocardial infarction. 10 We excluded trials thatpredominantly enrolled people at higher cardio-vascular risk, such as those with diabetes or priorcardiovascular events. Therefore, our relative risk estimates were unlikely to be in uenced byhigher-risk subgroups within the trial populations.

Our analysis has limitations. First, as with allmeta-analyses, our conclusions are limited by

the availability of individual trials. We cannotexclude the possibility of publication bias, al -though it seems unlikely that there are unidenti-

ed trials large enough to in uence the directionor signi cance of our ndings.

Second, although we used accepted tech-niques for metaregression in an attempt to iden-tify factors associated with greater or lesser ben-e t from statin use among low-risk participants,statistical power for these analyses was relativelylow given the number of available trials.

Third, the duration of all trials was relativelyshort given that people at low cardiovascular risk

might require treatment with statins for decades.Fourth, we used the number needed to treat to

summarize the bene ts of statins across differentanalyses. Whether the number needed to treat isappropriate for use in meta-analyses is contro-versial; however, we believe that the simplicityand transparency of this method outweigh its

potential disadvantages.21,60

Fifth, although we reported a wide range of clinically relevant outcomes as speci ed in ourprotocol, the de nition for some (e.g., unstableangina) will have varied across trials.

Sixth, in an attempt at unbiased identi cationof people at low cardiovascular risk, we based ourinclusion criterion on the observed incidence of events among placebo recipients, anticipating thatclinicians could apply our ndings in practice byusing their risk prediction instrument of choice toidentify patients with an estimated 10-year risk of less than 20% for cardiovascular-related death or

nonfatal myocardial infarction. Because risk pre-diction tools may overestimate true rates of eventsin contemporary practice, we re peated analysesafter including only trials with an estimated 10-year risk of less than 20% based on mean partici-pant characteristics and two widely used risk equations; 11,12 we reached similar conclusions.

Finally, given that most of the trials includedin our review were at moderate risk of bias, wecannot exclude the possibility that other factorsbesides statin use may have in uenced the ob -served differences in health outcomes betweentreatment groups.

ConclusionBoth low- and high-potency statins were ef ca-cious in preventing death and cardiovascular-related morbidity in people at low risk of car -diovascular events (whose 10-year risk of cardiovascular-related death or nonfatal myocar-dial infarction is less than 20%), most of whomdid not have a history of coronary artery diseaseor diabetes. However, the number needed to treatto prevent one adverse outcome was relativelyhigh for any statin. Whether high-potency statins

improve outcomes to a greater extent than low-potency statins is uncertain based on current data.

References1. Cholesterol Treatment Trialists (CTT) Collaborators, Kearney

PM, Blackwell L, et al. Ef cacy of cholesterol-lowering therapyin 18,686 people with diabetes in 14 randomised trials of statins:a meta-analysis. Lancet 2008;371:117-25.

2. National Cholesterol Education Program (NCEP) Expert Panel. Detection, evaluation, and treatment of high blood cholesterol inadults (Adult Treatment Panel III) . Bethesda (MD): NationalInstitutes of Health; 2002.

3. Genest J, McPherson R, Frohlich J, et al. 2009 Canadian Cardio-vascular Society/Canadian guidelines for the diagnosis and treatment of dyslipidemia and prevention of cardiovascular disease inthe adult 2009 recommendations. Can J Cardiol 2009; 25:567-79.

E1196 CMAJ, November 8, 2011, 183(16)


9/14

4. Ridker PM, Danielson E, Fonseca FA, et al.; JUPITER StudyGroup. Rosuvastatin to prevent vascular events in men and wo menwith elevated C-reactive protein. N Engl J Med 2008; 359: 2195-207.

5. Canadian Agency for Drugs and Technologies in Health. Requirements for health technology assessment template for reports . Ottawa (ON): The Agency; 2008.

6. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statementfor reporting systematic reviews and meta-analyses of studiesthat evaluate health care interventions: explanation and elabora-tion. PLoS Med 2009;6:e1000100.

7. Ward S, Lloyd Jones M, Pandor A, et al. A systematic review

and economic evaluation of statins for the prevention of coro-nary events. Health Technol Assess 2007;11:1-160.

8. Pletcher MJ, Lazar L, Bibbins-Domingo K, et al. Comparingimpact and cost-effectiveness of primary prevention strategiesfor lipid-lowering. Ann Intern Med 2009;150:243-54.

9. Pearson ES. The application of statistical methods to industrialstandardization and quality control . London (UK): British Stan-dards Institution; 1960.

10. Grundy SM, Cleeman JI, Merz CN, et al. Implications of recentclinical trials for the National Cholesterol Education ProgramAdult Treatment Panel III Guidelines. J Am Coll Cardiol 2004;44:720-32.

11. DAgostino RB Sr, Vasan RS, Pencina MJ, et al. General cardio-vascular risk pro le for use in primary care: the FraminghamHeart Study. Circulation 2008;117:743-53.

12. Executive Summary of the Third Report of the National Choles-terol Education Program (NCEP) Expert Panel on Detection.Evaluation, and treatment of high blood cholesterol in adults

(Adult Treatment Panel III). JAMA 2001;285:2486-97.13. Juni P, Altman DG, Egger M. Assessing the quality of ran-domised controlled trials. In: Egger M, Davey Smith G, AltmanDG, editors. Systematic reviews in health care. Vol 2. London(UK): BMJ Books; 2001. p. 87-108.

14. Schulz KF, Chalmers I, Hayes RJ, et al. Empirical evidence of bias. JAMA 1995;273:408-12.

15. Jadad AR, Moore RA, Carrol D, et al. Assessing the quality of reports of randomized clinical trials: Is blinding necessary? ControlClin Trials 1996;17:1-12.

16. Cho MK, Bero LA. The quality of drug studies published insymposium proceedings. Ann Intern Med 1996;124:485-9.

17. DerSimonian R, Laird N. Meta-analysis in clinical trials. ControlClin Trials 1986;7:177-88.

18. Thompson SG, Higgins JP. How should meta-regression analy-ses be undertaken and interpreted? Stat Med 2002;21:1559-73.

19. Higgins JPT, Thompson SG, Deeks JJ, et al. Measuring incon-sistency in meta-analyses. BMJ 2003;327:557-60.

20. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analy-sis detected by a simple, graphical test. BMJ 1997;315: 629-34.

21. Schunemann HJ, Oxman AD, Vist GE, et al. Interpreting resultsand drawing conclusions. In: Higgins JPT, Green S, editors.Cochrane handbook for systematic reviews of interventions .Oxford (UK): The Cochrane Collaboration; 2011.

22. Bucher HC, Guyatt GH, Grif th LE, et al. The results of directand indirect treatment comparisons in meta-analysis of random-ized controlled trials. J Clin Epidemiol 1997;50:683-91.

23. Bone HG, Kiel DP, Lindsay RS, et al. Effects of atorvastatin onbone in postmenopausal women with dyslipidemia: a double-blind, placebo-controlled, dose-ranging trial. J Clin Endocrinol

Metab 2007;92:4671-7.24. Nakamura H, Arakawa K, Itakura H, et al. Primary prevention of

cardiovascular disease with pravastatin in Japan (MEGA Study):a prospective randomised controlled trial. Lancet 2006; 368:1155-63.

25. Di Lullo L, Addesse R, Comegna C, et al. Effects of uvastatintreatment on lipid pro le, C-reactive protein trend, and renalfunction in dyslipidemic patients with chronic renal failure. AdvTher 2005;22:601-12.

26. Holmberg B, Brannstrom M, Bucht B, et al. Safety and ef cacyof atorvastatin in patients with severe renal dysfunction. Scand J Urol Nephrol 2005;39:503-10.

27. Baigent C, Landray M, Leaper C, et al. First United KingdomHeart and Renal Protection (UK-HARP-I) study: biochemicalef cacy and safety of simvastatin and safety of low-dose aspirinin chronic kidney disease. Am J Kidney Dis 2005;45:473-84.

28. Anderssen SA, Hjelstuen AK, Hjermann I, et al. Fluvastatin andlifestyle modi cation for reduction of carotid intima-mediathickness and left ventricular mass progression in drug-treatedhypertensives [Hypertension High Risk Management trial(HYRIM)]. Atherosclerosis 2005;178:387-97.

29. Muldoon MF, Ryan CM, Sereika SM, et al. Randomized trial of the effects of simvastatin on cognitive functioning in hypercho-lesterolemic adults. Am J Med 2004;117:823-9.

30. Zanchetti A, Crepaldi G, Bond MG, et al. Different effects of anti-

hypertensive regimens based on fosinopril or hydrochlorothiazidewith or without lipid lowering by pravastatin on progression of asymptomatic carotid atherosclerosis: principal results of PHYLLIS a randomized double-blind trial. Stroke 2004;35: 2807-12.

31. Asselbergs FW, Diercks GFH, Hillege HL, et al.; Preventionof Renal and Vascular Endstage Disease Intervention Trial (PREVEND IT) Investigators. Effects of fosinopril and prava -statin on cardiovascular events in subjects with microalbumin-uria. Circulation 2004;110:2809-16.

32. Rejnmark L, Buus NH, Vestergaard P, et al. Effects of simvas-tatin on bone turnover and BMD: a 1-year randomized controlled

trial in postmenopausal osteopenic women. J Bone Miner Res2004;19:737-44.

33. Bruckert E, Lievre M, Giral P, et al. Short-term ef cacy andsafety of extended-release uvastatin in a large cohort of elderlypatients. Am J Geriatr Cardiol 2003;12:225-31.

34. Crouse JR III, Raichlen JS, Riley WA, et al. Effect of rosuva -statin on progression of carotid intima-media thickness in low-risk individuals with subclinical atherosclerosis: the METEORtrial. JAMA 2007;297:1344-53.

35. Dernellis J, Panaretou M. Effect of C-reactive protein reductionon paroxysmal atrial brillation. Am Heart J 2005;150:1064.

36. Sever PS, Dahlof B, Poulter NR, et al. Prevention of coronaryand stroke events with atorvastatin in hypertensive patients whohave average or lower-than-average cholesterol concentrations,in the Anglo-Scandinavian Cardiac Outcomes Trial LipidLowering Arm (ASCOTLLA): a multicentre randomised con-trolled trial. Lancet. 2003;361:1149-58.

37. Pravastatin use and risk of coronary events and cerebral infarc-

tion in japanese men with moderate hypercholesterolemia: theKyushu Lipid Intervention Study [KLIS]. J Atheroscler Thromb2000; 7:110-21.

38. Furberg CD, Adams HP Jr, Applegate WB, et al. Effect of lova -statin on early carotid atherosclerosis and cardiovascular events.Asymptomatic Carotid Artery Progression Study (ACAPS)Research Group. Circulation 1994;90:1679-87.

39. Downs JR, Clear eld M, Weis S, et al. Primary prevention of acute coronary events with lovastatin in men and women withaverage cholesterol levels: results of AFCAPS/TexCAPS. AirForce/Texas Coronary Atherosclerosis Prevention Study. JAMA1998;279:1615-22.

40. ALLHAT Of cers and Coordinators for the ALLHAT Collabo-rative Research Group. Major outcomes in moderately hyperc-holesterolemic, hypertensive patients randomized to pravastatinvs usual care: The Antihypertensive and Lipid-Lowering Treat-ment to Prevent Heart Attack Trial (ALLHAT LLT). JAMA2002;288:2998-3007.

41. Salonen R, Nyyssonen K, Porkkala E, et al. Kuopio Atherosclero-sis Prevention Study (KAPS). A population-based primary preven-tive trial of the effect of LDL lowering on atherosclerotic progres-sion in carotid and femoral arteries. Circulation 1995; 92:1758-64.

42. Mercuri M, Bond MG, Sirtori CR, et al. Pravastatin reducescarotid intima-media thickness progression in an asymptomatichypercholesterolemic mediterranean population: the CarotidAtherosclerosis Italian Ultrasound Study [CAIUS]. Am J Med 1996; 101: 627-34.

43. Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heartdisease with pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 1995;333:1301-7.

44. Bak AA, Huizer J, Leijten PA, et al. Diet and pravastatin in mod-erate hypercholesterolaemia: a randomized trial in 215 middle-aged men free from cardiovascular disease. J Intern Med 1998;244:371-8.

45. Bradford RH, Shear CL, Chremos AN, et al. Expanded ClinicalEvaluation of Lovastatin (EXCEL) study results. I. Ef cacy inmodifying plasma lipoproteins and adverse event pro le in 8245patients with moderate hypercholesterolemia. Arch Intern Med 1991 ;151:43-9.

46. Feldman HH, Doody RS, Kivipelto M, et al. Randomized con-trolled trial of atorvastatin in mild to moderate Alzheimer dis-ease: LEADe. Neurology 2010;74:956-64.

47. Chan KL, Teo K, Dumesnil JG, et al. Effect of lipid lowering withrosuvastatin on progression of aortic stenosis: results of the AorticStenosis Progression Observation: Measuring Effects of Rosuva -statin (ASTRONOMER) Trial. Circulation 2010;121:306-14.

48. Fassett RG, Robertson IK, Ball MJ, et al. Effect of atorvastatin onkidney function in chronic kidney disease: a randomised double-blind placebo-controlled trial [LORD study]. Atherosclerosis2010; 213:218-24.

49. Ruggenenti P, Perna A, Tonelli M, et al. Effects of add-on u-vastatin therapy in patients with chronic proteinuric nephropathyon dual reninangiotensin system blockade: the ESPLANADEtrial. Clin J Am Soc Nephrol 2010;5:1928-38.

Research

CMAJ, November 8, 2011, 183(16) E1197


10/14

50. Yun KH, Shin IS, Park EM, et al. Effect of additional statin ther-apy on endothelial function and prognosis in patients withvasospastic angina. Korean Circ J 2008;38:638-43.

51. Higgins JPT, Deeks JJ, Altman DG. Special topics in statistics.In: Higgins JPT, Green S, editors. Cochrane handbook for sys-tematic reviews of interventions . 5.1.0 ed. Oxford (UK): TheCochrane Collaboration; 2011.

52. Baigent C, Keech A, Kearney PM, et al. Ef cacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90 056 participants in 14 randomised trials of statins.

Lancet 2005;366:1267-78.

53. Chan AW, Hrobjartsson A, Haahr MT, et al. Empirical evidencefor selective reporting of outcomes in randomized trials: compar-ison of protocols to published articles. JAMA 2004;291:2457-65.

54. Ioannidis JP, Lau J. Completeness of safety reporting in randomizedtrials: an evaluation of 7 medical areas. JAMA 2001; 285: 437-43.

55. Sattar N, Preiss D, Murray HM, et al. Statins and risk of incidentdiabetes: a collaborative meta-analysis of randomised statin tri-als. Lancet 2010;375:735-42.

56. Ray KK, Seshasai SR, Erqou S, et al. Statins and all-cause mor-tality in high-risk primary prevention: a meta-analysis of 11 ran-domized controlled trials involving 65,229 participants. Arch

Intern Med 2010;170:1024-31.57. Colhoun HM, Betteridge DJ, Durrington PN, et al. Primary pre-

vention of cardiovascular disease with atorvastatin in type 2 dia-betes in the Collaborative Atorvastatin Diabetes Study (CARDS):multicentre randomised placebo-controlled trial. Lancet 2004;364:685-96.

58. Knopp RH, dEmden M, Smilde JG, et al. Ef cacy and safety of atorvastatin in the prevention of cardiovascular end points in sub-

jects with type 2 diabetes: the Atorvastatin Study for Preventionof Coronary Heart Disease Endpoints in non-insulin-dependentdiabetes mellitus (ASPEN). Diabetes Care 2006;29:1478-85.

59. Shepherd J, Blauw GJ, Murphy MB, et al. Pravastatin in elderlyindividuals at risk of vascular disease (PROSPER): a randomisedcontrolled trial. Lancet 2002;360:1623-30.

60. McAlister FA. The number needed to treat turns 20 andcontinues to be used and misused. CMAJ 2008;179:549-53.

Af liations: From the Department of Medicine (Tonelli,Lloyd, McAlister), University of Alberta, Edmonton, Alta.;the Department of Community Health Sciences (Clement,Hemmelgarn, Manns), University of Calgary, Calgary, Alta.;the Alberta Kidney Disease Network (Tonelli, Lloyd, Conly,

Hemmelgarn, Klarenbach, Wiebe, Manns), Calgary/Edmon-ton, Alta.; the Department of Epidemiology and CommunityMedicine (Husereau, Klarenbach, Wiebe), University of Ottawa, Ottawa, Ont.; and the Department of Medicine(Hemmelgarn, Manns), University of Calgary, Calgary, Alta.The complete list of members of the Alberta Kidney DiseaseNetwork is available at www.akdn.info.

Contributors: All the authors contributed to the conceptionand design. Marcello Tonelli, Fiona Clement, Braden Manns,

Anita Lloyd and Jonathan Conly contributed to the acquisi-tion of data and drafted the report. All of the authors con-tributed to the analysis and interpretation of data, criticallyrevised the report for important intellectual content andapproved the nal version submitted for publication.

Funding: This study was jointly funded by the CanadianAgency for Drugs and Technology in Health and the AlbertaHeritage Foundation for Medical Research InterdisciplinaryTeam Grants Program (which supports the InterdisciplinaryChronic Disease Collaboration). The study sponsors had norole in the design of the study, the collection, analysis orinterpretation of data, the writing of the report or the decisionto submit the article for publication.

Acknowledgements: The authors thank Dr. Steven Grover forhis helpful comments on an earlier draft of this manuscript.

Marcello Tonelli, Finlay McAlister and Braden Manns aresupported by Alberta Innovates Health Solutions (formerlythe Alberta Heritage Foundation for Medical Research(AHFMR) Health Scholar Awards. Brenda Hemmelgarn andScott Klarenbach were supported by Population Health Investi-gator Awards from Alberta Innovates Health Solutions. Marcello Tonelli was supported by a Government of CanadaResearch Chair in the optimal care of people with chronic kid-ney disease. Fiona Clement was supported by a postdoctoralfellowship award from the Canadian Health Services ResearchFoundation and AHFMR. Marcello Tonelli, Brenda Hemmel-garn, Scott Klarenbach, Finlay McAlister and Braden Mannswere supported by an alternative funding plan from the Govern-ment of Alberta and the Universities of Alberta and Calgary.

Research

E1198 CMAJ, November 8, 2011, 183(16)

Box 1: Full names of trials included in the meta-analysis

ACAPS: Asymptomatic Carotid Artery Progression Study AFCAPS/TexCAPS: Air Force/Texas Coronary Atherosclerosis Prevention Study ALLHATLLT: Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial ASCOTLLA: Anglo-Scandinavian Cardiac Outcomes Trial Lipid Lowering Arm ASTRONOMER: Aortic Stenosis Progression Observation: Measuring the Effects of Rosuva statin CAIUS: Carotid Atherosclerosis Italian Ultrasound Study ESPLANADE: European Study for Preventing by Lipid-lowering Agents aNd ACE-inhibition Dialysis

Endpoints EXCEL: Expanded Clinical Evaluation of Lovastatin Study HYRIM: Hypertension High-Risk Management Trial JUPITER: Justi cation for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuva -

statin KAPS: Kuopio Atherosclerosis Prevention Study KLIS: Kyushu Lipid Intervention Study LEADe: Lipitors Effect in Alzheimers Dementia LORD: Lipid Lowering and Onset of Renal Disease MEGA: Management of Elevated Cholesterol in the Primary Prevention Group of Adult Japanese METEOR: Measuring Effects on Intima-Media Thickness: an Evaluation of Rosuvastatin PHYLLIS: Plaque Hypertension Lipid-Lowering Italian Study PREVEND-IT: Prevention of Renal and Vascular Endstage Disease Intervention Trial UK-HARP-I: First United Kingdom Heart and Renal Protection WOSCOPS: West of Scotland Coronary Prevention Study


11/14


12/14


13/14


14/14

statins for primary prevention

Documents