32
Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis of 31 Randomized Controlled Trials James S. McKinney, MD; William J. Kostis, PhD, MD Background and Purpose—Statin therapy decreases the risk of ischemic stroke. An increased risk of intracerebral hemorrhage (ICH) has been observed in some studies. To investigate this issue, we performed a meta-analysis of randomized controlled trials using statins that reported ICH. Methods—We performed a literature search of Medline, Web of Science, and The Cochrane Library through January 25, 2012, and identified additional randomized controlled trials by reviewing reference lists of retrieved studies and prior meta-analyses. All randomized controlled trials of statin therapy that reported ICH or hemorrhagic stroke were included. The primary outcome variable was ICH. Thirty-one randomized controlled trials were included. All analyses used random effects models and heterogeneity was not observed in any of the analyses. Results—A total of 91 588 subjects were included in the active group and 91 215 in the control group. There was no significant difference in incidence of ICH observed in the active treatment group versus control (OR, 1.08; 95% CI, 0.88 –1.32; P0.47). ICH risk was not related to the degree of low-density lipoprotein reduction or achieved low-density lipoprotein cholesterol. Total stroke (OR, 0.84; 95% CI, 0.78 – 0.91; P0.0001) and all-cause mortality (OR, 0.92; CI, 0.87– 0.96; P0.0007) were significantly reduced in the active therapy group. There was no evidence of publication bias. Conclusions—Active statin therapy was not associated with significant increase in ICH in this meta-analysis of 31 randomized controlled trials of statin therapy. A significant reduction in all stroke and all-cause mortality was observed with statin therapy. (Stroke. 2012;43:2149-2156.) Key Words: hemorrhagic stroke intracerebral hemorrhage meta-analysis statin H ypercholesterolemia is associated with an increased risk of ischemic stroke. 1–5 Lipid-lowering therapy with 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) is effective in reducing cardiovascular mortality by preventing myocardial infarction and ischemic stroke. 6 –24 The clinical benefit may not be limited to the lipid-lowering properties of statins but also derived from other “pleiotropic” effects, including anti-inflammatory and antithrombotic ef- fects. 25 Despite the significant reductions in ischemic stroke observed in clinical trials, a large population cohort study and subsequent meta-analyses have found no reduction in stroke mortality with statin therapy. 26,27 The Stroke Prevention by Aggressive Reduction in Cho- lesterol Levels (SPARCL) study of high-dose atorvastatin in secondary stroke prevention demonstrated a significant over- all reduction in recurrent stroke but no difference in fatal stroke. 22 A post hoc analysis of the SPARCL trial found that treatment with atorvastatin was independently associated with an increased risk of hemorrhagic stroke (hazard ratio, 1.68; 95% CI, 1.09 –2.59). 28 This was particularly true of subjects enrolled with an index hemorrhagic stroke who had a 5-fold increase in risk of recurrent hemorrhage. 28 A 2009 Cochrane review of lipid-lowering therapy and stroke reported a significant increase in the odds of hemor- rhagic stroke with statin therapy (OR, 1.72; 95% CI, 1.20 – 2.46). 29 However, this analysis only included 2 trials, Heart Protection Study (HPS) and SPARCL. A subsequent meta- analysis performed by the Cholesterol Treatment Trialists’ (CTT) Collaboration, which included 20 clinical trials of statin therapy that reported intracerebral hemorrhage (ICH) occurrence, observed a nonsignificant trend toward increased risk of hemorrhagic stroke (relative risk, 1.15; 95% CI, 0.93–1.41; P0.2). 30 This analysis showed no effect on stroke mortality with statin therapy. 30 The CTT Collaboration study excluded trials enrolling 1000 subjects and those with a 2-year follow-up period. 30 The goals of the current study were to examine the risk of hemorrhagic stroke in patients treated with statins, to assess their effect on total stroke and Received February 29, 2012; accepted March 30, 2012. From the Cardiovascular Institute of New Jersey and the Department of Neurology (J.S.M.), University of Medicine and Dentistry of New Jersey–Robert Wood Johnson Medical School, New Brunswick, NJ; and the Cardiology Division (W.J.K.), Massachusetts General Hospital, Boston, MA. The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112. 655894/-/DC1. Correspondence to James S. McKinney, MD, Department of Neurology, UMDNJ-Robert Wood Johnson Medical School, 125 Paterson Street, New Brunswick, NJ 08901. E-mail [email protected] © 2012 American Heart Association, Inc. Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.112.655894 2149 by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from by guest on July 18, 2018 http://stroke.ahajournals.org/ Downloaded from

Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Embed Size (px)

Citation preview

Page 1: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Statin Therapy and the Risk of Intracerebral HemorrhageA Meta-Analysis of 31 Randomized Controlled Trials

James S. McKinney, MD; William J. Kostis, PhD, MD

Background and Purpose—Statin therapy decreases the risk of ischemic stroke. An increased risk of intracerebralhemorrhage (ICH) has been observed in some studies. To investigate this issue, we performed a meta-analysis ofrandomized controlled trials using statins that reported ICH.

Methods—We performed a literature search of Medline, Web of Science, and The Cochrane Library through January 25,2012, and identified additional randomized controlled trials by reviewing reference lists of retrieved studies and priormeta-analyses. All randomized controlled trials of statin therapy that reported ICH or hemorrhagic stroke were included.The primary outcome variable was ICH. Thirty-one randomized controlled trials were included. All analyses usedrandom effects models and heterogeneity was not observed in any of the analyses.

Results—A total of 91 588 subjects were included in the active group and 91 215 in the control group. There was nosignificant difference in incidence of ICH observed in the active treatment group versus control (OR, 1.08; 95% CI,0.88–1.32; P�0.47). ICH risk was not related to the degree of low-density lipoprotein reduction or achieved low-densitylipoprotein cholesterol. Total stroke (OR, 0.84; 95% CI, 0.78–0.91; P�0.0001) and all-cause mortality (OR, 0.92; CI,0.87–0.96; P�0.0007) were significantly reduced in the active therapy group. There was no evidence of publicationbias.

Conclusions—Active statin therapy was not associated with significant increase in ICH in this meta-analysis of 31randomized controlled trials of statin therapy. A significant reduction in all stroke and all-cause mortality was observedwith statin therapy. (Stroke. 2012;43:2149-2156.)

Key Words: hemorrhagic stroke � intracerebral hemorrhage � meta-analysis � statin

Hypercholesterolemia is associated with an increased riskof ischemic stroke.1–5 Lipid-lowering therapy with

3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors(statins) is effective in reducing cardiovascular mortality bypreventing myocardial infarction and ischemic stroke.6–24

The clinical benefit may not be limited to the lipid-loweringproperties of statins but also derived from other “pleiotropic”effects, including anti-inflammatory and antithrombotic ef-fects.25 Despite the significant reductions in ischemic strokeobserved in clinical trials, a large population cohort study andsubsequent meta-analyses have found no reduction in strokemortality with statin therapy.26,27

The Stroke Prevention by Aggressive Reduction in Cho-lesterol Levels (SPARCL) study of high-dose atorvastatin insecondary stroke prevention demonstrated a significant over-all reduction in recurrent stroke but no difference in fatalstroke.22 A post hoc analysis of the SPARCL trial found thattreatment with atorvastatin was independently associatedwith an increased risk of hemorrhagic stroke (hazard ratio,

1.68; 95% CI, 1.09–2.59).28 This was particularly true ofsubjects enrolled with an index hemorrhagic stroke who hada �5-fold increase in risk of recurrent hemorrhage.28

A 2009 Cochrane review of lipid-lowering therapy andstroke reported a significant increase in the odds of hemor-rhagic stroke with statin therapy (OR, 1.72; 95% CI, 1.20–2.46).29 However, this analysis only included 2 trials, HeartProtection Study (HPS) and SPARCL. A subsequent meta-analysis performed by the Cholesterol Treatment Trialists’(CTT) Collaboration, which included 20 clinical trials ofstatin therapy that reported intracerebral hemorrhage (ICH)occurrence, observed a nonsignificant trend toward increasedrisk of hemorrhagic stroke (relative risk, 1.15; 95% CI,0.93–1.41; P�0.2).30 This analysis showed no effect onstroke mortality with statin therapy.30 The CTT Collaborationstudy excluded trials enrolling �1000 subjects and those witha �2-year follow-up period.30 The goals of the current studywere to examine the risk of hemorrhagic stroke in patientstreated with statins, to assess their effect on total stroke and

Received February 29, 2012; accepted March 30, 2012.From the Cardiovascular Institute of New Jersey and the Department of Neurology (J.S.M.), University of Medicine and Dentistry of New

Jersey–Robert Wood Johnson Medical School, New Brunswick, NJ; and the Cardiology Division (W.J.K.), Massachusetts General Hospital, Boston, MA.The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112.

655894/-/DC1.Correspondence to James S. McKinney, MD, Department of Neurology, UMDNJ-Robert Wood Johnson Medical School, 125 Paterson Street, New

Brunswick, NJ 08901. E-mail [email protected]© 2012 American Heart Association, Inc.

Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.112.655894

2149

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 2: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

all-cause mortality, and determine whether the occurrence ofhemorrhagic stroke was related to the low-density lipoprotein(LDL) change achieved by therapy. We performed a meta-analysis including all available randomized controlled trialsof statin therapy that reported hemorrhagic stroke rate.

MethodsThe methods used in this meta-analysis are similar to those previ-ously reported.31 We performed a literature search through January25, 2012, using Medline, Web of Science, and the Cochrane Librarydatabases identifying randomized controlled trials (RCTs) of statintherapy and ICH or hemorrhagic stroke published in the Englishlanguage. Search criteria are presented in online-only Data Supple-ment Figure I. We identified additional RCTs by reviewing referencelists of identified studies and previously published meta-analyses ofstatin therapy. We included all trials with the following criteria:subjects aged �18 years, randomized controlled design, blindedoutcomes assessment, and recorded data on hemorrhagic stroke orICH. Both primary and secondary prevention trials were included aswere trials comparing active therapy with control therapy of “usualcare,” placebo, or lower-dose statins. Studies were included ifhemorrhagic stroke rates were reported in the first published report,subsequent publications, or meta-analyses. All studies were adjudi-cated by authors, and disagreements were resolved through discus-sion and consensus. Based on the search criteria, 31 RCTs of statintherapy that reported ICH or hemorrhagic stroke as an outcome wereincluded in the analysis (Figure 1).

Statistical Methods

Study Selection

Data Extraction and Quality AssessmentData on the statin used in the active treatment group, type of therapy(placebo, low-dose statin, or usual care) used in the control group,inclusion criteria, average follow-up, and the numbers of patients,total strokes, ICH, and deaths in the active and control groups were

retrieved by the authors. Data on randomization, allocation conceal-ment, comparison of baseline characteristics, defined eligibilitycriteria, type of control, blinding (patients, investigators, assessmentof vital status), percent lost to follow-up, and use of intention-to-treatanalysis were assessed.

Data Synthesis and AnalysisICH rate, total stroke rate, and all-cause mortality in the active andcontrol groups were calculated. Statistical analyses were performedusing Comprehensive Meta-Analysis 2.232 and JMP 9.0 (SASInstitute, Cary, NC). ORs and 95% CIs were calculated for the 3rates described using the intention-to-treat approach. Weightedpooled treatment effects were calculated for each of the 3 event ratesusing random-effects models. Heterogeneity of the effects wasevaluated by the use of the Q statistic.33 Publication bias wasexamined by constructing funnel plots,32 by Duval and Tweedie’sTrim and Fill,33 and the fail-safe N models of Rosenthal34 andOrwin.35

Sensitivity AnalysesSeveral additional sensitivity analyses were performed. First, 1 study(Collaborative Atorvastatin Diabetes Study [CARDS]) reported 0hemorrhagic strokes in both the atorvastatin and placebo groups.19,25

Therefore, to avoid inclusion of an undefined OR, CARDS was notconsidered in our primary analysis, which included only the 30 otherstudies. To examine the effects of excluding this study from ourprimary analysis, a sensitivity analysis was performed by imputing 1ICH (instead of 0) in each of the 2 groups (active and control) inCARDS and including this with the other 30 studies. A secondsensitivity analysis was performed by removing 1 study at a time (ofthe 30) in an iterative fashion. This was performed to evaluatewhether individual trials may have overly influenced the findings ofthis study. Third, a cumulative meta-analysis was performed toinvestigate how well the estimated OR of hemorrhagic strokeconverged with the iterative addition of progressively larger studies.

ResultsThirty-one trials of statin therapy that randomized a totalof 182 803 subjects were identified and included in the

Figure 1. Literature search profile.

2150 Stroke August 2012

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 3: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

analysis.6–24,36–47 Of the 31 RCTs, 6 studies11,17,18,20,24,42 comparedhigh-dose with low-dose statin therapy, and the remain-der6 –10,12–15,19,21–23,36 – 41,43– 47 compared statin therapy withplacebo or “usual care.” A total of 91 588 subjects were

randomized to active therapy and 91 215 subjects to controltherapy. The median length of follow-up was 46.8 months.

The patient and LDL characteristics for RCTs included inthis analysis are presented in online-only Data Supplement

Table. Cerebrovascular Events and Death

Trial, Year

Subjects

Enrolled

(N�182 803)

All

Stroke–

Active

All

Stroke–

Control

Ischemic

Stroke–

Active

Ischemic

Stroke–

Control

ICH–

Active

ICH–

Control

Other

Stroke–

Active

Other

Stroke–

Control

Fatal

Stroke–

Active

Fatal

Stroke–

Control

Total

Mortality–

Active

Total

Mortality–

Control

4S, 19946 4444 44 64 29 49 0 2 15 13 14 12 182 256

AF-TEXCAPS,

19988

6605 14 17 1 1 1 0 12 16 . . . . . . 80 77

ALLHAT-LLT,

200212

10 355 209 231 71 83 17 5 121 143 53 56 631 641

ASCOT-LLA,

200316

10 305 89 121 74 95 11 20 4 6 . . . . . . 185 212

A-to-Z,

200417

4497 28 35 22 31 6 0 0 4 . . . . . . 130 104

AURORA,

200936

2773 94 81 57 55 25 21 12 5 40 36 636 660

CARE, 19967 4159 54 78 48 64 2 6 4 8 5 1 180 196

CORONA,

200737

5011 126 145 73 90 15 9 15 16 35 39 728 759

GISSI-P,

200038

4271 20 19 15 13 1 0 4 6 4 4 72 88

GREACE,

200214

1600 9 17 . . . . . . 1 1 9 17 0 1 23 40

HPS, 200213 20 536 444 585 290 409 51 53 103 134 96 119 1328 1507

JUPITER,

200823

17 802 33 64 23 47 6 9 4 8 3 6 198 247

LIPID, 19989 9014 224 272 200 255 17 9 7 8 22 27 717 888

MEGA,

200621

7832 50 62 34 46 16 14 0 2 . . . . . . 55 79

PROSPER,

200215

5804 135 131 91 88 8 10 36 33 22 14 298 305

PROVE-IT,

200418

4162 21 19 10 12 4 1 7 6 . . . . . . 46 66

TNT, 200520 10 001 117 155 96 130 16 17 5 8 . . . . . . 284 282

SEARCH,

201024

12 064 255 279 233 255 24 25 0 0 57 67 964 970

SPARCL,

200622

4731 265 311 218 274 55 33 7 12 24 41 216 211

CARDS,

200419

2841 21 39 9 24 0 0 12 15 1 5 61 82

ASPEN,

200639

2410 34 38 14 15 4 2 16 21 . . . . . . 70 68

GISSI-HF,

200840

4574 82 66 63 53 11 3 8 10 38 29 657 644

4D, 200541 1252 59 44 47 33 5 8 10 6 27 13 297 320

IDEAL, 200542 8888 151 174 129 158 6 6 16 10 . . . . . . 366 374

MIRACL,

200110

3086 12 24 . . . . . . 0 3 12 21 3 2 64 68

PATE, 200111 665 11 18 11 15 0 3 0 0 2 2 14 20

ACAPS,

199443

919 0 5 . . . . . . 0 3 0 2 . . . . . . 1 8

ALERT,

200344

2102 93 91 67 66 10 17 5 5 11 10 143 138

BONE, 200745 604 1 0 . . . . . . 1 0 . . . . . . . . . . . . 0 0

CLAPT,

199946

226 0 1 . . . . . . 0 1 . . . . . . . . . . . . 0 2

SHARP,

201147

9270 171 210 114 157 45 37 18 19 68 78 1142 1115

All studies

total (%)

2866 (3.13) 3396 (3.72) 2039 (2.23) 2518 (2.76) 358 (0.39) 318 (0.35) 462 (0.50) 554 (0.61) 525 (0.57) 562 (0.62) 9768 (10.7) 10 427 (11.4)

ICH indicates intracerebral hemorrhage.

McKinney and Kostis Statin Therapy and Risk of ICH 2151

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 4: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Tables I and II. The mean age of was 62.6�5.2 years. A totalof 67.0% of patients were male and 78.1% were white. Only11.0% of subjects had a prior documented history of stroke.On average 24.7% had diabetes, 53.0% had hypertension,59.1% had cardiovascular diseases, and 21.2% activelysmoked cigarettes at study enrollment. A total of 61.0% ofsubjects were treated with aspirin or oral anticoagulants.Stroke events are presented in the Table.

Intracerebral HemorrhageICH occurred in 358 subjects (0.39%) in the active treatmentgroup versus 318 (0.35%) in the control group. In the primaryanalysis assessing ICH risk in 30 studies of statin treatment,active therapy was not associated with an increase in ICH(OR, 1.08; 95% CI, 0.88–1.32; P�0.47; Figure 2). The P forheterogeneity (interaction) was 0.38, Q�30.705, and degreesof freedom�29. No statistically significant difference in therisk of ICH was observed when the meta-analyses werestratified by control group or primary versus secondaryprevention studies (online-only Data Supplement Figure II).

The sensitivity analysis that included all 31 studies (includ-ing CARDS) produced nearly identical results as the primaryanalysis (OR, 1.08; 95% CI, 0.88–1.31; P�0.46). In all 30

iterations of the sensitivity analysis performed by removing 1study at a time, there was no association between ICH andactive treatment. The point estimate of the OR varied from1.04 to 1.11 with a lower limit of 0.84 to 0.91 and P of 0.27to 0.77. During the cumulative meta-analysis, there was noassociation between ICH and active treatment in 28 of 30iterations. The point estimate of the OR ranged from 0.20(95% CI, 0.01–4.17; P�0.30) to 1.90 (95% CI, 0.40–8.95;P�0.42).

We did not observe evidence of publication bias asdemonstrated by a symmetrical funnel plot. The OR and CI ofthe overall effect (OR, 1.08; 95% CI, 0.88–1.31) remainedunchanged using Duval and Tweedie’s Trim and Fillmethod.33 Orwin’s fail-safe N revealed that 25 additionalstudies with a mean OR of 1.0 would be needed to bring a 5%increase (OR, 1.05) at the P�0.05 level.

A metaregression was performed to study the relationshipbetween the LDL effect of statin therapy and ICH risk. Therewas no relationship between the effects of active versuscontrol therapy with measures of LDL cholesterol. Specifi-cally, there were no significant relationships of the log OR(active/control) of each study versus (1) the difference (activeminus control) of the LDL cholesterol drop (difference

Study name Subgroup within study Statistics for each study Stroke / Total Odds ratio and 95% CI

Odds Lower Upper ratio p-Value limit limit Active Control

4S PLACEBO 0.20 0.2989 0.01 4.17 0 / 2221 2 / 2223AF-TEXCAPS PLACEBO 3.00 0.5014 0.12 73.62 1 / 3304 0 / 3301ALLHAT-LLT PLACEBO 3.42 0.0158 1.26 9.27 17 / 5170 5 / 5185ASCOT-LLA PLACEBO 0.55 0.1071 0.26 1.14 11 / 5168 20 / 5137ATOZ LOW DOSE 12.84 0.0820 0.72 228.14 6 / 2265 0 / 2232AURORA PLACEBO 1.19 0.5608 0.66 2.14 25 / 1389 21 / 1384CARE PLACEBO 0.33 0.1774 0.07 1.65 2 / 2081 6 / 2078CORONA PLACEBO 1.66 0.2308 0.72 3.80 15 / 2514 9 / 2497GISSI-P PLACEBO 2.99 0.5019 0.12 73.55 1 / 2138 0 / 2133GREACE PLACEBO 1.00 1.0000 0.06 16.02 1 / 800 1 / 800HPS PLACEBO 0.96 0.8434 0.65 1.41 51 / 10269 53 / 10267JUPITER PLACEBO 0.67 0.4415 0.24 1.87 6 / 8901 9 / 8901LIPID PLACEBO 1.89 0.1236 0.84 4.24 17 / 4512 9 / 4502MEGA PLACEBO 1.17 0.6632 0.57 2.41 16 / 3866 14 / 3966PROSPER PLACEBO 0.81 0.6490 0.32 2.04 8 / 2891 10 / 2913PROVE-IT LOW DOSE 3.94 0.2205 0.44 35.25 4 / 2099 1 / 2063TNT LOW DOSE 0.94 0.8666 0.48 1.87 16 / 4995 17 / 5006SEARCH LOW DOSE 0.96 0.8871 0.55 1.68 24 / 6031 25 / 6033SPARCL PLACEBO 1.68 0.0191 1.09 2.60 55 / 2365 33 / 2366ASPEN PLACEBO 1.98 0.4296 0.36 10.85 4 / 1211 2 / 1199GISSI-HF PLACEBO 3.69 0.0454 1.03 13.23 11 / 2285 3 / 22894D PLACEBO 0.64 0.4299 0.21 1.96 5 / 619 8 / 633IDEAL LOW DOSE 1.00 0.9969 0.32 3.11 6 / 4439 6 / 4449MIRACL PLACEBO 0.14 0.1992 0.01 2.78 0 / 1538 3 / 1548PATE LOW DOSE 0.60 0.1963 0.28 1.30 11 / 331 18 / 334ACAPS PLACEBO 0.14 0.1965 0.01 2.75 0 / 460 3 / 459ALERT PLACEBO 0.59 0.1817 0.27 1.28 10 / 1050 17 / 1052BONE PLACEBO 0.74 0.8540 0.03 18.28 1 / 485 0 / 119CLAPT PLACEBO 0.34 0.5060 0.01 8.34 0 / 112 1 / 114SHARP PLACEBO 1.21 0.3916 0.78 1.87 45 / 4650 37 / 4620

1.08 0.4687 0.88 1.32

0.01 0.1 1 10 100

Favors Active Favors Control

ICH - Statin Meta-Analysis: ICH

All studies

Figure 2. Forrest plot of random-effects meta-analysis of randomized trials of statins and intracerebral hemorrhage.

2152 Stroke August 2012

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 5: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

between baseline and follow-up) in mg/dL (slope, 0.0043;SE, 0.0054; 95% CI, �0.4831 to 0.0149; P�0.43]; (2) thisdifference expressed as percentage of baseline (slope, 0.0054;SE, 0.0075; 95% CI, �0.0092 to 0.0200; P�0.47]; and (3)the achieved LDL cholesterol in the active treatment groupduring follow-up (slope, �0.0049; SE, 0.0043; 95% CI,�0.0133 to 0.0035; P�0.26).

All StrokeThere was a total of 6262 strokes in this meta-analysis. Theoverall stroke rate was 3.13% in the active group versus3.72% in the control group. Active therapy resulted in asignificant reduction in total stroke (OR, 0.84; 95% CI,0.78–0.91; P�0.0001; Figure 3). The P for heterogeneity(interaction) was 0.31, Q�33.315, and degrees of free-dom�30. The number needed to treat with active statintherapy to prevent any stroke during study follow-up was 200(absolute risk reduction�0.5%, P�0.0001).

All-Cause MortalityThere were 20 195 deaths recorded in the 31 trials included inthis analysis. There was a significantly lower rate of all-causemortality in the active group (10.67%) than in the control

group (11.43%; OR, 0.92; 95% CI, 0.87–0.96; P�0.0007;Figure 4). The P for heterogeneity (interaction) was 0.31,Q�32.272, and degrees of freedom�29. The number neededto treat with active statin therapy to prevent 1 death was 167(absolute risk reduction�0.6%, P�0.0001).

DiscussionEpidemiological studies have reported increased rates ofhemorrhagic stroke and ICH-related mortality in populationswith low cholesterol levels.1,4,48,49 It has been hypothesizedthat cholesterol may be important for cerebrovascular wallintegrity and that low levels may increase the risk of vesselrupture and ICH.50 Furthermore, several studies have reportedan association of hemorrhagic stroke with statin use.

The SPARCL trial of high-dose atorvastatin in secondarystroke prevention reported an excess of ICH with activetreatment compared with placebo (55 versus 33; P�0.02).22

Similarly, there was a 2-fold increase in hemorrhagic strokein the HPS among patients with prior stroke treated withsimvastatin.51 Three previous meta-analyses of statin therapyhave found no increased risk of hemorrhagic stroke.25,30,52 Ina meta-analysis of 11 trials, Amarenco and Labreuche25

reported a negligible risk of hemorrhagic stroke with statin

Study name Subgroup within study Statistics for each study Stroke / Total Odds ratio and 95% CI

Odds Lower Upper ratio p-Value limit limit Active Control

4S PLACEBO 0.68 0.0533 0.46 1.01 44 / 2221 64 / 2223AF-TEXCAPS PLACEBO 0.82 0.5880 0.40 1.67 14 / 3304 17 / 3301ALLHAT-LLT PLACEBO 0.90 0.2982 0.75 1.09 209 / 5170 231 / 5185ASCOT-LLA PLACEBO 0.73 0.0234 0.55 0.96 89 / 5168 121 / 5137ATOZ LOW DOSE 0.79 0.3448 0.48 1.30 28 / 2265 35 / 2232AURORA PLACEBO 1.17 0.3223 0.86 1.59 94 / 1389 81 / 1384CARE PLACEBO 0.68 0.0340 0.48 0.97 54 / 2081 78 / 2078CORONA PLACEBO 0.86 0.2138 0.67 1.09 126 / 2514 145 / 2497GISSI-P PLACEBO 1.05 0.8780 0.56 1.97 20 / 2138 19 / 2133GREACE PLACEBO 0.52 0.1197 0.23 1.18 9 / 800 17 / 800HPS PLACEBO 0.75 0.0000 0.66 0.85 444 / 10269 585 / 10267JUPITER PLACEBO 0.51 0.0019 0.34 0.78 33 / 8901 64 / 8901LIPID PLACEBO 0.81 0.0251 0.68 0.97 224 / 4512 272 / 4502MEGA PLACEBO 0.83 0.3151 0.57 1.20 50 / 3866 62 / 3966PROSPER PLACEBO 1.04 0.7533 0.81 1.33 135 / 2891 131 / 2913PROVE-IT LOW DOSE 1.09 0.7928 0.58 2.03 21 / 2099 19 / 2063TNT LOW DOSE 0.75 0.0209 0.59 0.96 117 / 4995 155 / 5006SEARCH LOW DOSE 0.91 0.2900 0.77 1.08 255 / 6031 279 / 6033SPARCL PLACEBO 0.83 0.0416 0.70 0.99 265 / 2365 311 / 2366CARDS PLACEBO 0.53 0.0184 0.31 0.90 21 / 1429 39 / 1412ASPEN PLACEBO 0.88 0.6022 0.55 1.41 34 / 1211 38 / 1199GISSI-HF PLACEBO 1.25 0.1786 0.90 1.74 82 / 2285 66 / 22894D PLACEBO 1.41 0.0979 0.94 2.12 59 / 619 44 / 633IDEAL LOW DOSE 0.87 0.2012 0.69 1.08 151 / 4439 174 / 4449MIRACL PLACEBO 0.50 0.0507 0.25 1.00 12 / 1538 24 / 1548PATE LOW DOSE 0.60 0.1963 0.28 1.30 11 / 331 18 / 334ACAPS PLACEBO 0.09 0.1030 0.00 1.63 0 / 460 5 / 459ALERT PLACEBO 1.03 0.8667 0.76 1.39 93 / 1050 91 / 1052BONE PLACEBO 0.74 0.8540 0.03 18.28 1 / 485 0 / 119CLAPT PLACEBO 0.34 0.5060 0.01 8.34 0 / 112 1 / 114SHARP PLACEBO 0.80 0.0356 0.65 0.99 171 / 4650 210 / 4620

0.84 0.0000 0.78 0.91

0.5 1 2

Favors Active Favors Control

ICH - Statin Meta-Analysis: All Stroke

All studies

Figure 3. Forrest plot of random-effects meta-analysis of randomized trials of statins and all stroke.

McKinney and Kostis Statin Therapy and Risk of ICH 2153

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 6: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

therapy (relative risk, 1.03; 95% CI, 0.75–1.41). Studyingintensive LDL reduction with statins, the CTT collaborationperformed a meta-analysis of 26 RCTs and reported anonsignificant trend toward an increased risk of ICH (relativerisk, 1.15; 95% CI, 0.93–1.41).30 A more recent meta-analysisincluding 23 RCTs found no evidence of increased risk ofICH.53

We performed a comprehensive meta-analysis of previ-ously conducted randomized clinical trials of statin therapyfor which data on ICH or hemorrhagic stroke were available.We included 31 RCTs with a total of 182 803 subjects. In ouranalysis, active therapy was associated with a nonsignificantincrease in the risk of ICH (OR, 1.08; 95% CI, 0.88–1.32).

The small nonsignificant observed excess of ICH was notrelated to the LDL effects of statin therapy. We found norelationship between the achieved LDL level or the degree ofLDL reduction and the risk of hemorrhagic stroke in thisanalysis. This suggests that any potential increase in ICH riskcould be attributable to other non-LDL effects of statintherapy. This is in agreement with a large meta-analysis ofobservational cohort studies that found no correlation withfatal stroke and baseline total cholesterol in almost 900 000subjects.27 In addition to their lipid-lowering properties,

statins may have antithrombotic properties by inhibitingplatelet aggregation and enhancing fibrinolysis.54,55 The an-tithrombotic affects of statins could account for a theoreti-cally increased risk of bleeding complications.

When stratifying our analysis by type of prevention, therewas a nonsignificant trend toward increased odds of ICH insecondary prevention studies (OR, 1.26; 95% CI, 0.91–1.73)compared with primary prevention studies (OR, 0.96; 95%CI, 0.75–1.23). The present study was not limited to trials ofstatin therapy for secondary stroke prevention. Post hocanalyses of HPS and SPARCL suggest that patients with priorstroke may be at particularly increased risk of hemorrhagicstroke.28,51 However, a recent large retrospective cohort studyfound no evidence of increased risk of ICH in patients treatedwith statins after an ischemic stroke.52 Westover et al,56 usinga decision-analytic approach, found that avoiding statins,particularly in patients with a history of lobar hemorrhage(and who are at high risk for recurrent hemorrhage), wasfavored over a wide range of clinical parameters. Our analysisdid not evaluate the risk of statin therapy in this subgroup ofpatients with prior ICH. Based on current data, caution shouldbe used in treating these patients with statins, and furtherresearch is warranted.

Study name Subgroup within study Statistics for each study Dead / Total Odds ratio and 95% CI

Odds Lower Upper ratio p-Value limit limit Active Control

4S PLACEBO 0.69 0.0002 0.56 0.84 182 / 2221 256 / 2223AF-TEXCAPS PLACEBO 1.04 0.8130 0.76 1.43 80 / 3304 77 / 3301ALLHAT-LLT PLACEBO 0.99 0.8071 0.88 1.11 631 / 5170 641 / 5185ASCOT-LLA PLACEBO 0.86 0.1493 0.71 1.05 185 / 5168 212 / 5137ATOZ LOW DOSE 1.25 0.1036 0.96 1.62 130 / 2265 104 / 2232AURORA PLACEBO 0.93 0.3162 0.80 1.08 636 / 1389 660 / 1384CARE PLACEBO 0.91 0.3791 0.74 1.12 180 / 2081 196 / 2078CORONA PLACEBO 0.93 0.2650 0.83 1.05 728 / 2514 759 / 2497GISSI-P PLACEBO 0.81 0.1928 0.59 1.11 72 / 2138 88 / 2133GREACE PLACEBO 0.56 0.0309 0.33 0.95 23 / 800 40 / 800HPS PLACEBO 0.86 0.0003 0.80 0.93 1328 / 10269 1507 / 10267JUPITER PLACEBO 0.80 0.0189 0.66 0.96 198 / 8901 247 / 8901LIPID PLACEBO 0.77 0.0000 0.69 0.86 717 / 4512 888 / 4502MEGA PLACEBO 0.71 0.0532 0.50 1.00 55 / 3866 79 / 3966PROSPER PLACEBO 0.98 0.8393 0.83 1.16 298 / 2891 305 / 2913PROVE-IT LOW DOSE 0.68 0.0458 0.46 0.99 46 / 2099 66 / 2063TNT LOW DOSE 1.01 0.9096 0.85 1.20 284 / 4995 282 / 5006SEARCH LOW DOSE 0.99 0.8879 0.90 1.09 964 / 6031 970 / 6033SPARCL PLACEBO 1.03 0.7962 0.84 1.25 216 / 2365 211 / 2366CARDS PLACEBO 0.72 0.0617 0.51 1.02 61 / 1429 82 / 1412ASPEN PLACEBO 1.02 0.9084 0.72 1.44 70 / 1211 68 / 1199GISSI-HF PLACEBO 1.03 0.6431 0.91 1.17 657 / 2285 644 / 22894D PLACEBO 0.90 0.3628 0.72 1.13 297 / 619 320 / 633IDEAL LOW DOSE 0.98 0.7832 0.84 1.14 366 / 4439 374 / 4449MIRACL PLACEBO 0.95 0.7507 0.67 1.34 64 / 1538 68 / 1548PATE LOW DOSE 0.69 0.3056 0.34 1.40 14 / 331 20 / 334ACAPS PLACEBO 0.12 0.0485 0.02 0.99 1 / 460 8 / 459ALERT PLACEBO 1.04 0.7357 0.81 1.34 143 / 1050 138 / 1052CLAPT PLACEBO 0.20 0.3006 0.01 4.21 0 / 112 2 / 114SHARP PLACEBO 1.02 0.6336 0.93 1.13 1142 / 4650 1115 / 4620

0.92 0.0007 0.87 0.96

0.5 1 2

Favors Active Favors Control

ICH - Statin Meta-Analysis: Death

All studies

Figure 4. Forrest plot of random-effects meta-analysis of randomized trials of statins and all-cause mortality.

2154 Stroke August 2012

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 7: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

There are several limitations to this analysis. First, we didnot have access to patient-level data and there may beunaccounted patient variables that influenced the rates ofhemorrhagic stroke. Second, hemorrhagic stroke is a nebu-lous term that may include ICH, subarachnoid hemorrhage,subdural or epidural hemorrhages, or hemorrhagic transfor-mation of an ischemic stroke. Although ICH is the conditionof interest, it is possible that prior RCTs that reportedhemorrhagic stroke rates included some of these other causesof intracranial bleeding. Lastly, although we performed themost comprehensive analysis yet published of statin therapyand ICH risk in randomized trials, it is possible that there areother trials, particularly those published in languages otherthan English, that were excluded. Although we included allpertinent trials, regardless of size, no publication bias orheterogeneity was found in our analysis.

Statin therapy was not associated with a significant in-creased risk of ICH. There was no effect on ICH risk relatedto the degree of decline in LDL or to the achieved level. Thesignificant reduction in total stroke and all-cause mortalitymore than offset any slight increase in ICH risk. Thesefindings support the current recommendations to prescribestatins in otherwise appropriate patients.

ConclusionsIn this meta-analysis of 31 RCTs, statin therapy was notassociated with a significant increase in odds of ICH. How-ever, statin therapy is associated with significant reductions intotal stroke and death, thereby negating any potential hemor-rhage risk in an unselected patient population.

Sources of FundingThis study was funded in part by the Robert Wood JohnsonFoundation and the Schering-Plough Foundation.

DisclosuresNone.

References1. Iso H, Jacobs DR Jr, Wentworth D, Neaton JD, Cohen JD. Serum

cholesterol levels and six-year mortality from stroke in 350 977 menscreened for the multiple risk factor intervention trial. N Engl J Med.1989;320:904–910.

2. Leppala JM, Virtamo J, Fogelholm R, Albanes D, Heinonen OP. Dif-ferent risk factors for different stroke subtypes: association of bloodpressure, cholesterol, and antioxidants. Stroke. 1999;30:2535–2540.

3. Horenstein RB, Smith DE, Mosca L. Cholesterol predicts stroke mortalityin the women’s pooling project. Stroke. 2002;33:1863–1868.

4. Zhang X, Patel A, Horibe H, Wu Z, Barzi F, Rodgers A, et al. Choles-terol, coronary heart disease, and stroke in the Asia Pacific region. Int JEpidemiol. 2003;32:563–572.

5. Kurth T, Everett BM, Buring JE, Kase CS, Ridker PM, Gaziano JM.Lipid levels and the risk of ischemic stroke in women. Neurology.2007;68:556–562.

6. Randomised trial of cholesterol lowering in 4444 patients with coronaryheart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet.1994;344:1383–1389.

7. Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG,et al. The effect of pravastatin on coronary events after myocardialinfarction in patients with average cholesterol levels. Cholesterol andrecurrent events trial investigators. N Engl J Med. 1996;335:1001–1009.

8. Downs JR, Clearfield M, Weis S, Whitney E, Shapiro DR, Beere PA, etal. Primary prevention of acute coronary events with lovastatin in menand women with average cholesterol levels: results of AFCAPS/

TEXCAPS. Air Force/Texas Coronary Atherosclerosis Prevention Study.JAMA. 1998;279:1615–1622.

9. Prevention of cardiovascular events and death with pravastatin in patientswith coronary heart disease and a broad range of initial cholesterol levels.The Long-Term Intervention With Pravastatin in Ischaemic Disease(LIPID) study group. N Engl J Med. 1998;339:1349–1357.

10. Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters D,et al. Effects of atorvastatin on early recurrent ischemic events in acutecoronary syndromes: the MIRACL study: a randomized controlled trial.JAMA. 2001;285:1711–1718.

11. Ito H, Ouchi Y, Ohashi Y, Saito Y, Ishikawa T, Nakamura H, et al. Acomparison of low versus standard dose pravastatin therapy for theprevention of cardiovascular events in the elderly: the Pravastatin Anti-Atherosclerosis Trial in the Elderly (PATE). J Atheroscler Thromb.2001;8:33–44.

12. Major outcomes in moderately hypercholesterolemic, hypertensive patientsrandomized to pravastatin vs usual care: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT). JAMA.2002;288:2998–3007.

13. MRC/BHF Heart Protection Study of cholesterol lowering with simva-statin in 20 536 high-risk individuals: a randomised placebo-controlledtrial. Lancet. 2002;360:7–22.

14. Athyros VG, Papageorgiou AA, Mercouris BR, Athyrou VV, SymeonidisAN, Basayannis EO, et al. Treatment with atorvastatin to the nationalcholesterol educational program goal versus ‘usual’ care in secondarycoronary heart disease prevention. The Greek Atorvastatin and Coronary-Heart-Disease Evaluation (GREACE) study. Curr Med Res Opin. 2002;18:220–228.

15. Shepherd J, Blauw GJ, Murphy MB, Bollen EL, Buckley BM, Cobbe SM,et al. Pravastatin in Elderly Individuals at Risk of Vascular Disease(PROSPER): a randomised controlled trial. Lancet. 2002;360:1623–1630.

16. Sever PS, Dahlof B, Poulter NR, Wedel H, Beevers G, Caulfield M, et al.Prevention of coronary and stroke events with atorvastatin in hyper-tensive patients who have average or lower-than-average cholesterolconcentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial–LipidLowering Arm (ASCOT-LLA): a multicentre randomised controlled trial.Lancet. 2003;361:1149–1158.

17. de Lemos JA, Blazing MA, Wiviott SD, Lewis EF, Fox KA, White HD,et al. Early intensive vs a delayed conservative simvastatin strategy inpatients with acute coronary syndromes: phase Z of the A to Z Trial.JAMA. 2004;292:1307–1316.

18. Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder R,et al. Intensive versus moderate lipid lowering with statins after acutecoronary syndromes. N Engl J Med. 2004;350:1495–1504.

19. Colhoun HM, Betteridge DJ, Durrington PN, Hitman GA, Neil HA,Livingstone SJ, et al. Primary prevention of cardiovascular disease withatorvastatin in type 2 diabetes in the Collaborative Atorvastatin DiabetesStudy (CARDS): multicentre randomised placebo-controlled trial.Lancet. 2004;364:685–696.

20. LaRosa JC, Grundy SM, Waters DD, Shear C, Barter P, Fruchart JC, etal. Intensive lipid lowering with atorvastatin in patients with stablecoronary disease. N Engl J Med. 2005;352:1425–1435.

21. Nakamura H, Arakawa K, Itakura H, Kitabatake A, Goto Y, Toyota T, etal. Primary prevention of cardiovascular disease with pravastatin in Japan(Mega study): a prospective randomised controlled trial. Lancet. 2006;368:1155–1163.

22. Amarenco P, Bogousslavsky J, Callahan A III, Goldstein LB, HennericiM, Rudolph AE, et al. High-dose atorvastatin after stroke or transientischemic attack. N Engl J Med. 2006;355:549–559.

23. Ridker PM, Danielson E, Fonseca FA, Genest J, Gotto AM Jr, KasteleinJJ, et al. Rosuvastatin to prevent vascular events in men and women withelevated C-reactive protein. N Engl J Med. 2008;359:2195–2207.

24. Armitage J, Bowman L, Wallendszus K, Bulbulia R, Rahimi K, HaynesR, et al. Intensive lowering of LDL cholesterol with 80 mg versus 20 mgsimvastatin daily in 12 064 survivors of myocardial infarction: adouble-blind randomised trial. Lancet. 2010;376:1658–1669.

25. Amarenco P, Labreuche J. Lipid management in the prevention of stroke:review and updated meta-analysis of statins for stroke prevention. LancetNeurol. 2009;8:453–463.

26. Cholesterol, diastolic blood pressure, and stroke: 13 000 strokes in450 000 people in 45 prospective cohorts. Prospective studies collabo-ration. Lancet. 1995;346:1647–1653.

27. Lewington S, Whitlock G, Clarke R, Sherliker P, Emberson J, Halsey J,et al. Blood cholesterol and vascular mortality by age, sex, and blood

McKinney and Kostis Statin Therapy and Risk of ICH 2155

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 8: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

pressure: a meta-analysis of individual data from 61 prospective studieswith 55 000 vascular deaths. Lancet. 2007;370:1829–1839.

28. Goldstein MR, Mascitelli L, Pezzetta F. Hemorrhagic stroke in thestroke prevention by Aggressive Reduction in Cholesterol LevelsStudy. Neurology. 2009;72:1448; author reply 1448 –1449.

29. Manktelow BN, Potter JF. Interventions in the management of serumlipids for preventing stroke recurrence. Cochrane Database Syst Rev.2009;3:CD002091.

30. Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, et al.Efficacy and safety of more intensive lowering of LDL cholesterol: ameta-analysis of data from 170 000 participants in 26 randomised trials.Lancet. 2010;376:1670–1681.

31. Kostis WJ, Moreyra AE, Cheng JQ, Dobrzynski JM, Kostis JB. Contin-uation of mortality reduction after the end of randomized therapy inclinical trials of lipid-lowering therapy. J Clin Lipidol. 2011;5:97–104.

32. Borenstein M. Introduction to Meta-Analysis. Chichester, UK: JohnWiley & Sons; 2009.

33. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method oftesting and adjusting for publication bias in meta-analysis. Biometrics.2000;56:455–463.

34. Rosenthal R. The ’file drawer problem’ and tolerance for null results.Psychol Bull. 1979;86:638–641.

35. Orwin RG. A fail-safe N for effect size in meta-analysis. J Educ Stat.1983;8:157–159.

36. Fellstrom BC, Jardine AG, Schmieder RE, Holdaas H, Bannister K,Beutler J, et al. Rosuvastatin and cardiovascular events in patientsundergoing hemodialysis. N Engl J Med. 2009;360:1395–1407.

37. Kjekshus J, Apetrei E, Barrios V, Bohm M, Cleland JG, Cornel JH, et al.Rosuvastatin in older patients with systolic heart failure. N Engl J Med.2007;357:2248–2261.

38. Results of the low-dose (20 mg) pravastatin GISSI Prevenzione Trial in4271 patients with recent myocardial infarction: do stopped trials con-tribute to overall knowledge? Gissi Prevenzione Investigators (GruppoItaliano Per Lo Studio Della Sopravvivenza Nell’infarto Miocardico). ItalHeart J. 2000;1:810–820.

39. Knopp RH, d’Emden M, Smilde JG, Pocock SJ. Efficacy and safety ofatorvastatin in the prevention of cardiovascular end points in subjectswith type 2 diabetes: the Atorvastatin Study for Prevention of CoronaryHeart Disease Endpoints in Non-Insulin-Dependent Diabetes Mellitus(ASPEN). Diabetes Care. 2006;29:1478–1485.

40. Tavazzi L, Maggioni AP, Marchioli R, Barlera S, Franzosi MG, Latini R,et al. Effect of rosuvastatin in patients with chronic heart failure (theGISSI-HF Trial): a randomised, double-blind, placebo-controlled trial.Lancet. 2008;372:1231–1239.

41. Wanner C, Krane V, Marz W, Olschewski M, Mann JF, Ruf G, et al.Atorvastatin in patients with type 2 diabetes mellitus undergoing hemo-dialysis. N Engl J Med. 2005;353:238–248.

42. Pedersen TR, Faergeman O, Kastelein JJ, Olsson AG, Tikkanen MJ,Holme I, et al. High-dose atorvastatin vs usual-dose simvastatin forsecondary prevention after myocardial infarction: the IDEAL study: arandomized controlled trial. JAMA. 2005;294:2437–2445.

43. Furberg CD, Adams HP Jr, Applegate WB, Byington RP, Espeland MA,Hartwell T, et al. Effect of lovastatin on early carotid atherosclerosis andcardiovascular events. Asymptomatic Carotid Artery Progression Study(ACAPS) research group. Circulation. 1994;90:1679–1687.

44. Holdaas H, Fellstrom B, Jardine AG, Holme I, Nyberg G, Fauchald P, etal. Effect of fluvastatin on cardiac outcomes in renal transplant recipients:a multicentre, randomised, placebo-controlled trial. Lancet. 2003;361:2024–2031.

45. Bone HG, Kiel DP, Lindsay RS, Lewiecki EM, Bolognese MA, LearyET, et al. Effects of atorvastatin on bone in postmenopausal women withdyslipidemia: a double-blind, placebo-controlled, dose-ranging trial.J Clin Endocrinol Metab. 2007;92:4671–4677.

46. Kleemann A, Eckert S, von Eckardstein A, Lepper W, Schernikau U,Gleichmann U, et al. Effects of lovastatin on progression of non-dilatedand dilated coronary segments and on restenosis in patients after PTCA.The Cholesterol Lowering Atherosclerosis PTCA Trial (CLAPT). EurHeart J. 1999;20:1393–1406.

47. Baigent C, Landray MJ, Reith C, Emberson J, Wheeler DC, Tomson C,et al. The effects of lowering LDL cholesterol with simvastatin plusezetimibe in patients with chronic kidney disease (study of heart and renalprotection): a randomised placebo-controlled trial. Lancet. 2011;377:2181–2192.

48. Sturgeon JD, Folsom AR, Longstreth WT Jr, Shahar E, Rosamond WD,Cushman M. Risk factors for intracerebral hemorrhage in a pooled pro-spective study. Stroke. 2007;38:2718–2725.

49. Iribarren C, Reed DM, Burchfiel CM, Dwyer JH. Serum total cholesteroland mortality. Confounding factors and risk modification in Japanese-American men. JAMA. 1995;273:1926–1932.

50. Bjorkhem I, Meaney S. Brain cholesterol: long secret life behind a barrier.Arterioscler Thromb Vasc Biol. 2004;24:806–815.

51. Collins R, Armitage J, Parish S, Sleight P, Peto R. Effects of cholester-ol-lowering with simvastatin on stroke and other major vascular events in20 536 people with cerebrovascular disease or other high-risk conditions.Lancet. 2004;363:757–767.

52. Hackam DG, Austin PC, Huang A, Juurlink DN, Mamdani MM, PatersonJM, et al. Statins and intracerebral hemorrhage: a retrospective cohortstudy. Arch Neurol. 2012;69:39–45.

53. Hackam DG, Woodward M, Newby LK, Bhatt DL, Shao M, Smith EE,et al. Statins and intracerebral hemorrhage: collaborative systematicreview and meta-analysis. Circulation. 2011;124:2233–2242.

54. Serebruany VL, Miller M, Pokov AN, Malinin AI, Lowry DR, TanguayJF, et al. Effect of statins on platelet par-1 thrombin receptor in patientswith the metabolic syndrome (from the PAR-1 Inhibition by Statins[PARIS] study). Am J Cardiol. 2006;97:1332–1336.

55. Serebruany VL, Malinin AI, Hennekens CH. Statins increase risk ofhemorrhagic stroke by inhibition of the PAR-1 receptor. CerebrovascDis. 2007;24:477–479.

56. Westover MB, Bianchi MT, Eckman MH, Greenberg SM. Statin usefollowing intracerebral hemorrhage: a decision analysis. Arch Neurol.2010;68:573–579.

2156 Stroke August 2012

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 9: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

James S. McKinney and William J. KostisRandomized Controlled Trials

Statin Therapy and the Risk of Intracerebral Hemorrhage: A Meta-Analysis of 31

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 2012 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/STROKEAHA.112.655894

2012;43:2149-2156; originally published online May 15, 2012;Stroke. 

http://stroke.ahajournals.org/content/43/8/2149World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://stroke.ahajournals.org/content/suppl/2016/04/10/STROKEAHA.112.655894.DC3 http://stroke.ahajournals.org/content/suppl/2013/10/02/STROKEAHA.112.655894.DC2 http://stroke.ahajournals.org/content/suppl/2012/05/15/STROKEAHA.112.655894.DC1

Data Supplement (unedited) at:

  http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer process is available in the

Request Permissions in the middle column of the Web page under Services. Further information about thisOnce the online version of the published article for which permission is being requested is located, click

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Strokein Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on July 18, 2018http://stroke.ahajournals.org/

Dow

nloaded from

Page 10: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

SUPPLEMENTAL MATERIAL Supplemental Figure 1. Literature search criteria for Medline (A), Web of Science (B), and the Cochrane Library (C), January 25, 2012.

Page 11: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

A.

Page 12: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

B.

Page 13: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis
Page 14: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

C.

Page 15: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Supplemental Figure 2. Forrest plot of random-effects meta-analysis of statin therapy and intracerebral hemorrhage stratified by type of control group (A) and type of prevention (B).

Page 16: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

A.

Page 17: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Group bySubgroup within study

Subgroup within study Statistics for each study Odds ratio and 95% CI

Odds Lower Upper ratio p-Value limit limit

PRIMARY AF-TEXCAPSPRIMARY 3.00 0.5014 0.12 73.62PRIMARY ALLHAT-LLTPRIMARY 3.42 0.0158 1.26 9.27PRIMARY ASCOT-LLAPRIMARY 0.55 0.1071 0.26 1.14PRIMARY AURORA PRIMARY 1.19 0.5608 0.66 2.14PRIMARY HPS PRIMARY 0.96 0.8434 0.65 1.41PRIMARY JUPITER PRIMARY 0.67 0.4415 0.24 1.87PRIMARY MEGA PRIMARY 1.17 0.6632 0.57 2.41PRIMARY PROSPER PRIMARY 0.81 0.6490 0.32 2.04PRIMARY ASPEN PRIMARY 1.98 0.4296 0.36 10.85PRIMARY PATE PRIMARY 0.60 0.1963 0.28 1.30PRIMARY ACAPS PRIMARY 0.14 0.1965 0.01 2.75PRIMARY ALERT PRIMARY 0.59 0.1817 0.27 1.28PRIMARY BONE PRIMARY 0.74 0.8540 0.03 18.28PRIMARY SHARP PRIMARY 1.21 0.3916 0.78 1.87PRIMARY 0.96 0.7663 0.75 1.23SECONDARY 4S SECONDARY 0.20 0.2989 0.01 4.17SECONDARY ATOZ SECONDARY 12.84 0.0820 0.72 228.14SECONDARY CARE SECONDARY 0.33 0.1774 0.07 1.65SECONDARY CORONA SECONDARY 1.66 0.2308 0.72 3.80SECONDARY GISSI-P SECONDARY 2.99 0.5019 0.12 73.55SECONDARY GREACE SECONDARY 1.00 1.0000 0.06 16.02SECONDARY LIPID SECONDARY 1.89 0.1236 0.84 4.24SECONDARY PROVE-IT SECONDARY 3.94 0.2205 0.44 35.25SECONDARY TNT SECONDARY 0.94 0.8666 0.48 1.87SECONDARY SEARCH SECONDARY 0.96 0.8871 0.55 1.68SECONDARY SPARCL SECONDARY 1.68 0.0191 1.09 2.60SECONDARY GISSI-HF SECONDARY 3.69 0.0454 1.03 13.23SECONDARY 4D SECONDARY 0.64 0.4299 0.21 1.96SECONDARY IDEAL SECONDARY 1.00 0.9969 0.32 3.11SECONDARY MIRACL SECONDARY 0.14 0.1992 0.01 2.78SECONDARY CLAPT SECONDARY 0.34 0.5060 0.01 8.34SECONDARY 1.26 0.1638 0.91 1.73Overall 1.06 0.5409 0.87 1.29

0.01 0.1 1 10 100

Favors Active Favors Control

ICH Meta-Analysis: ICH OR by Type of Prevention

B.

Page 18: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Supplemental Table 1. Description of Trials, Patient Demographics Trial, Year Active Drug Control Total

Subjects Age

(y ears) Gender

(% male) Race

(% white) Follow-up (months) DM (%) HTN (%) Stroke

(%) CVD (%) Smoking (%)

ASA/OA (%)

4S, 19941 Simvastatin Placebo 4444 63 81 --- 64.8 4.5 26.0 --- 100 26 37 AF-TEXCAPS, 19982 Lovastatin Placebo 6605 58 85 89 62.4 12.2 22.0 --- 0 13 17

ALLHAT-LLT, 20023 Pravastatin Usual care 10355 66 51 41 57.6 35.2 100.0 --- 14 23 31

ASCOT-LLA, 20034 Atorvastatin Placebo 10305 63 81 95 39.6 24.6 100.0 9.7 19 33 17

A-to-Z, 20045 Simvastatin Low-dose 4497 61 76 --- 24 23.5 50.0 --- 100 41 98 AURORA, 20096 Rosuvastatin Placebo 2773 64 62 85 45.6 26.0 39.6 --- 40 15 42

CARE, 19967 Pravastatin Placebo 4159 59 86 93 60 15.5 42.5 --- 100 21 83 CORONA, 20078 Rosuvastatin Placebo 5011 73 76 --- 32.8 29.5 63.0 12.5 100 9 60

GISSI-P, 20009 Pravastatin Usual care 4271 60 86 --- 23 13.7 36.5 --- 100 14 50 GREACE, 200210 Atorvastatin Usual care 1600 59 79 --- 36 19.5 43.0 --- 100 5 88

HPS, 200211 Simvastatin Placebo 20536 65 75 --- 60 19.0 41.0 --- 87 14 63 JUPITER, 200812 Rosuvastatin Placebo 17802 66 62 71 22.8 0.0 56.7 N/A 0 16 17

LIPID, 199813 Pravastatin Placebo 9014 62 83 --- 72 9.0 41.5 3.6 100 63 83 MEGA, 200614 Pravastatin Placebo 7832 58 32 0 63.6 21.0 42.0 0 0 21 39 PROSPER, 200215 Pravastatin Placebo 5804 75 48 --- 38.4 11.8 61.9 11 44 27 36

PROVE-IT, 200416 Atorvastatin Low dose 4162 58 78 91 24 17.6 50.0 --- 100 37 100

TNT, 200517 Atorvastatin Low dose 10001 61 81 94 58.8 15.0 54.3 5.2 100 13 88 SEARCH, 201018 Simvastatin Low dose 12064 64 83 --- 80.4 11.0 42.0 7 100 30 98

SPARCL, 200619 Atorvastatin Placebo 4731 63 60 --- 58.8 17.0 62.0 69 100 19 87

CARDS, 200420 Atorvastatin Placebo 2841 62 68 95 46.8 100.0 84.0 0 0 23 15 ASPEN, 200621 Atorvastatin Placebo 2410 61 66 84 48 100.0 55.0 --- 21 12 --- GISSI-HF, 200822 Rosuvastatin Placebo 4574 68 77 --- 46.8 26 54.0 4.5 100 14 75

4D, 200523 Atorvastatin Placebo 1252 66 54 --- 46.8 100.0 88.0 18 84 9 52 IDEAL, 200524 Atorvastatin Low dose 8888 62 81 --- 57.6 12 33 8.2 39 29 79 MIRACL, 200125 Atorvastatin Placebo 3086 65 65 85 4 23 55 8.6 100 28 91

PATE, 200126 Pravastatin Low dose 665 73 21 --- 46.8 30 50 12.9 72 29 --- ACAPS, 199427 Lovastatin Placebo 919 62 52 92 34.1 2.3 28.8 0 0 11.9 100 ALERT, 200328 Fluvastatin Placebo 2102 50 66 --- 61.2 19 75 5.8 3.1 19 43 BONE, 200729 Atorvastatin Placebo 604 59 0 84 13 0 --- 0 --- --- --- CLAPT, 199930 Lovastatin Usual care 226 54 100 --- 24 5.7 41.2 --- 50.9 10.2 --- SHARP, 201131 Simvastatin Placebo 9720 62 63 72 58.8 23 --- --- --- 13 --- All Studies (mean±SD) 63±5 67±21 78±26 46±18 25±27 53±20 11±16 59±43 21±12 61±30

DM indicated diabetes mellitus; HTN, hypertension; CVD, cardiovascular disease; ASA, aspirin; OA, oral anticoagulant.

Page 19: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Supplemental Table 2. Description of Trials, LDL Characteristics Trial, Year

Baseline LDL-Active

Follow-up LDL-Acive ΔLDL-Active Baseline LDL-

Control Follow-up

LDL-Control ΔLDL-Control ΔLDL-Active

– ΔLDL-Control

Mean Baseline LDL

ΔLDL-Acitve - ΔLDL-Control

(%) 4S, 19941 188 122 66 188 190 -2 68 188 36 AF-TEXCAPS, 19982 150 115 35 150 158 -8 43 150 29

ALLHAT-LLT, 20023 146 111 35 146 134 12 23 146 16

ASCOT-LLA, 20034 133 87 46 133 130 3 43 133 32

A-to-Z, 20045 112 63 49 111 77 34 15 112 13 AURORA, 20096 100 58 42 99 97 2 40 100 40 CARE, 19967 139 98 41 139 136 3 38 139 27 CORONA, 20078 137 76 61 136 138 -2 63 137 46 GISSI-P, 20009 152 130 22 152 147 5 17 152 11 GREACE, 200210 180 97 83 179 169 10 73 180 41 HPS, 200211 131 89 42 131 128 3 39 131 30 JUPITER, 200812 108 54 54 108 108 0 54 108 50 LIPID, 199813 150 105 45 150 148 2 43 150 29 MEGA, 200614 157 127 30 157 148 9 21 157 13 PROSPER, 200215 147 97 50 147 147 0 50 147 34 PROVE-IT, 200416 106 62 44 106 95 11 33 106 31 TNT, 200517 97 77 20 98 101 -3 23 98 24 SEARCH, 201018 97 81 16 97 93 4 12 97 12 SPARCL, 200619 133 72 61 134 128 6 55 134 41 CARDS, 200420 118 72 46 117 120 -3 49 118 42 ASPEN, 200621 113 93 20 114 112 2 18 114 16 GISSI-HF, 200822 122 83 39 121 130 -9 48 122 40 4D, 200523 121 72 49 125 120 5 44 123 36 IDEAL, 200524 122 82 40 121 104 17 23 122 19 MIRACL, 200125 124 72 52 124 135 -11 63 124 51 PATE, 200126 163 123 40 163 133 30 10 163 6 ACAPS, 199427 157 113 44 155 155 0 44 156 28 ALERT, 200328 159 108 51 159 146 13 38 159 24 BONE, 200729 156 89 67 159 159 0 67 157 43 CLAPT, 199930 181 126 55 183 162 21 34 182 19 SHARP, 201131 108 75 33 108 105 3 30 108 28 All Studies (mean±SD) 137±25 91±22 45±14 137±25 132±25 5±10 40±17 137±25 30±12

LDL indicated low density lipoprotein.

Page 20: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Supplemental References. 1. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: The scandinavian simvastatin survival

study (4s). Lancet. 1994;344:1383-1389. 2. Downs JR, Clearfield M, Weis S, Whitney E, Shapiro DR, Beere PA, et al. Primary prevention of acute coronary events with

lovastatin in men and women with average cholesterol levels: Results of afcaps/texcaps. Air force/texas coronary atherosclerosis prevention study. JAMA. 1998;279:1615-1622.

3. Major outcomes in moderately hypercholesterolemic, hypertensive patients randomized to pravastatin vs usual care: The antihypertensive and lipid-lowering treatment to prevent heart attack trial (allhat-llt). JAMA. 2002;288:2998-3007.

4. Sever PS, Dahlof B, Poulter NR, Wedel H, Beevers G, Caulfield M, et al. Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the anglo-scandinavian cardiac outcomes trial--lipid lowering arm (ascot-lla): A multicentre randomised controlled trial. Lancet. 2003;361:1149-1158.

5. de Lemos JA, Blazing MA, Wiviott SD, Lewis EF, Fox KA, White HD, et al. Early intensive vs a delayed conservative simvastatin strategy in patients with acute coronary syndromes: Phase z of the a to z trial. JAMA. 2004;292:1307-1316.

6. Fellstrom BC, Jardine AG, Schmieder RE, Holdaas H, Bannister K, Beutler J, et al. Rosuvastatin and cardiovascular events in patients undergoing hemodialysis. N Engl J Med. 2009;360:1395-1407.

7. Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and recurrent events trial investigators. N Engl J Med. 1996;335:1001-1009.

8. Kjekshus J, Apetrei E, Barrios V, Bohm M, Cleland JG, Cornel JH, et al. Rosuvastatin in older patients with systolic heart failure. N Engl J Med. 2007;357:2248-2261.

9. Results of the low-dose (20 mg) pravastatin gissi prevenzione trial in 4271 patients with recent myocardial infarction: Do stopped trials contribute to overall knowledge? Gissi prevenzione investigators (gruppo italiano per lo studio della sopravvivenza nell'infarto miocardico). Ital Heart J. 2000;1:810-820.

10. Athyros VG, Papageorgiou AA, Mercouris BR, Athyrou VV, Symeonidis AN, Basayannis EO, et al. Treatment with atorvastatin to the national cholesterol educational program goal versus 'usual' care in secondary coronary heart disease prevention. The greek atorvastatin and coronary-heart-disease evaluation (greace) study. Curr Med Res Opin. 2002;18:220-228.

11. Mrc/bhf heart protection study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: A randomised placebo-controlled trial. Lancet. 2002;360:7-22.

12. Ridker PM, Danielson E, Fonseca FA, Genest J, Gotto AM, Jr., Kastelein JJ, et al. Rosuvastatin to prevent vascular events in men and women with elevated c-reactive protein. N Engl J Med. 2008;359:2195-2207.

Page 21: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

13. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. The long-term intervention with pravastatin in ischaemic disease (lipid) study group. N Engl J Med. 1998;339:1349-1357.

14. Nakamura H, Arakawa K, Itakura H, Kitabatake A, Goto Y, Toyota T, et al. Primary prevention of cardiovascular disease with pravastatin in japan (mega study): A prospective randomised controlled trial. Lancet. 2006;368:1155-1163.

15. Shepherd J, Blauw GJ, Murphy MB, Bollen EL, Buckley BM, Cobbe SM, et al. Pravastatin in elderly individuals at risk of vascular disease (prosper): A randomised controlled trial. Lancet. 2002;360:1623-1630.

16. Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder R, et al. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med. 2004;350:1495-1504.

17. LaRosa JC, Grundy SM, Waters DD, Shear C, Barter P, Fruchart JC, et al. Intensive lipid lowering with atorvastatin in patients with stable coronary disease. N Engl J Med. 2005;352:1425-1435.

18. Armitage J, Bowman L, Wallendszus K, Bulbulia R, Rahimi K, Haynes R, et al. Intensive lowering of ldl cholesterol with 80 mg versus 20 mg simvastatin daily in 12,064 survivors of myocardial infarction: A double-blind randomised trial. Lancet. 2010;376:1658-1669.

19. Amarenco P, Bogousslavsky J, Callahan A, 3rd, Goldstein LB, Hennerici M, Rudolph AE, et al. High-dose atorvastatin after stroke or transient ischemic attack. N Engl J Med. 2006;355:549-559.

20. Colhoun HM, Betteridge DJ, Durrington PN, Hitman GA, Neil HA, Livingstone SJ, et al. Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the collaborative atorvastatin diabetes study (cards): Multicentre randomised placebo-controlled trial. Lancet. 2004;364:685-696.

21. Knopp RH, d'Emden M, Smilde JG, Pocock SJ. Efficacy and safety of atorvastatin in the prevention of cardiovascular end points in subjects with type 2 diabetes: The atorvastatin study for prevention of coronary heart disease endpoints in non-insulin-dependent diabetes mellitus (aspen). Diabetes Care. 2006;29:1478-1485.

22. Tavazzi L, Maggioni AP, Marchioli R, Barlera S, Franzosi MG, Latini R, et al. Effect of rosuvastatin in patients with chronic heart failure (the gissi-hf trial): A randomised, double-blind, placebo-controlled trial. Lancet. 2008;372:1231-1239.

23. Wanner C, Krane V, Marz W, Olschewski M, Mann JF, Ruf G, et al. Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis. N Engl J Med. 2005;353:238-248.

24. Pedersen TR, Faergeman O, Kastelein JJ, Olsson AG, Tikkanen MJ, Holme I, et al. High-dose atorvastatin vs usual-dose simvastatin for secondary prevention after myocardial infarction: The ideal study: A randomized controlled trial. JAMA. 2005;294:2437-2445.

25. Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters D, et al. Effects of atorvastatin on early recurrent ischemic events in acute coronary syndromes: The miracl study: A randomized controlled trial. JAMA. 2001;285:1711-1718.

Page 22: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

26. Ito H, Ouchi Y, Ohashi Y, Saito Y, Ishikawa T, Nakamura H, et al. A comparison of low versus standard dose pravastatin therapy for the prevention of cardiovascular events in the elderly: The pravastatin anti-atherosclerosis trial in the elderly (pate). J Atheroscler Thromb. 2001;8:33-44.

27. Furberg CD, Adams HP, Jr., Applegate WB, Byington RP, Espeland MA, Hartwell T, et al. Effect of lovastatin on early carotid atherosclerosis and cardiovascular events. Asymptomatic carotid artery progression study (acaps) research group. Circulation. 1994;90:1679-1687.

28. Holdaas H, Fellstrom B, Jardine AG, Holme I, Nyberg G, Fauchald P, et al. Effect of fluvastatin on cardiac outcomes in renal transplant recipients: A multicentre, randomised, placebo-controlled trial. Lancet. 2003;361:2024-2031.

29. Bone HG, Kiel DP, Lindsay RS, Lewiecki EM, Bolognese MA, Leary ET, et al. Effects of atorvastatin on bone in postmenopausal women with dyslipidemia: A double-blind, placebo-controlled, dose-ranging trial. J Clin Endocrinol Metab. 2007;92:4671-4677.

30. Kleemann A, Eckert S, von Eckardstein A, Lepper W, Schernikau U, Gleichmann U, et al. Effects of lovastatin on progression of non-dilated and dilated coronary segments and on restenosis in patients after ptca. The cholesterol lowering atherosclerosis ptca trial (clapt). Eur Heart J. 1999;20:1393-1406.

31. Baigent C, Landray MJ, Reith C, Emberson J, Wheeler DC, Tomson C, et al. The effects of lowering ldl cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (study of heart and renal protection): A randomised placebo-controlled trial. Lancet. 2011;377:2181-2192.

Page 23: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

Stroke 日本語版 Vol. 7, No. 330

スタチン療法と脳内出血のリスク31件の無作為化比較試験のメタ解析Statin Therapy and the Risk of Intracerebral HemorrhageA Meta-Analysis of 31 Randomized Controlled Trials

James S. McKinney, MD1; William J. Kostis, PhD, MD2

1 Cardiovascular Institute of New Jersey and Department of Neurology, University of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School, New Brunswick, NJ; and 2 Cardiology Division, Massachusetts General Hospital, Boston, MA.

Abstract

背景および目的:スタチン療法は虚血性脳卒中のリスクを低下させる。しかし,いくつかの試験で脳内出血(ICH)のリスク上昇が観察されている。この問題を検討するため,ICHが報告されたスタチンを用いた無作為化比較試験のメタ解析を行った。方法:2012 年 1 月 25 日までのMedline,Web of Science,およびCochrane Library の文献検索を行った。また,抽出した試験の一覧および以前のメタ解析のレビューから追加の無作為化比較試験を同定した。スタチン療法の無作為化比較試験で,ICHまたは出血性脳卒中が報告された全試験を対象とした。主要評価項目は ICHとした。31 件の無作為化比較試験が対象となった。すべての解析で変量効果モデルを使用し,いずれの解析でも不均一性は認められなかった。

結果:合計 91,588 例の被験者が実薬群に含まれ,91,215例が対照群に含まれた。実薬投与群と対照群の比較ではICHの発生率に有意差は認められなかった(OR,1.08;95% CI,0.88 ~ 1.32;p = 0.47)。ICHのリスクは,低比重リポ蛋白減少または低比重リポ蛋白コレステロール減少の達成程度と関連していなかった。全脳卒中(OR,0.84;95% CI,0.78 ~ 0.91;p < 0.0001)および全死因死亡(OR,0.92;CI,0.87 ~ 0.96;p = 0.0007)は実薬治療群で有意に少なかった。出版バイアスのエビデンスは認められなかった。結論:スタチン療法の無作為化比較試験 31 件のメタ解析では,スタチンの実薬治療は ICHの有意な増加と関連していなかった。スタチン療法では全脳卒中および全死因死亡の有意な減少が認められた。

Stroke 2012; 43: 2149-2156

図2 スタチンと脳内出血に関する無作為化比較試験の変量効果メタ解析のフォレストプロット。

ICH- スタチンのメタ解析:ICH

試験名試験内のサブグループ 各試験の統計値 脳卒中/合計 オッズ比および 95% CI

オッズ比 p 値 下限 上限 実薬 対照薬4S プラセボ 0.20 0.2989 0.01 4.17 0 / 2,221 2 / 2,223AF-TEXCAPS プラセボ 3.00 0.5014 0.12 73.62 1 / 3,304 0 / 3,301ALLHAT-LLT プラセボ 3.42 0.0158 1.26 9.27 17 / 5,170 5 / 5,185ASCOT-LLA プラセボ 0.55 0.1071 0.26 1.14 11 / 5,168 20 / 5,137ATOZ 低用量 12.84 0.0820 0.72 228.14 6 / 2,265 0 / 2,232AURORA プラセボ 1.19 0.5608 0.66 2.14 25 / 1,389 21 / 1,384CARE プラセボ 0.33 0.1774 0.07 1.65 2 / 2,081 6 / 2,078CORONA プラセボ 1.66 0.2308 0.72 3.80 15 / 2,514 9 / 2,497GISSI-P プラセボ 2.99 0.5019 0.12 73.55 1 / 2,138 0 / 2,133GREACE プラセボ 1.00 1.0000 0.06 16.02 1 / 800 1 / 800HPS プラセボ 0.96 0.8434 0.65 1.41 51 / 10,269 53 / 10,267JUPITER プラセボ 0.67 0.4415 0.24 1.87 6 / 8,901 9 / 8,901LIPID プラセボ 1.89 0.1236 0.84 4.24 17 / 4,512 9 / 4,502MEGA プラセボ 1.17 0.6632 0.57 2.41 16 / 3,866 14 / 3,966PROSPER プラセボ 0.81 0.6490 0.32 2.04 8 / 2,891 10 / 2,913PROVE-IT 低用量 3.94 0.2205 0.44 35.25 4 / 2,099 1 / 2,063TNT 低用量 0.94 0.8666 0.48 1.87 16 / 4,995 17 / 5,006SEARCH 低用量 0.96 0.8871 0.55 1.68 24 / 6,031 25 / 6,033SPARCL プラセボ 1.68 0.0191 1.09 2.60 55 / 2,365 33 / 2,366ASPEN プラセボ 1.98 0.4296 0.36 10.85 4 / 1,211 2 / 1,199GISSI-HF プラセボ 3.69 0.0454 1.03 13.23 11 / 2,285 3 / 2,2894D プラセボ 0.64 0.4299 0.21 1.96 5 / 619 8 / 633IDEAL 低用量 1.00 0.9969 0.32 3.11 6 / 4,439 6 / 4,449MIRACL プラセボ 0.14 0.1992 0.01 2.78 0 / 1,538 3 / 1,548PATE 低用量 0.60 0.1963 0.28 1.30 11 / 331 18 / 334ACAPS プラセボ 0.14 0.1965 0.01 2.75 0 / 460 3 / 459ALERT プラセボ 0.59 0.1817 0.27 1.28 10 / 1,050 17 / 1,052BONE プラセボ 0.74 0.8540 0.03 18.28 1 / 485 0 / 119CLAPT プラセボ 0.34 0.5060 0.01 8.34 0 / 112 1 / 114SHARP プラセボ 1.21 0.3916 0.78 1.87 45 / 4,650 37 / 4,620

1.08 0.4687 0.88 1.32

全試験

0.01 0.1 1 10 100実薬が優位 対照薬が優位

Page 24: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

113

Tratamiento con estatinas y riesgo de hemorragia intracerebral

Un metanálisis de 31 ensayos controlados y aleatorizadosJames S. McKinney, MD; William J. Kostis, PhD, MD

Recibido el 29 de febrero de 2012; aceptado el 30 de marzo de 2012.Cardiovascular Institute of New Jersey and the Department of Neurology (J.S.M.), University of Medicine and Dentistry of New Jersey–Robert Wood

Johnson Medical School, New Brunswick, NJ; y Cardiology Division (W.J.K.), Massachusetts General Hospital, Boston, MA.El suplemento de datos de este artículo, disponible solamente online, puede consultarse en http://stroke.ahajournals.org/lookup/suppl/

doi:10.1161/STROKEAHA.112.655894/-/DC1.Remitir la correspondencia a James S. McKinney, MD, Department of Neurology, UMDNJ-Robert Wood Johnson Medical School, 125 Paterson

Street, New Brunswick, NJ 08901. Correo electrónico [email protected]© 2012 American Heart Association, Inc.

Puede accederse a Stroke en http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.112.655894

Antecedentes y objetivo—El tratamiento con estatinas reduce el riesgo de ictus isquémico. En algunos estudios se ha ob-servado un aumento del riesgo de hemorragia intracerebral (HIC). Con objeto de investigar esta cuestión, llevamos a cabo un metanálisis de ensayos controlados y aleatorizados de uso de estatinas en los que se ha presentado información sobre la HIC.

Métodos—Realizamos una búsqueda bibliográfica en Medline, Web of Science y The Cochrane Library hasta el 25 de enero de 2012, e identificamos otros ensayos controlados y aleatorizados adicionales mediante el examen de las listas de bibliografía de los estudios identificados y de los metanálisis anteriores. Se incluyeron en el metanálisis todos los ensayos controlados y aleatorizados del tratamiento con estatinas en los que se presentaba información sobre la HIC o el ictus hemorrágico. La variable de valoración primaria fue la HIC. Se incluyeron 31 ensayos controlados y aleatorizados. Todos los análisis utilizaron modelos de efectos aleatorios y no se observó heterogeneidad en ninguno de ellos.

Resultados—Se incluyó a un total de 91.588 participantes en el grupo de tratamiento activo y a 91.215 en el grupo control. No se observaron diferencias de incidencia de HIC en el grupo de tratamiento activo frente al grupo control (OR, 1,08; IC del 95%, 0,88–1,32; p = 0,47). El riesgo de HIC no estaba relacionado con el grado de reducción de las lipoproteínas de baja densidad ni con el nivel alcanzado de colesterol de lipoproteínas de baja densidad. El total de ictus (OR, 0,84; IC del 95%, 0,78–0,91; p < 0,0001) y la mortalidad por todas las causas (OR, 0,92; IC, 0,87-0,96; p = 0,0007) presentaron una reducción significativa en el grupo de tratamiento activo. No se observó evidencia alguna de sesgo de publicación.

Conclusiones—El tratamiento activo con estatinas no se asoció a un aumento significativo de las HIC en este metanálisis de 31 ensayos controlados y aleatorizados del tratamiento con estatinas. Se observó una reducción significativa del total de ictus y de la mortalidad por todas las causas con el tratamiento de estatinas. (Traducido del inglés: Statin Thera-py and the Risk of Intracerebral Hemorrhage. A Meta-Analysis of 31 Randomized Controlled Trials. Stroke. 2012;43:2149-2156.)

Palabras clave: hemorrhagic stroke n intracerebral hemorrhage n meta-analysis n statin

La hipercolesterolemia se asocia a un aumento del riesgo de ictus isquémico1–5. El tratamiento hipolipemiante con

el empleo de inhibidores de la 3-hidroxi-3-metilglutaril-coen-zima A reductasa (estatinas) es eficaz para reducir la mortali-dad cardiovascular mediante la prevención del infarto agudo de miocardio y del ictus isquémico6–24. Es posible que el be-neficio clínico no se limite a las propiedades hipolipemiantes de las estatinas sino que sea consecuencia también de otros efectos “pleiotrópicos”, incluidos los antiinflamatorios y anti-trombóticos25. A pesar de las reducciones significativas de los ictus isquémicos que se han observado en los ensayos clínicos, en un estudio de cohorte realizado en una población amplia y en un metanálisis posterior no se ha observado una reducción de la mortalidad por ictus con el tratamiento con estatinas26,27.

El estudio Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) del tratamiento con dosis altas de atorvastatina en la prevención secundaria del ictus puso de manifiesto una reducción global significativa de las recurren-cias de ictus, pero no una diferencia en los ictus mortales22. En un análisis post hoc del ensayo SPARCL se observó que el tra-tamiento con atorvastatina se asociaba de manera independien-te con un aumento del riesgo de ictus hemorrágico (razón de riesgos, 1,68; IC del 95%, 1,09–2,59)28. Esto era especialmen-te apreciable en los individuos que se incorporaron al ensayo tras sufrir un ictus hemorrágico, en los que hubo un aumento > 5 veces en el riesgo de una recurrencia hemorrágica28.

Una revisión Cochrane de 2009 sobre el tratamiento hipo-lipemiante y el ictus indicó un aumento significativo de las

Page 25: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

114 Stroke Noviembre 2012

Figura 1. Perfil de la búsquedabibliográfica.

4.098 registros identificadosa través de la base

de datos

4.101 registros examinados

171 artículos completosexaminados en cuanto

a la elegibilidad

31 estudios incluidosen el metanálisis

140 estudios excluidos

50: ausencia de datos sobreel subtipo de ictus

53: informe duplicadode un ensayo

37: revisión/estudiono aleatorizado

3.930 excluidos

34 registros identificadosa través de otras fuentes

31 títulos duplicados

probabilidades de sufrir un ictus hemorrágico con el tratamien-to con estatinas (OR, 1,72; IC del 95%, 1,20–2,46)29. Sin em-bargo, este análisis incluyó tan solo 2 ensayos, el Heart Pro-tection Study (HPS) y el SPARCL. Un posterior metanálisis llevado a cabo por la Cholesterol Treatment Trialists’ (CTT) Collaboration, en el que se incluyeron 20 ensayos clínicos del tratamiento con estatinas en los que se presentaban datos so-bre la frecuencia de la hemorragia intracerebral (HIC), observó una tendencia no significativa al aumento del riesgo de ictus hemorrágico (riesgo relativo, 1,15; IC del 95%, 0,93–1,41; p = 0,2)30. Este análisis no mostró efecto alguno sobre la morta-lidad por ictus con el tratamiento de estatinas30. El estudio de la CTT Collaboration excluyó a los ensayos con un número de participantes < 1.000 y a los que tenían un periodo de segui-miento < 2 años30. Los objetivos del presente estudio fueron examinar el riesgo de ictus hemorrágico en pacientes tratados con estatinas, evaluar su efecto sobre el total de ictus y sobre la mortalidad por todas las causas, y determinar si la aparición de ictus hemorrágicos estaba relacionada con el cambio al-canzado mediante el tratamiento en las lipoproteínas de baja densidad (LDL). Llevamos a cabo un metanálisis en el que se incluyeron todos los ensayos controlados y aleatorizados dis-ponibles del tratamiento con estatinas en los que se presenta-ban datos sobre la frecuencia del ictus hemorrágico.

MétodosLos métodos utilizados en este metanálisis son similares a los des-critos anteriormente31. Realizamos una búsqueda bibliográfica hasta el 25 de enero de 2012, con el empleo de las bases de datos Medline, Web of Science y The Cochrane Library, para identificar los ensayos controlados y aleatorizados (ECA) del tratamiento con estatinas y la HIC o el ictus hemorrágico publicados en lengua inglesa. Los crite-

rios de búsqueda se presentan la Figura I del Suplemento de Datos Online. Identificamos otros ECA adicionales mediante el examen de las listas de bibliografía de los estudios identificados y de los me-tanálisis del tratamiento con estatinas publicados con anterioridad. Incluimos todos los ensayos que cumplían los siguientes criterios: participantes de edad ≥ 18 años, diseño controlado y aleatorizado, evaluación ciega de las variables de valoración y registro de datos sobre ictus hemorrágicos o HIC. Se incluyeron ensayos tanto de prevención primaria como de prevención secundaria, así como los ensayos de comparación del tratamiento activo con un tratamiento de control consistente en la “asistencia habitual”, un placebo o dosis inferiores de estatinas. Se incluyeron los estudios que presentaban las tasas de ictus hemorrágicos en el primer informe publicado, en publicaciones posteriores o en metanálisis. La valoración de los cri-terios de aceptación de todos los estudios la realizaron los autores, y las discrepancias se resolvieron mediante discusión y consenso. Con la aplicación de los criterios de búsqueda, se incluyeron en el análisis 31 ECA del tratamiento con estatinas que presentaban datos de HIC o ictus hemorrágico como variable de valoración (Figura 1).

Métodos estadísticos

Selección de los estudios

Extracción de los datos y evaluación de la calidadLos autores extrajeron los datos relativos a la estatina utilizada en el grupo de tratamiento activo, tipo de tratamiento (placebo, dosis bajas de estati-na o asistencia habitual) utilizado en el grupo control, criterios de inclu-sión, media de seguimiento y número de pacientes, total de ictus, HIC y muertes en los grupos de tratamiento activo y de control. Se evaluaron los datos relativos a la asignación aleatoria, la ocultación de la asignación, la comparación de las características basales, los criterios de elegibilidad definidos, el tipo de control, el enmascaramiento aplicado (pacientes, in-vestigadores, evaluación de la situación vital), el porcentaje de pérdida del seguimiento y el uso de un análisis por intención de tratar.

Page 26: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

McKinney y Kostis Mal Tratamiento con estatinas y riesgo de hemorragia intracerebral 115

Síntesis y análisis de los datosSe calcularon las tasas de HIC, total de ictus y mortalidad por todas las causas en los grupos de tratamiento activo y de control. Los aná-lisis estadísticos se realizaron con Comprehensive Meta-Analysis 2.232 y JMP 9.0 (SAS Institute, Cary, NC, EEUU). Se calcularon los

valores de OR y de IC del 95% para las tres tasas indicadas, utilizan-do un análisis por intención de tratar. Se calcularon los efectos del tratamiento combinando los datos ponderados para cada una de las tres tasas de episodios, utilizando modelos de efectos aleatorios. Se evaluó la heterogeneidad de los efectos con el empleo del estadístico

Tabla. Episodios cerebrovasculares y muerte

Ensayo, año

ParticipantesIncluidos

(N = 182.803)

Todos los ictus–Tratamiento

activoTodos los ictus

–Control

Ictus isquémico–Tratamiento

activo

Ictusisquémico–

ControlHIC–Tratamiento

activoHIC–

Control

Otros ictus–Tratamiento

activoOtros ictus–

Control

Ictus mortal–Tratamiento

activoIctus mortal–

Control

Mortalidad total–Tratamiento

activo

Mortalidad total–

Control

4S, 19946 4.444 44 64 29 49 0 2 15 13 14 12 182 256

AF-TEXCAPS,

19988

6.605 14 17 1 1 1 0 12 16 . . . . . . 80 77

ALLHAT-LLT,

200212

10.355 209 231 71 83 17 5 121 143 53 56 631 641

ASCOT-LLA,

200316

10.305 89 121 74 95 11 20 4 6 . . . . . . 185 212

A-to-Z,

200417

4.497 28 35 22 31 6 0 0 4 . . . . . . 130 104

AURORA,

200936

2.773 94 81 57 55 25 21 12 5 40 36 636 660

CARE, 19967 4.159 54 78 48 64 2 6 4 8 5 1 180 196

CORONA,

200737

5.011 126 145 73 90 15 9 15 16 35 39 728 759

GISSI-P,

200038

4.271 20 19 15 13 1 0 4 6 4 4 72 88

GREACE,

200214

1.600 9 17 . . . . . . 1 1 9 17 0 1 23 40

HPS, 200213 20.536 444 585 290 409 51 53 103 134 96 119 1.328 1.507

JUPITER,

200823

17.802 33 64 23 47 6 9 4 8 3 6 198 247

LIPID, 19989 9.014 224 272 200 255 17 9 7 8 22 27 717 888

MEGA,

200621

7.832 50 62 34 46 16 14 0 2 . . . . . . 55 79

PROSPER,

200215

5.804 135 131 91 88 8 10 36 33 22 14 298 305

PROVE-IT,

200418

4.162 21 19 10 12 4 1 7 6 . . . . . . 46 66

TNT, 200520 10.001 117 155 96 130 16 17 5 8 . . . . . . 284 282

SEARCH,

201024

12.064 255 279 233 255 24 25 0 0 57 67 964 970

SPARCL,

200622

4.731 265 311 218 274 55 33 7 12 24 41 216 211

CARDS,

200419

2.841 21 39 9 24 0 0 12 15 1 5 61 82

ASPEN,

200639

2.410 34 38 14 15 4 2 16 21 . . . . . . 70 68

GISSI-HF,

200840

4.574 82 66 63 53 11 3 8 10 38 29 657 644

4D, 200541 1.252 59 44 47 33 5 8 10 6 27 13 297 320

IDEAL, 200542 8.888 151 174 129 158 6 6 16 10 . . . . . . 366 374

MIRACL,

200110

3.086 12 24 . . . . . . 0 3 12 21 3 2 64 68

PATE, 200111 665 11 18 11 15 0 3 0 0 2 2 14 20

ACAPS,

199443

919 0 5 . . . . . . 0 3 0 2 . . . . . . 1 8

ALERT,

200344

2.102 93 91 67 66 10 17 5 5 11 10 143 138

BONE, 200745 604 1 0 . . . . . . 1 0 . . . . . . . . . . . . 0 0

CLAPT,

199946

226 0 1 . . . . . . 0 1 . . . . . . . . . . . . 0 2

SHARP,

201147

9.270 171 210 114 157 45 37 18 19 68 78 1.142 1.115

Total de todos los estudios (%)

2.866 (3,13) 3.396 (3,72) 2.039 (2,23) 2.518 (2,76) 358 (0,39) 318 (0,35) 462 (0,50) 554 (0,61) 525 (0,57) 562 (0,62) 9.768 (10,7) 10.427 (11,4)

HIC indica hemorragia intracerebral.

Page 27: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

116 Stroke Noviembre 2012

Q33. Se evaluó el sesgo de publicación mediante la elaboración de gráficos de embudo32, con los modelos de Trim and Fill de Duval y Tweedie33 y con los modelos de N de fail-safe de Rosenthal34 y Orwin35.

Análisis de sensibilidadSe realizaron varios análisis de sensibilidad adicionales. En primer lugar, 1 estudio (Collaborative Atorvastatin Diabetes Study [CARDS]) presentó 0 ictus hemorrágicos tanto en el grupo de atorvastatina como en el de placebo19,25. Por consi-guiente, con objeto de evitar la inclusión de una OR no defi-nida, el CARDS no se tuvo en cuenta en el análisis principal, en el que solamente se incluyeron los otros 30 estudios. Con objeto de examinar los efectos que tenía la exclusión de este estudio en el análisis principal, se llevó a cabo un análisis de sensibilidad mediante la imputación de 1 HIC (en vez de 0) en cada uno de los 2 grupos (tratamiento activo y control) del CARDS e incluyendo este estudio junto con los otros 30. Se realizó un segundo análisis de sensibilidad excluyendo 1 estudio (de los 30) cada vez, de forma iterativa. Esto se hizo para evaluar si los ensayos individuales podían haber influi-do de manera manifiesta en los resultados de este estudio.

En tercer lugar, se realizó un metanálisis acumulativo para investigar el grado en el que la OR estimada de ictus hemo-rrágico convergía con la adición iterativa de estudios progre-sivamente más grandes.

ResultadosSe identificaron 31 ensayos del tratamiento con estatinas en los que se incluyó en la asignación aleatoria del tratamien-to a un total de 182.803 participantes, y se incluyeron di-chos ensayos en el análisis6–24,36–47. De los 31 ECA, 6 estu-dios11,17,18,20,24,42 comparaban el tratamiento de dosis altas con el de dosis bajas de estatinas, y los demás 6–10,12–15,19,21–23,36–

41,43– 47 comparaban el tratamiento con estatinas con un place-bo o con la “asistencia habitual”. A un total de 91.588 parti-cipantes se les asignó aleatoriamente el tratamiento activo y a 91.215 el tratamiento de control. La mediana de duración del seguimiento fue de 46,8 meses.

Las características de los pacientes y de las LDL en los ECA incluidos en este análisis se presentan en las Tablas I y II del Suplemento de Datos Online. La media de edad fue de 62,6±5,2 años. Un total del 67,0% de los pacientes eran

Nombre delestudio Subgrupo en el estudio

Análisis estadístico paracada estudio Ictus / Total Odds ratio e IC del 95%

Oddsratio

Valorde p

Límiteinferior

Límitesuperior

Tratamientoactivo Control

4S PLACEBO 0,20 0,2989 0,01 4,17 0 / 2221 2 / 2223AF-TEXCAPS PLACEBO 3,00 0,5014 0,12 73,62 1 / 3304 0 / 3301ALLHAT-LLT PLACEBO 3,42 0,0158 1,26 9,27 17 / 5170 5 / 5185ASCOT-LLA PLACEBO 0,55 0,1071 0,26 1,14 11 / 5168 20 / 5137ATOZ DOSIS BAJA 12,84 0,0820 0,72 228,14 6 / 2265 0 / 2232AURORA PLACEBO 1,19 0,5608 0,66 2,14 25 / 1389 21 / 1384CARE PLACEBO 0,33 0,1774 0,07 1,65 2 / 2081 6 / 2078CORONA PLACEBO 1,66 0,2308 0,72 3,80 15 / 2514 9 / 2497GISSI-P PLACEBO 2,99 0,5019 0,12 73,55 1 / 2138 0 / 2133GREACE PLACEBO 1,00 1,0000 0,06 16,02 1 / 800 1 / 800HPS PLACEBO 0,96 0,8434 0,65 1,41 51 / 10269 53 / 10267JUPITER PLACEBO 0,67 0,4415 0,24 1,87 6 / 8901 9 / 8901LIPID PLACEBO 1,89 0,1236 0,84 4,24 17 / 4512 9 / 4502MEGA PLACEBO 1,17 0,6632 0,57 2,41 16 / 3866 14 / 3966PROSPER PLACEBO 0,81 0,6490 0,32 2,04 8 / 2891 10 / 2913PROVE-IT DOSIS BAJA 3,94 0,2205 0,44 35,25 4 / 2099 1 / 2063TNT DOSIS BAJA 0,94 0,8666 0,48 1,87 16 / 4995 17 / 5006SEARCH DOSIS BAJA 0,96 0,8871 0,55 1,68 24 / 6031 25 / 6033SPARCL PLACEBO 1,68 0,0191 1,09 2,60 55 / 2365 33 / 2366ASPEN PLACEBO 1,98 0,4296 0,36 10,85 4 / 1211 2 / 1199GISSI-HF PLACEBO 3,69 0,0454 1,03 13,23 11 / 2285 3 / 22894D PLACEBO 0,64 0,4299 0,21 1,96 5 / 619 8 / 633IDEAL DOSIS BAJA 1,00 0,9969 0,32 3,11 6 / 4439 6 / 4449MIRACL PLACEBO 0,14 0,1992 0,01 2,78 0 / 1538 3 / 1548PATE DOSIS BAJA 0,60 0,1963 0,28 1,30 11 / 331 18 / 334ACAPS PLACEBO 0,14 0,1965 0,01 2,75 0 / 460 3 / 459ALERT PLACEBO 0,59 0,1817 0,27 1,28 10 / 1050 17 / 1052BONE PLACEBO 0,74 0,8540 0,03 18,28 1 / 485 0 / 119CLAPT PLACEBO 0,34 0,5060 0,01 8,34 0 / 112 1 / 114SHARP PLACEBO 1,21 0,3916 0,78 1,87 45 / 4650 37 / 4620

1,08 0,4687 0,88 1,32

0,01 0,1 1 10 100

Favorable altratamiento activo

Favorableal control

Metanálisis de HIC - Estatinas: HIC

Todos los estudiosFigura 2. Gráfico de Forrest del metanálisis con efectos aleatorios de ensayos aleatorizados de las estatinas y la hemorragia intracerebral.

Page 28: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

McKinney y Kostis Mal Tratamiento con estatinas y riesgo de hemorragia intracerebral 117

varones y un 78,1% eran blancos. Tan solo en un 11,0% de los participantes se habían documentado unos antecedentes previos de ictus. En promedio el 24,7% tenían diabetes, el 53,0% hipertensión, el 59,1% enfermedades cardiovasculares y el 21,2% eran fumadores activos de cigarrillos en el mo-mento de la inclusión en el estudio. Un total del 61,0% de los participantes fueron tratados con ácido acetilsalicílico o anticoagulantes orales. Los episodios de ictus se presentan en la Tabla.

Hemorragia intracerebralSe produjo una HIC en 358 participantes (0,39%) del grupo de tratamiento activo frente a 318 (0,35%) del grupo control. En el análisis principal de evaluación del riesgo de HIC en 30 estudios del tratamiento con estatinas, el tratamiento activo no se asoció a un aumento de las HIC (OR, 1,08; IC del 95%, 0,88–1,32; p = 0,47; Figura 2). El valor de p para la hetero-geneidad (interacción) fue de 0,38, Q = 30,705 y grados de libertad = 29. No hubo diferencias estadísticamente signifi-cativas en el riesgo de HIC al estratificar el metanálisis según el tipo de grupo control o los estudios de prevención primaria

frente a los de prevención secundaria (Suplemento de Datos Online, Figura II).

El análisis de sensibilidad en el que se incluyó la totalidad de los 31 estudios (incluido el CARDS) produjo unos resul-tados casi idénticos a los del análisis principal (OR, 1,08; IC del 95%, 0,88–1,31; p = 0,46). En la totalidad de las 30 ite-raciones del análisis de sensibilidad realizado excluyendo 1 estudio cada vez, no se observó asociación alguna entre la HIC y el tratamiento activo. La estimación puntual de la OR osciló entre 1,04 y 1,11, con un límite inferior de 0,84 a 0,91 y un valor de p de 0,27 a 0,77. En el metanálisis acumulativo, no hubo asociación alguna entre la HIC y el tratamiento ac-tivo en 28 de 30 iteraciones. La estimación puntual de la OR osciló entre 0,20 (IC del 95%, 0,01–4,17; p = 0,30) y 1,90 (IC del 95%, 0,40–8,95; p = 0,42).

No observamos evidencia alguna de sesgo de publicación puesto que el gráfico de embudo era simétrico. La OR y el IC del efecto global (OR, 1,08; IC del 95%, 0,88–1,31) se mantuvieron inalterados al aplicar el método de Trim and Fill de Duval y Tweedie33. La N fail-safe de Orwin reveló que serían necesarios otros 25 estudios adicionales con una media

4S PLACEBO 0,68 0,0533 0,46 1,01 44 / 2221 64 / 2223AF-TEXCAPS PLACEBO 0,82 0,5880 0,40 1,67 14 / 3304 17 / 3301ALLHAT-LLT PLACEBO 0,90 0,2982 0,75 1,09 209 / 5170 231 / 5185ASCOT-LLA PLACEBO 0,73 0,0234 0,55 0,96 89 / 5168 121 / 5137ATOZ DOSIS BAJA 0,79 0,3448 0,48 1,30 28 / 2265 35 / 2232AURORA PLACEBO 1,17 0,3223 0,86 1,59 94 / 1389 81 / 1384CARE PLACEBO 0,68 0,0340 0,48 0,97 54 / 2081 78 / 2078CORONA PLACEBO 0,86 0,2138 0,67 1,09 126 / 2514 145 / 2497GISSI-P PLACEBO 1,05 0,8780 0,56 1,97 20 / 2138 19 / 2133GREACE PLACEBO 0,52 0,1197 0,23 1,18 9 / 800 17 / 800HPS PLACEBO 0,75 0,0000 0,66 0,85 444 / 10269 585 / 10267JUPITER PLACEBO 0,51 0,0019 0,34 0,78 33 / 8901 64 / 8901LIPID PLACEBO 0,81 0,0251 0,68 0,97 224 / 4512 272 / 4502MEGA PLACEBO 0,83 0,3151 0,57 1,20 50 / 3866 62 / 3966PROSPER PLACEBO 1,04 0,7533 0,81 1,33 135 / 2891 131 / 2913PROVE-IT DOSIS BAJA 1,09 0,7928 0,58 2,03 21 / 2099 19 / 2063TNT DOSIS BAJA 0,75 0,0209 0,59 0,96 117 / 4995 155 / 5006SEARCH DOSIS BAJA 0,91 0,2900 0,77 1,08 255 / 6031 279 / 6033SPARCL PLACEBO 0,83 0,0416 0,70 0,99 265 / 2365 311 / 2366CARDS PLACEBO 0,53 0,0184 0,31 0,90 21 / 1429 39 / 1412ASPEN PLACEBO 0,88 0,6022 0,55 1,41 34 / 1211 38 / 1199GISSI-HF PLACEBO 1,25 0,1786 0,90 1,74 82 / 2285 66 / 22894D PLACEBO 1,41 0,0979 0,94 2,12 59 / 619 44 / 633IDEAL DOSIS BAJA 0,87 0,2012 0,69 1,08 151 / 4439 174 / 4449MIRACL PLACEBO 0,50 0,0507 0,25 1,00 12 / 1538 24 / 1548PATE DOSIS BAJA 0,60 0,1963 0,28 1,30 11 / 331 18 / 334ACAPS PLACEBO 0,09 0,1030 0,00 1,63 0 / 460 5 / 459ALERT PLACEBO 1,03 0,8667 0,76 1,39 93 / 1050 91 / 1052BONE PLACEBO 0,74 0,8540 0,03 18,28 1 / 485 0 / 119CLAPT PLACEBO 0,34 0,5060 0,01 8,34 0 / 112 1 / 114SHARP PLACEBO 0,80 0,0356 0,65 0,99 171 / 4650 210 / 4620

0,84 0,0000 0,78 0,91

0,5 1 2

Metanálisis de HIC - Estatinas: Total de ictus

Todos los estudios Figura 3. Gráfico de Forrest del metanálisis con efectos aleatorios de ensayos aleatorizados de las estatinas y el total de ictus.

Nombre delestudio Subgrupo en el estudio

Análisis estadístico paracada estudio Ictus / Total Odds ratio e IC del 95%

Oddsratio

Valorde p

Límiteinferior

Límitesuperior

Tratamientoactivo Control

Favorable altratamiento activo

Favorableal control

Page 29: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

118 Stroke Noviembre 2012

de OR de 1,0 para producir un aumento del 5% (OR, 1,05) al nivel de p = 0,05.

Se llevó a cabo una metarregresión para estudiar la relación entre el efecto del tratamiento con estatinas sobre las LDL y el riesgo de HIC. No hubo relación alguna entre los efectos del tratamiento activo en comparación con el de control y las determinaciones del colesterol de LDL. Concretamente, no se observaron relaciones significativas del log OR (activo/con-trol) de cada estudio respecto a (1) la diferencia (activo menos control) de la reducción de colesterol de LDL (diferencia entre valor basal y valor en el seguimiento) en mg/dL (pendiente, 0,0043; EE, 0,0054; IC del 95%, -0,4831 a 0,0149; p = 0,43]; (2) esta diferencia expresada como porcentaje de valor basal (pendiente, 0,0054; EE, 0,0075; IC del 95%, -0,0092 a 0,0200; p = 0,47]; y (3) el valor de colesterol de LDL alcanzado en el grupo de tratamiento activo durante el seguimiento (pendiente, -0,0049; EE, 0,0043; IC del 95%, -0,0133 a 0,0035; p = 0,26).

Total de ictusSe produjeron en total 6.262 ictus en este metanálisis. La tasa global de ictus fue del 3,13% en el grupo de tratamiento acti-vo frente al 3,72% en el grupo control. El tratamiento activo

produjo una reducción significativa en el número total de ic-tus (OR, 0,84; IC del 95%, 0,78–0,91; p < 0,0001; Figura 3). El valor de p para la heterogeneidad (interacción) fue de 0,31, Q = 33,315 y grados de libertad = 30. El número necesario a tratar con la medicación activa de estatinas para prevenir un ictus de cualquier tipo durante el seguimiento del estudio fue de 200 (reducción del riesgo absoluto = 0,5%, p < 0,0001).

Mortalidad por todas las causasSe registraron 20.195 muertes en los 31 ensayos incluidos en este análisis. Hubo una tasa de mortalidad por todas las causas significativamente inferior en el grupo de tratamiento activo (10,67%) en comparación con el grupo control (11,43%; OR, 0,92; IC del 95%, 0,87–0,96; p = 0,0007; Figura 4). El valor de p para la heterogeneidad (interacción) fue de 0,31, Q = 32,272 y grados de libertad = 29. El número necesario a tratar con la medicación activa de estatinas para prevenir 1 muerte fue de 167 (reducción del riesgo absoluto = 0,6%, p < 0,0001).

DiscusiónLos estudios epidemiológicos han descrito un aumento de las tasas de ictus hemorrágicos y de mortalidad debida a HIC

4S PLACEBO 0,69 0,0002 0,56 0,84 182 / 2221 256 / 2223AF-TEXCAPS PLACEBO 1,04 0,8130 0,76 1,43 80 / 3304 77 / 3301ALLHAT-LLT PLACEBO 0,99 0,8071 0,88 1,11 631 / 5170 641 / 5185ASCOT-LLA PLACEBO 0,86 0,1493 0,71 1,05 185 / 5168 212 / 5137ATOZ DOSIS BAJA 1,25 0,1036 0,96 1,62 130 / 2265 104 / 2232AURORA PLACEBO 0,93 0,3162 0,80 1,08 636 / 1389 660 / 1384CARE PLACEBO 0,91 0,3791 0,74 1,12 180 / 2081 196 / 2078CORONA PLACEBO 0,93 0,2650 0,83 1,05 728 / 2514 759 / 2497GISSI-P PLACEBO 0,81 0,1928 0,59 1,11 72 / 2138 88 / 2133GREACE PLACEBO 0,56 0,0309 0,33 0,95 23 / 800 40 / 800HPS PLACEBO 0,86 0,0003 0,80 0,93 1328 / 10269 1507 / 10267JUPITER PLACEBO 0,80 0,0189 0,66 0,96 198 / 8901 247 / 8901LIPID PLACEBO 0,77 0,0000 0,69 0,86 717 / 4512 888 / 4502MEGA PLACEBO 0,71 0,0532 0,50 1,00 55 / 3866 79 / 3966PROSPER PLACEBO 0,98 0,8393 0,83 1,16 298 / 2891 305 / 2913PROVE-IT DOSIS BAJA 0,68 0,0458 0,46 0,99 46 / 2099 66 / 2063TNT DOSIS BAJA 1,01 0,9096 0,85 1,20 284 / 4995 282 / 5006SEARCH DOSIS BAJA 0,99 0,8879 0,90 1,09 964 / 6031 970 / 6033SPARCL PLACEBO 1,03 0,7962 0,84 1,25 216 / 2365 211 / 2366CARDS PLACEBO 0,72 0,0617 0,51 1,02 61 / 1429 82 / 1412ASPEN PLACEBO 1,02 0,9084 0,72 1,44 70 / 1211 68 / 1199GISSI-HF PLACEBO 1,03 0,6431 0,91 1,17 657 / 2285 644 / 22894D PLACEBO 0,90 0,3628 0,72 1,13 297 / 619 320 / 633IDEAL DOSIS BAJA 0,98 0,7832 0,84 1,14 366 / 4439 374 / 4449MIRACL PLACEBO 0,95 0,7507 0,67 1,34 64 / 1538 68 / 1548PATE DOSIS BAJA 0,69 0,3056 0,34 1,40 14 / 331 20 / 334ACAPS PLACEBO 0,12 0,0485 0,02 0,99 1 / 460 8 / 459ALERT PLACEBO 1,04 0,7357 0,81 1,34 143 / 1050 138 / 1052CLAPT PLACEBO 0,20 0,3006 0,01 4,21 0 / 112 2 / 114SHARP PLACEBO 1,02 0,6336 0,93 1,13 1142 / 4650 1115 / 4620

0,92 0,0007 0,87 0,96

0,5 1 2

Metanálisis de HIC - Estatinas: Muerte

Todos los estudios Figura 4. Gráfico de Forrest del metanálisis con efectos aleatorios de ensayos aleatorizados de las estatinas y la mortalidad por todas las causas.

Favorable altratamiento activo

Favorableal control

Nombre delestudio Subgrupo en el estudio

Análisis estadístico paracada estudio Ictus / Total Odds ratio e IC del 95%

Oddsratio

Valorde p

Límiteinferior

Límitesuperior

Tratamientoactivo Control

Page 30: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

McKinney y Kostis Mal Tratamiento con estatinas y riesgo de hemorragia intracerebral 119

en poblaciones con niveles de colesterol bajos1,4,48,49. Se ha planteado la hipótesis de que el colesterol puede ser impor-tante para la integridad de la pared cerebrovascular y que los niveles bajos pueden elevar el riesgo de ruptura de los vasos sanguíneos y de HIC50. Además, varios estudios han descrito una asociación del ictus hemorrágico con el uso de estatinas.

El ensayo SPARCL de atorvastatina en dosis altas para la prevención secundaria del ictus describió un exceso de HIC con el tratamiento activo en comparación con placebo (55 frente a 33; p = 0,02)22. De igual modo, hubo un aumento al doble en los ictus hemorrágicos en el estudio HPS en los pacientes que habían sufrido un ictus previo y fueron trata-dos con simvastatina51. En tres metanálisis previos del tra-tamiento con estatinas no se ha observado un aumento del riesgo de ictus hemorrágico25,30,52. En un metanálisis de 11 ensayos, Amarenco y Labreuche25 describieron un riesgo de ictus hemorrágico desdeñable con el tratamiento de estatinas (riesgo relativo, 1,03; IC del 95%, 0,75–1,41). En su estudio de la reducción intensiva de las LDL con estatinas, la CTT Collaboration llevó a cabo un metanálisis de 26 ECA y des-cribió una tendencia no significativa al aumento del riesgo de HIC (riesgo relativo, 1,15; IC del 95%, 0,93–1,41)30. En un metanálisis más reciente en el que se incluyeron 23 ECA, no se observó evidencia alguna de aumento del riesgo de HIC53.

Nosotros hemos realizado un metanálisis exhaustivo de los ensayos clínicos aleatorizados realizados anteriormente sobre el tratamiento con estatinas en los que se presentaron datos relativos a la HIC o el ictus hemorrágico. Incluimos 31 ECA con un total de 182.803 participantes. En nuestro análisis, el tratamiento activo se asoció a un aumento no significativo del riesgo de HIC (OR, 1,08; IC del 95%, 0,88–1,32).

El pequeño exceso de HIC no significativo observado no estaba relacionado con los efectos del tratamiento con estati-nas sobre las LDL. No observamos ninguna relación entre el nivel de LDL alcanzado o el grado de reducción de las LDL y el riesgo de ictus hemorrágico en este análisis. Esto sugie-re que todo posible aumento del riesgo de HIC podría ser atribuible a otros efectos de las estatinas distintos de los que ejercen sobre las LDL. Esto concuerda con lo indicado por un gran metanálisis de estudios de cohorte observacionales en el que no se observó correlación alguna con los ictus mortales y el colesterol total basal en casi 900.000 participantes27. Ade-más de sus propiedades hipolipemiantes, las estatinas pueden tener propiedades antitrombóticas como consecuencia de la inhibición de la agregación plaquetaria y la potenciación de la fibrinólisis54,55. Los efectos antitrombóticos de las estati-nas podrían explicar un teórico aumento de las complicacio-nes hemorrágicas.

Al estratificar nuestro análisis según el tipo de prevención, hubo una tendencia no significativa al aumento de las proba-bilidades de HIC en los estudios de prevención secundaria (OR, 1,26; IC del 95%, 0,91–1,73) en comparación con los de prevención primaria (OR, 0,96; IC del 95%, 0,75–1,23). El presente estudio no se limitó a los ensayos del tratamiento con estatinas para la prevención secundaria del ictus. El aná-lisis post hoc de los ensayos HPS y SPARCL sugiere que los pacientes con un ictus previo pueden presentar un especial aumento del riesgo de ictus hemorrágico28,51. Sin embargo, en un amplio estudio de cohorte retrospectivo reciente no se

observó evidencia alguna de aumento del riesgo de HIC en pacientes tratados con estatinas después de sufrir un ictus is-quémico52. Westover y cols.56, con el empleo de un enfoque de análisis de decisión, observaron que la evitación de las estatinas, sobre todo en pacientes con antecedentes de hemo-rragia lobular (y que tienen un riesgo elevado de recurrencia hemorrágica), era favorable respecto a una amplia variedad de parámetros clínicos. Nuestro análisis no evaluó el riesgo del tratamiento con estatinas en este subgrupo de pacientes con una HIC previa. Teniendo en cuenta los datos actuales, es preciso tener precaución al tratar a estos pacientes con es-tatinas, y serán necesarias nuevas investigaciones al respecto.

Este análisis tiene varias limitaciones. En primer lugar, no tuvimos acceso a datos de los pacientes individuales y es po-sible que haya variables de estos no tenidas en cuenta que in-fluyeran en las tasas de ictus hemorrágico. En segundo lugar, ictus hemorrágico es un término nebuloso que puede incluir la HIC, la hemorragia subaracnoidea, subdural o epidural, o la transformación hemorrágica de un ictus isquémico. Aun-que la HIC es el trastorno de interés, es posible que en ECA previos que han presentado las tasas de ictus hemorrágicos se incluyeran algunas de otras de estas causas de hemorragia intracraneal. Finalmente, aunque hemos realizado el análisis más exhaustivo publicado hasta el momento del tratamiento con estatinas y el riesgo de HIC en ensayos aleatorizados, es posible que hubiera otros ensayos, en especial los publica-dos en lenguas distintas del inglés, que quedaran excluidos. Aunque incluimos todos los ensayos pertinentes, con inde-pendencia de su tamaño, en nuestro análisis no se observó ningún sesgo de publicación ni heterogeneidad.

El tratamiento con estatinas no se asoció a un aumento sig-nificativo del riesgo de HIC. No hubo efecto alguno sobre el riesgo de HIC asociado al grado de reducción de las LDL o al nivel de LDL alcanzado. La reducción significativa del total de ictus y de la mortalidad por todas las causas compensó sobradamente todo ligero aumento del riesgo de HIC que pu-diera haber. Estos resultados respaldan las recomendaciones actuales para la prescripción de estatinas en los pacientes en los que, por lo demás, son apropiados.

ConclusionesEn este metanálisis de 31 ECA, el tratamiento con estatinas no se asoció a un aumento significativo de las probabilida-des de HIC. Sin embargo, el tratamiento con estatinas sí se asocia a reducciones significativas del total de ictus y de las muertes, lo cual anula todo posible riesgo hemorrágico en una población de pacientes no seleccionada.

Fuentes de financiaciónEste estudio fue financiado en parte por la Robert Wood Johnson Foundation y la Schering-Plough Foundation.

DeclaracionesNinguna.

Bibliografía1. Iso H, Jacobs DR Jr, Wentworth D, Neaton JD, Cohen JD. Serum

cholesterol levels and six-year mortality from stroke in 350 977 menscreened for the multiple risk factor intervention trial. N Engl J Med.1989;320:904–910.

2. Leppala JM, Virtamo J, Fogelholm R, Albanes D, Heinonen OP. Dif-ferent risk factors for different stroke subtypes: association of bloodpressure, cholesterol, and antioxidants. Stroke. 1999;30:2535–2540.

3. Horenstein RB, Smith DE, Mosca L. Cholesterol predicts stroke mortalityin the women’s pooling project. Stroke. 2002;33:1863–1868.

4. Zhang X, Patel A, Horibe H, Wu Z, Barzi F, Rodgers A, et al. Choles-terol, coronary heart disease, and stroke in the Asia Pacific region. Int JEpidemiol. 2003;32:563–572.

5. Kurth T, Everett BM, Buring JE, Kase CS, Ridker PM, Gaziano JM.Lipid levels and the risk of ischemic stroke in women. Neurology.2007;68:556–562.

6. Randomised trial of cholesterol lowering in 4444 patients with coronaryheart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet.1994;344:1383–1389.

7. Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG,et al. The effect of pravastatin on coronary events after myocardialinfarction in patients with average cholesterol levels. Cholesterol andrecurrent events trial investigators. N Engl J Med. 1996;335:1001–1009.

8. Downs JR, Clearfield M, Weis S, Whitney E, Shapiro DR, Beere PA, etal. Primary prevention of acute coronary events with lovastatin in menand women with average cholesterol levels: results of AFCAPS/

Page 31: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

120 Stroke Noviembre 20121. Iso H, Jacobs DR Jr, Wentworth D, Neaton JD, Cohen JD. Serumcholesterol levels and six-year mortality from stroke in 350 977 menscreened for the multiple risk factor intervention trial. N Engl J Med.1989;320:904–910.

2. Leppala JM, Virtamo J, Fogelholm R, Albanes D, Heinonen OP. Dif-ferent risk factors for different stroke subtypes: association of bloodpressure, cholesterol, and antioxidants. Stroke. 1999;30:2535–2540.

3. Horenstein RB, Smith DE, Mosca L. Cholesterol predicts stroke mortalityin the women’s pooling project. Stroke. 2002;33:1863–1868.

4. Zhang X, Patel A, Horibe H, Wu Z, Barzi F, Rodgers A, et al. Choles-terol, coronary heart disease, and stroke in the Asia Pacific region. Int JEpidemiol. 2003;32:563–572.

5. Kurth T, Everett BM, Buring JE, Kase CS, Ridker PM, Gaziano JM.Lipid levels and the risk of ischemic stroke in women. Neurology.2007;68:556–562.

6. Randomised trial of cholesterol lowering in 4444 patients with coronaryheart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet.1994;344:1383–1389.

7. Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG,et al. The effect of pravastatin on coronary events after myocardialinfarction in patients with average cholesterol levels. Cholesterol andrecurrent events trial investigators. N Engl J Med. 1996;335:1001–1009.

8. Downs JR, Clearfield M, Weis S, Whitney E, Shapiro DR, Beere PA, etal. Primary prevention of acute coronary events with lovastatin in menand women with average cholesterol levels: results of AFCAPS/

TEXCAPS. Air Force/Texas Coronary Atherosclerosis Prevention Study.JAMA. 1998;279:1615–1622.

9. Prevention of cardiovascular events and death with pravastatin in patientswith coronary heart disease and a broad range of initial cholesterol levels.The Long-Term Intervention With Pravastatin in Ischaemic Disease(LIPID) study group. N Engl J Med. 1998;339:1349–1357.

10. Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters D,et al. Effects of atorvastatin on early recurrent ischemic events in acutecoronary syndromes: the MIRACL study: a randomized controlled trial.JAMA. 2001;285:1711–1718.

11. Ito H, Ouchi Y, Ohashi Y, Saito Y, Ishikawa T, Nakamura H, et al. Acomparison of low versus standard dose pravastatin therapy for theprevention of cardiovascular events in the elderly: the Pravastatin Anti-Atherosclerosis Trial in the Elderly (PATE). J Atheroscler Thromb.2001;8:33–44.

12. Major outcomes in moderately hypercholesterolemic, hypertensive patientsrandomized to pravastatin vs usual care: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT). JAMA.2002;288:2998–3007.

13. MRC/BHF Heart Protection Study of cholesterol lowering with simva-statin in 20 536 high-risk individuals: a randomised placebo-controlledtrial. Lancet. 2002;360:7–22.

14. Athyros VG, Papageorgiou AA, Mercouris BR, Athyrou VV, SymeonidisAN, Basayannis EO, et al. Treatment with atorvastatin to the nationalcholesterol educational program goal versus ‘usual’ care in secondarycoronary heart disease prevention. The Greek Atorvastatin and Coronary-Heart-Disease Evaluation (GREACE) study. Curr Med Res Opin. 2002;18:220–228.

15. Shepherd J, Blauw GJ, Murphy MB, Bollen EL, Buckley BM, Cobbe SM,et al. Pravastatin in Elderly Individuals at Risk of Vascular Disease(PROSPER): a randomised controlled trial. Lancet. 2002;360:1623–1630.

16. Sever PS, Dahlof B, Poulter NR, Wedel H, Beevers G, Caulfield M, et al.Prevention of coronary and stroke events with atorvastatin in hyper-tensive patients who have average or lower-than-average cholesterolconcentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial–LipidLowering Arm (ASCOT-LLA): a multicentre randomised controlled trial.Lancet. 2003;361:1149–1158.

17. de Lemos JA, Blazing MA, Wiviott SD, Lewis EF, Fox KA, White HD,et al. Early intensive vs a delayed conservative simvastatin strategy inpatients with acute coronary syndromes: phase Z of the A to Z Trial.JAMA. 2004;292:1307–1316.

18. Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder R,et al. Intensive versus moderate lipid lowering with statins after acutecoronary syndromes. N Engl J Med. 2004;350:1495–1504.

19. Colhoun HM, Betteridge DJ, Durrington PN, Hitman GA, Neil HA,Livingstone SJ, et al. Primary prevention of cardiovascular disease withatorvastatin in type 2 diabetes in the Collaborative Atorvastatin DiabetesStudy (CARDS): multicentre randomised placebo-controlled trial.Lancet. 2004;364:685–696.

20. LaRosa JC, Grundy SM, Waters DD, Shear C, Barter P, Fruchart JC, etal. Intensive lipid lowering with atorvastatin in patients with stablecoronary disease. N Engl J Med. 2005;352:1425–1435.

21. Nakamura H, Arakawa K, Itakura H, Kitabatake A, Goto Y, Toyota T, etal. Primary prevention of cardiovascular disease with pravastatin in Japan(Mega study): a prospective randomised controlled trial. Lancet. 2006;368:1155–1163.

22. Amarenco P, Bogousslavsky J, Callahan A III, Goldstein LB, HennericiM, Rudolph AE, et al. High-dose atorvastatin after stroke or transientischemic attack. N Engl J Med. 2006;355:549–559.

23. Ridker PM, Danielson E, Fonseca FA, Genest J, Gotto AM Jr, KasteleinJJ, et al. Rosuvastatin to prevent vascular events in men and women withelevated C-reactive protein. N Engl J Med. 2008;359:2195–2207.

24. Armitage J, Bowman L, Wallendszus K, Bulbulia R, Rahimi K, HaynesR, et al. Intensive lowering of LDL cholesterol with 80 mg versus 20 mgsimvastatin daily in 12 064 survivors of myocardial infarction: adouble-blind randomised trial. Lancet. 2010;376:1658–1669.

25. Amarenco P, Labreuche J. Lipid management in the prevention of stroke:review and updated meta-analysis of statins for stroke prevention. LancetNeurol. 2009;8:453–463.

26. Cholesterol, diastolic blood pressure, and stroke: 13 000 strokes in450 000 people in 45 prospective cohorts. Prospective studies collabo-ration. Lancet. 1995;346:1647–1653.

27. Lewington S, Whitlock G, Clarke R, Sherliker P, Emberson J, Halsey J,et al. Blood cholesterol and vascular mortality by age, sex, and blood

TEXCAPS. Air Force/Texas Coronary Atherosclerosis Prevention Study.JAMA. 1998;279:1615–1622.

9. Prevention of cardiovascular events and death with pravastatin in patientswith coronary heart disease and a broad range of initial cholesterol levels.The Long-Term Intervention With Pravastatin in Ischaemic Disease(LIPID) study group. N Engl J Med. 1998;339:1349–1357.

10. Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters D,et al. Effects of atorvastatin on early recurrent ischemic events in acutecoronary syndromes: the MIRACL study: a randomized controlled trial.JAMA. 2001;285:1711–1718.

11. Ito H, Ouchi Y, Ohashi Y, Saito Y, Ishikawa T, Nakamura H, et al. Acomparison of low versus standard dose pravastatin therapy for theprevention of cardiovascular events in the elderly: the Pravastatin Anti-Atherosclerosis Trial in the Elderly (PATE). J Atheroscler Thromb.2001;8:33–44.

12. Major outcomes in moderately hypercholesterolemic, hypertensive patientsrandomized to pravastatin vs usual care: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT). JAMA.2002;288:2998–3007.

13. MRC/BHF Heart Protection Study of cholesterol lowering with simva-statin in 20 536 high-risk individuals: a randomised placebo-controlledtrial. Lancet. 2002;360:7–22.

14. Athyros VG, Papageorgiou AA, Mercouris BR, Athyrou VV, SymeonidisAN, Basayannis EO, et al. Treatment with atorvastatin to the nationalcholesterol educational program goal versus ‘usual’ care in secondarycoronary heart disease prevention. The Greek Atorvastatin and Coronary-Heart-Disease Evaluation (GREACE) study. Curr Med Res Opin. 2002;18:220–228.

15. Shepherd J, Blauw GJ, Murphy MB, Bollen EL, Buckley BM, Cobbe SM,et al. Pravastatin in Elderly Individuals at Risk of Vascular Disease(PROSPER): a randomised controlled trial. Lancet. 2002;360:1623–1630.

16. Sever PS, Dahlof B, Poulter NR, Wedel H, Beevers G, Caulfield M, et al.Prevention of coronary and stroke events with atorvastatin in hyper-tensive patients who have average or lower-than-average cholesterolconcentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial–LipidLowering Arm (ASCOT-LLA): a multicentre randomised controlled trial.Lancet. 2003;361:1149–1158.

17. de Lemos JA, Blazing MA, Wiviott SD, Lewis EF, Fox KA, White HD,et al. Early intensive vs a delayed conservative simvastatin strategy inpatients with acute coronary syndromes: phase Z of the A to Z Trial.JAMA. 2004;292:1307–1316.

18. Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder R,et al. Intensive versus moderate lipid lowering with statins after acutecoronary syndromes. N Engl J Med. 2004;350:1495–1504.

19. Colhoun HM, Betteridge DJ, Durrington PN, Hitman GA, Neil HA,Livingstone SJ, et al. Primary prevention of cardiovascular disease withatorvastatin in type 2 diabetes in the Collaborative Atorvastatin DiabetesStudy (CARDS): multicentre randomised placebo-controlled trial.Lancet. 2004;364:685–696.

20. LaRosa JC, Grundy SM, Waters DD, Shear C, Barter P, Fruchart JC, etal. Intensive lipid lowering with atorvastatin in patients with stablecoronary disease. N Engl J Med. 2005;352:1425–1435.

21. Nakamura H, Arakawa K, Itakura H, Kitabatake A, Goto Y, Toyota T, etal. Primary prevention of cardiovascular disease with pravastatin in Japan(Mega study): a prospective randomised controlled trial. Lancet. 2006;368:1155–1163.

22. Amarenco P, Bogousslavsky J, Callahan A III, Goldstein LB, HennericiM, Rudolph AE, et al. High-dose atorvastatin after stroke or transientischemic attack. N Engl J Med. 2006;355:549–559.

23. Ridker PM, Danielson E, Fonseca FA, Genest J, Gotto AM Jr, KasteleinJJ, et al. Rosuvastatin to prevent vascular events in men and women withelevated C-reactive protein. N Engl J Med. 2008;359:2195–2207.

24. Armitage J, Bowman L, Wallendszus K, Bulbulia R, Rahimi K, HaynesR, et al. Intensive lowering of LDL cholesterol with 80 mg versus 20 mgsimvastatin daily in 12 064 survivors of myocardial infarction: adouble-blind randomised trial. Lancet. 2010;376:1658–1669.

25. Amarenco P, Labreuche J. Lipid management in the prevention of stroke:review and updated meta-analysis of statins for stroke prevention. LancetNeurol. 2009;8:453–463.

26. Cholesterol, diastolic blood pressure, and stroke: 13 000 strokes in450 000 people in 45 prospective cohorts. Prospective studies collabo-ration. Lancet. 1995;346:1647–1653.

27. Lewington S, Whitlock G, Clarke R, Sherliker P, Emberson J, Halsey J,et al. Blood cholesterol and vascular mortality by age, sex, and blood

pressure: a meta-analysis of individual data from 61 prospective studieswith 55 000 vascular deaths. Lancet. 2007;370:1829–1839.

28. Goldstein MR, Mascitelli L, Pezzetta F. Hemorrhagic stroke in thestroke prevention by Aggressive Reduction in Cholesterol LevelsStudy. Neurology. 2009;72:1448; author reply 1448 –1449.

29. Manktelow BN, Potter JF. Interventions in the management of serumlipids for preventing stroke recurrence. Cochrane Database Syst Rev.2009;3:CD002091.

30. Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, et al.Efficacy and safety of more intensive lowering of LDL cholesterol: ameta-analysis of data from 170 000 participants in 26 randomised trials.Lancet. 2010;376:1670–1681.

31. Kostis WJ, Moreyra AE, Cheng JQ, Dobrzynski JM, Kostis JB. Contin-uation of mortality reduction after the end of randomized therapy inclinical trials of lipid-lowering therapy. J Clin Lipidol. 2011;5:97–104.

32. Borenstein M. Introduction to Meta-Analysis. Chichester, UK: JohnWiley & Sons; 2009.

33. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method oftesting and adjusting for publication bias in meta-analysis. Biometrics.2000;56:455–463.

34. Rosenthal R. The ’file drawer problem’ and tolerance for null results.Psychol Bull. 1979;86:638–641.

35. Orwin RG. A fail-safe N for effect size in meta-analysis. J Educ Stat.1983;8:157–159.

36. Fellstrom BC, Jardine AG, Schmieder RE, Holdaas H, Bannister K,Beutler J, et al. Rosuvastatin and cardiovascular events in patientsundergoing hemodialysis. N Engl J Med. 2009;360:1395–1407.

37. Kjekshus J, Apetrei E, Barrios V, Bohm M, Cleland JG, Cornel JH, et al.Rosuvastatin in older patients with systolic heart failure. N Engl J Med.2007;357:2248–2261.

38. Results of the low-dose (20 mg) pravastatin GISSI Prevenzione Trial in4271 patients with recent myocardial infarction: do stopped trials con-tribute to overall knowledge? Gissi Prevenzione Investigators (GruppoItaliano Per Lo Studio Della Sopravvivenza Nell’infarto Miocardico). ItalHeart J. 2000;1:810–820.

39. Knopp RH, d’Emden M, Smilde JG, Pocock SJ. Efficacy and safety ofatorvastatin in the prevention of cardiovascular end points in subjectswith type 2 diabetes: the Atorvastatin Study for Prevention of CoronaryHeart Disease Endpoints in Non-Insulin-Dependent Diabetes Mellitus(ASPEN). Diabetes Care. 2006;29:1478–1485.

40. Tavazzi L, Maggioni AP, Marchioli R, Barlera S, Franzosi MG, Latini R,et al. Effect of rosuvastatin in patients with chronic heart failure (theGISSI-HF Trial): a randomised, double-blind, placebo-controlled trial.Lancet. 2008;372:1231–1239.

41. Wanner C, Krane V, Marz W, Olschewski M, Mann JF, Ruf G, et al.Atorvastatin in patients with type 2 diabetes mellitus undergoing hemo-dialysis. N Engl J Med. 2005;353:238–248.

42. Pedersen TR, Faergeman O, Kastelein JJ, Olsson AG, Tikkanen MJ,Holme I, et al. High-dose atorvastatin vs usual-dose simvastatin forsecondary prevention after myocardial infarction: the IDEAL study: arandomized controlled trial. JAMA. 2005;294:2437–2445.

Page 32: Statin Therapy and the Risk of Intracerebral Hemorrhagestroke.ahajournals.org/content/strokeaha/43/8/2149.full.pdf · Statin Therapy and the Risk of Intracerebral Hemorrhage A Meta-Analysis

McKinney y Kostis Mal Tratamiento con estatinas y riesgo de hemorragia intracerebral 121

43. Furberg CD, Adams HP Jr, Applegate WB, Byington RP, Espeland MA,Hartwell T, et al. Effect of lovastatin on early carotid atherosclerosis andcardiovascular events. Asymptomatic Carotid Artery Progression Study(ACAPS) research group. Circulation. 1994;90:1679–1687.

44. Holdaas H, Fellstrom B, Jardine AG, Holme I, Nyberg G, Fauchald P, etal. Effect of fluvastatin on cardiac outcomes in renal transplant recipients:a multicentre, randomised, placebo-controlled trial. Lancet. 2003;361:2024–2031.

45. Bone HG, Kiel DP, Lindsay RS, Lewiecki EM, Bolognese MA, LearyET, et al. Effects of atorvastatin on bone in postmenopausal women withdyslipidemia: a double-blind, placebo-controlled, dose-ranging trial.J Clin Endocrinol Metab. 2007;92:4671–4677.

46. Kleemann A, Eckert S, von Eckardstein A, Lepper W, Schernikau U,Gleichmann U, et al. Effects of lovastatin on progression of non-dilatedand dilated coronary segments and on restenosis in patients after PTCA.The Cholesterol Lowering Atherosclerosis PTCA Trial (CLAPT). EurHeart J. 1999;20:1393–1406.

47. Baigent C, Landray MJ, Reith C, Emberson J, Wheeler DC, Tomson C,et al. The effects of lowering LDL cholesterol with simvastatin plusezetimibe in patients with chronic kidney disease (study of heart and renalprotection): a randomised placebo-controlled trial. Lancet. 2011;377:2181–2192.

48. Sturgeon JD, Folsom AR, Longstreth WT Jr, Shahar E, Rosamond WD,Cushman M. Risk factors for intracerebral hemorrhage in a pooled pro-spective study. Stroke. 2007;38:2718–2725.

49. Iribarren C, Reed DM, Burchfiel CM, Dwyer JH. Serum total cholesteroland mortality. Confounding factors and risk modification in Japanese-American men. JAMA. 1995;273:1926–1932.

50. Bjorkhem I, Meaney S. Brain cholesterol: long secret life behind a barrier.Arterioscler Thromb Vasc Biol. 2004;24:806–815.

51. Collins R, Armitage J, Parish S, Sleight P, Peto R. Effects of cholester-ol-lowering with simvastatin on stroke and other major vascular events in20 536 people with cerebrovascular disease or other high-risk conditions.Lancet. 2004;363:757–767.

52. Hackam DG, Austin PC, Huang A, Juurlink DN, Mamdani MM, PatersonJM, et al. Statins and intracerebral hemorrhage: a retrospective cohortstudy. Arch Neurol. 2012;69:39–45.

53. Hackam DG, Woodward M, Newby LK, Bhatt DL, Shao M, Smith EE,et al. Statins and intracerebral hemorrhage: collaborative systematicreview and meta-analysis. Circulation. 2011;124:2233–2242.

54. Serebruany VL, Miller M, Pokov AN, Malinin AI, Lowry DR, TanguayJF, et al. Effect of statins on platelet par-1 thrombin receptor in patientswith the metabolic syndrome (from the PAR-1 Inhibition by Statins[PARIS] study). Am J Cardiol. 2006;97:1332–1336.

55. Serebruany VL, Malinin AI, Hennekens CH. Statins increase risk ofhemorrhagic stroke by inhibition of the PAR-1 receptor. CerebrovascDis. 2007;24:477–479.

56. Westover MB, Bianchi MT, Eckman MH, Greenberg SM. Statin usefollowing intracerebral hemorrhage: a decision analysis. Arch Neurol.2010;68:573–579.

43. Furberg CD, Adams HP Jr, Applegate WB, Byington RP, Espeland MA,Hartwell T, et al. Effect of lovastatin on early carotid atherosclerosis andcardiovascular events. Asymptomatic Carotid Artery Progression Study(ACAPS) research group. Circulation. 1994;90:1679–1687.

44. Holdaas H, Fellstrom B, Jardine AG, Holme I, Nyberg G, Fauchald P, etal. Effect of fluvastatin on cardiac outcomes in renal transplant recipients:a multicentre, randomised, placebo-controlled trial. Lancet. 2003;361:2024–2031.

45. Bone HG, Kiel DP, Lindsay RS, Lewiecki EM, Bolognese MA, LearyET, et al. Effects of atorvastatin on bone in postmenopausal women withdyslipidemia: a double-blind, placebo-controlled, dose-ranging trial.J Clin Endocrinol Metab. 2007;92:4671–4677.

46. Kleemann A, Eckert S, von Eckardstein A, Lepper W, Schernikau U,Gleichmann U, et al. Effects of lovastatin on progression of non-dilatedand dilated coronary segments and on restenosis in patients after PTCA.The Cholesterol Lowering Atherosclerosis PTCA Trial (CLAPT). EurHeart J. 1999;20:1393–1406.

47. Baigent C, Landray MJ, Reith C, Emberson J, Wheeler DC, Tomson C,et al. The effects of lowering LDL cholesterol with simvastatin plusezetimibe in patients with chronic kidney disease (study of heart and renalprotection): a randomised placebo-controlled trial. Lancet. 2011;377:2181–2192.

48. Sturgeon JD, Folsom AR, Longstreth WT Jr, Shahar E, Rosamond WD,Cushman M. Risk factors for intracerebral hemorrhage in a pooled pro-spective study. Stroke. 2007;38:2718–2725.

49. Iribarren C, Reed DM, Burchfiel CM, Dwyer JH. Serum total cholesteroland mortality. Confounding factors and risk modification in Japanese-American men. JAMA. 1995;273:1926–1932.

50. Bjorkhem I, Meaney S. Brain cholesterol: long secret life behind a barrier.Arterioscler Thromb Vasc Biol. 2004;24:806–815.

51. Collins R, Armitage J, Parish S, Sleight P, Peto R. Effects of cholester-ol-lowering with simvastatin on stroke and other major vascular events in20 536 people with cerebrovascular disease or other high-risk conditions.Lancet. 2004;363:757–767.

52. Hackam DG, Austin PC, Huang A, Juurlink DN, Mamdani MM, PatersonJM, et al. Statins and intracerebral hemorrhage: a retrospective cohortstudy. Arch Neurol. 2012;69:39–45.

53. Hackam DG, Woodward M, Newby LK, Bhatt DL, Shao M, Smith EE,et al. Statins and intracerebral hemorrhage: collaborative systematicreview and meta-analysis. Circulation. 2011;124:2233–2242.

54. Serebruany VL, Miller M, Pokov AN, Malinin AI, Lowry DR, TanguayJF, et al. Effect of statins on platelet par-1 thrombin receptor in patientswith the metabolic syndrome (from the PAR-1 Inhibition by Statins[PARIS] study). Am J Cardiol. 2006;97:1332–1336.

55. Serebruany VL, Malinin AI, Hennekens CH. Statins increase risk ofhemorrhagic stroke by inhibition of the PAR-1 receptor. CerebrovascDis. 2007;24:477–479.

56. Westover MB, Bianchi MT, Eckman MH, Greenberg SM. Statin usefollowing intracerebral hemorrhage: a decision analysis. Arch Neurol.2010;68:573–579.