Statin Therapy and the Risk of Intracerebral Hemorrhage
Transcripción
Statin Therapy and the Risk of Intracerebral Hemorrhage
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 Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 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; P⫽0.47). 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; 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 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 䡲 intracerebral hemorrhage 䡲 meta-analysis 䡲 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 effects.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 Cholesterol Levels (SPARCL) study of high-dose atorvastatin in secondary stroke prevention demonstrated a significant overall 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 hemorrhagic 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 metaanalysis 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; P⫽0.2).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 2150 Stroke August 2012 Figure 1. Literature search profile. Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 all-cause mortality, and determine whether the occurrence of hemorrhagic stroke was related to the low-density lipoprotein (LDL) change achieved by therapy. We performed a metaanalysis including all available randomized controlled trials of statin therapy that reported hemorrhagic stroke rate. Methods The methods used in this meta-analysis are similar to those previously reported.31 We performed a literature search through January 25, 2012, using Medline, Web of Science, and the Cochrane Library databases identifying randomized controlled trials (RCTs) of statin therapy and ICH or hemorrhagic stroke published in the English language. Search criteria are presented in online-only Data Supplement Figure I. We identified additional RCTs by reviewing reference lists of identified studies and previously published meta-analyses of statin therapy. We included all trials with the following criteria: subjects aged ⱖ18 years, randomized controlled design, blinded outcomes assessment, and recorded data on hemorrhagic stroke or ICH. Both primary and secondary prevention trials were included as were trials comparing active therapy with control therapy of “usual care,” placebo, or lower-dose statins. Studies were included if hemorrhagic stroke rates were reported in the first published report, subsequent publications, or meta-analyses. All studies were adjudicated by authors, and disagreements were resolved through discussion and consensus. Based on the search criteria, 31 RCTs of statin therapy that reported ICH or hemorrhagic stroke as an outcome were included in the analysis (Figure 1). Statistical Methods Study Selection Data Extraction and Quality Assessment Data 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 concealment, comparison of baseline characteristics, defined eligibility criteria, type of control, blinding (patients, investigators, assessment of vital status), percent lost to follow-up, and use of intention-to-treat analysis were assessed. Data Synthesis and Analysis ICH rate, total stroke rate, and all-cause mortality in the active and control groups were calculated. Statistical analyses were performed using Comprehensive Meta-Analysis 2.232 and JMP 9.0 (SAS Institute, Cary, NC). ORs and 95% CIs were calculated for the 3 rates described using the intention-to-treat approach. Weighted pooled treatment effects were calculated for each of the 3 event rates using random-effects models. Heterogeneity of the effects was evaluated by the use of the Q statistic.33 Publication bias was examined by constructing funnel plots,32 by Duval and Tweedie’s Trim and Fill,33 and the fail-safe N models of Rosenthal34 and Orwin.35 Sensitivity Analyses Several additional sensitivity analyses were performed. First, 1 study (Collaborative Atorvastatin Diabetes Study [CARDS]) reported 0 hemorrhagic strokes in both the atorvastatin and placebo groups.19,25 Therefore, to avoid inclusion of an undefined OR, CARDS was not considered in our primary analysis, which included only the 30 other studies. To examine the effects of excluding this study from our primary analysis, a sensitivity analysis was performed by imputing 1 ICH (instead of 0) in each of the 2 groups (active and control) in CARDS and including this with the other 30 studies. A second sensitivity analysis was performed by removing 1 study at a time (of the 30) in an iterative fashion. This was performed to evaluate whether individual trials may have overly influenced the findings of this study. Third, a cumulative meta-analysis was performed to investigate how well the estimated OR of hemorrhagic stroke converged with the iterative addition of progressively larger studies. Results Thirty-one trials of statin therapy that randomized a total of 182 803 subjects were identified and included in the McKinney and Kostis Table. Statin Therapy and Risk of ICH 2151 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 Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 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 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) All studies total (%) ICH indicates intracerebral hemorrhage. analysis.6–24,36–47 Of the 31 RCTs, 6 studies11,17,18,20,24,42 compared high-dose with low-dose statin therapy, and the remainder6 –10,12–15,19,21–23,36 – 41,43– 47 compared statin therapy with placebo or “usual care.” A total of 91 588 subjects were randomized to active therapy and 91 215 subjects to control therapy. The median length of follow-up was 46.8 months. The patient and LDL characteristics for RCTs included in this analysis are presented in online-only Data Supplement 2152 Stroke August 2012 ICH - Statin Meta-Analysis: ICH Study name Subgroup within study Statistics for each study Odds ratio Stroke / Total p-Value Lower limit Upper limit 4S PLACEBO 0.20 0.2989 0.01 4.17 0 / 2221 2 / 2223 AF-TEXCAPS PLACEBO 3.00 0.5014 0.12 73.62 1 / 3304 0 / 3301 ALLHAT-LLT PLACEBO 3.42 0.0158 1.26 9.27 17 / 5170 5 / 5185 ASCOT-LLA PLACEBO 0.55 0.1071 0.26 1.14 11 / 5168 20 / 5137 Active Control ATOZ LOW DOSE 12.84 0.0820 0.72 228.14 6 / 2265 0 / 2232 AURORA PLACEBO 1.19 0.5608 0.66 2.14 25 / 1389 21 / 1384 6 / 2078 CARE PLACEBO 0.33 0.1774 0.07 1.65 2 / 2081 CORONA PLACEBO 1.66 0.2308 0.72 3.80 15 / 2514 9 / 2497 GISSI-P PLACEBO 2.99 0.5019 0.12 73.55 1 / 2138 0 / 2133 GREACE PLACEBO 1.00 1.0000 0.06 16.02 1 / 800 1 / 800 HPS PLACEBO 0.96 0.8434 0.65 1.41 51 / 10269 53 / 10267 9 / 8901 JUPITER PLACEBO 0.67 0.4415 0.24 1.87 6 / 8901 LIPID PLACEBO 1.89 0.1236 0.84 4.24 17 / 4512 9 / 4502 MEGA PLACEBO 1.17 0.6632 0.57 2.41 16 / 3866 14 / 3966 10 / 2913 Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 PROSPER PLACEBO 0.81 0.6490 0.32 2.04 8 / 2891 PROVE-IT LOW DOSE 3.94 0.2205 0.44 35.25 4 / 2099 1 / 2063 TNT LOW DOSE 0.94 0.8666 0.48 1.87 16 / 4995 17 / 5006 SEARCH LOW DOSE 0.96 0.8871 0.55 1.68 24 / 6031 25 / 6033 SPARCL PLACEBO 1.68 0.0191 1.09 2.60 55 / 2365 33 / 2366 ASPEN PLACEBO 1.98 0.4296 0.36 10.85 4 / 1211 2 / 1199 GISSI-HF PLACEBO 3.69 0.0454 1.03 13.23 11 / 2285 3 / 2289 4D PLACEBO 0.64 0.4299 0.21 1.96 5 / 619 8 / 633 IDEAL LOW DOSE 1.00 0.9969 0.32 3.11 6 / 4439 6 / 4449 MIRACL PLACEBO 0.14 0.1992 0.01 2.78 0 / 1538 3 / 1548 PATE LOW DOSE 0.60 0.1963 0.28 1.30 11 / 331 18 / 334 ACAPS PLACEBO 0.14 0.1965 0.01 2.75 0 / 460 3 / 459 ALERT PLACEBO 0.59 0.1817 0.27 1.28 10 / 1050 17 / 1052 BONE PLACEBO 0.74 0.8540 0.03 18.28 1 / 485 0 / 119 CLAPT PLACEBO 0.34 0.5060 0.01 8.34 0 / 112 1 / 114 SHARP PLACEBO 45 / 4650 37 / 4620 1.21 0.3916 0.78 1.87 1.08 0.4687 0.88 1.32 Odds ratio and 95% CI 0.01 0.1 Favors Active 1 10 100 Favors Control All studies Figure 2. Forrest plot of random-effects meta-analysis of randomized trials of statins and intracerebral hemorrhage. Tables I and II. The mean age of was 62.6⫾5.2 years. A total of 67.0% of patients were male and 78.1% were white. Only 11.0% 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% actively smoked cigarettes at study enrollment. A total of 61.0% of subjects were treated with aspirin or oral anticoagulants. Stroke events are presented in the Table. Intracerebral Hemorrhage ICH occurred in 358 subjects (0.39%) in the active treatment group versus 318 (0.35%) in the control group. In the primary analysis 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 for heterogeneity (interaction) was 0.38, Q⫽30.705, and degrees of freedom⫽29. No statistically significant difference in the risk of ICH was observed when the meta-analyses were stratified by control group or primary versus secondary prevention studies (online-only Data Supplement Figure II). The sensitivity analysis that included all 31 studies (including CARDS) produced nearly identical results as the primary analysis (OR, 1.08; 95% CI, 0.88 –1.31; P⫽0.46). In all 30 iterations of the sensitivity analysis performed by removing 1 study at a time, there was no association between ICH and active treatment. The point estimate of the OR varied from 1.04 to 1.11 with a lower limit of 0.84 to 0.91 and P of 0.27 to 0.77. During the cumulative meta-analysis, there was no association between ICH and active treatment in 28 of 30 iterations. 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 as demonstrated by a symmetrical funnel plot. The OR and CI of the overall effect (OR, 1.08; 95% CI, 0.88 –1.31) remained unchanged using Duval and Tweedie’s Trim and Fill method.33 Orwin’s fail-safe N revealed that 25 additional studies 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 relationship between the LDL effect of statin therapy and ICH risk. There was no relationship between the effects of active versus control therapy with measures of LDL cholesterol. Specifically, there were no significant relationships of the log OR (active/control) of each study versus (1) the difference (active minus control) of the LDL cholesterol drop (difference McKinney and Kostis Statin Therapy and Risk of ICH 2153 ICH - Statin Meta-Analysis: All Stroke Study name 4S Subgroup within study PLACEBO Statistics for each study Odds ratio p-Value Lower limit 0.68 0.0533 0.46 Stroke / Total Upper limit 1.01 Active Control 44 / 2221 64 / 2223 AF-TEXCAPS PLACEBO 0.82 0.5880 0.40 1.67 14 / 3304 17 / 3301 ALLHAT-LLT PLACEBO 0.90 0.2982 0.75 1.09 209 / 5170 231 / 5185 ASCOT-LLA PLACEBO 0.73 0.0234 0.55 0.96 89 / 5168 121 / 5137 ATOZ LOW DOSE 0.79 0.3448 0.48 1.30 28 / 2265 35 / 2232 AURORA PLACEBO 1.17 0.3223 0.86 1.59 94 / 1389 81 / 1384 CARE PLACEBO 0.68 0.0340 0.48 0.97 54 / 2081 78 / 2078 CORONA PLACEBO 0.86 0.2138 0.67 1.09 126 / 2514 145 / 2497 GISSI-P PLACEBO 1.05 0.8780 0.56 1.97 20 / 2138 19 / 2133 GREACE PLACEBO 0.52 0.1197 0.23 1.18 9 / 800 17 / 800 HPS PLACEBO 0.75 0.0000 0.66 0.85 444 / 10269 585 / 10267 JUPITER PLACEBO 0.51 0.0019 0.34 0.78 33 / 8901 64 / 8901 LIPID PLACEBO 0.81 0.0251 0.68 0.97 224 / 4512 272 / 4502 MEGA PLACEBO 0.83 0.3151 0.57 1.20 50 / 3866 62 / 3966 PROSPER PLACEBO 1.04 0.7533 0.81 1.33 135 / 2891 131 / 2913 PROVE-IT LOW DOSE 1.09 0.7928 0.58 2.03 21 / 2099 19 / 2063 TNT LOW DOSE 0.75 0.0209 0.59 0.96 117 / 4995 155 / 5006 Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 SEARCH LOW DOSE 0.91 0.2900 0.77 1.08 255 / 6031 279 / 6033 SPARCL PLACEBO 0.83 0.0416 0.70 0.99 265 / 2365 311 / 2366 CARDS PLACEBO 0.53 0.0184 0.31 0.90 21 / 1429 39 / 1412 ASPEN PLACEBO 0.88 0.6022 0.55 1.41 34 / 1211 38 / 1199 GISSI-HF PLACEBO 1.25 0.1786 0.90 1.74 82 / 2285 66 / 2289 4D PLACEBO 1.41 0.0979 0.94 2.12 59 / 619 44 / 633 IDEAL LOW DOSE 0.87 0.2012 0.69 1.08 151 / 4439 174 / 4449 MIRACL PLACEBO 0.50 0.0507 0.25 1.00 12 / 1538 24 / 1548 PATE LOW DOSE 0.60 0.1963 0.28 1.30 11 / 331 18 / 334 ACAPS PLACEBO 0.09 0.1030 0.00 1.63 0 / 460 5 / 459 ALERT PLACEBO 1.03 0.8667 0.76 1.39 93 / 1050 91 / 1052 BONE PLACEBO 0.74 0.8540 0.03 18.28 1 / 485 0 / 119 CLAPT PLACEBO 0.34 0.5060 0.01 8.34 0 / 112 1 / 114 SHARP PLACEBO 0.80 0.0356 0.65 0.99 171 / 4650 210 / 4620 0.84 0.0000 0.78 0.91 Odds ratio and 95% CI 0.5 1 Favors Active 2 Favors Control All studies Figure 3. Forrest plot of random-effects meta-analysis of randomized trials of statins and all stroke. 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) this difference 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 group during follow-up (slope, ⫺0.0049; SE, 0.0043; 95% CI, ⫺0.0133 to 0.0035; P⫽0.26). All Stroke There was a total of 6262 strokes in this meta-analysis. The overall stroke rate was 3.13% in the active group versus 3.72% in the control group. Active therapy resulted in a significant 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 freedom⫽30. The number needed to treat with active statin therapy to prevent any stroke during study follow-up was 200 (absolute risk reduction⫽0.5%, P⬍0.0001). All-Cause Mortality There were 20 195 deaths recorded in the 31 trials included in this analysis. There was a significantly lower rate of all-cause mortality 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 needed to treat with active statin therapy to prevent 1 death was 167 (absolute risk reduction⫽0.6%, P⬍0.0001). Discussion Epidemiological studies have reported increased rates of hemorrhagic stroke and ICH-related mortality in populations with low cholesterol levels.1,4,48,49 It has been hypothesized that cholesterol may be important for cerebrovascular wall integrity and that low levels may increase the risk of vessel rupture and ICH.50 Furthermore, several studies have reported an association of hemorrhagic stroke with statin use. The SPARCL trial of high-dose atorvastatin in secondary stroke prevention reported an excess of ICH with active treatment compared with placebo (55 versus 33; P⫽0.02).22 Similarly, there was a 2-fold increase in hemorrhagic stroke in the HPS among patients with prior stroke treated with simvastatin.51 Three previous meta-analyses of statin therapy have found no increased risk of hemorrhagic stroke.25,30,52 In a meta-analysis of 11 trials, Amarenco and Labreuche25 reported a negligible risk of hemorrhagic stroke with statin 2154 Stroke August 2012 ICH - Statin Meta-Analysis: Death Study name Subgroup within study Statistics for each study Dead / Total Odds ratio p-Value Lower limit PLACEBO 0.69 0.0002 0.56 0.84 AF-TEXCAPS PLACEBO 1.04 0.8130 0.76 1.43 80 / 3304 77 / 3301 ALLHAT-LLT PLACEBO 0.99 0.8071 0.88 1.11 631 / 5170 641 / 5185 ASCOT-LLA PLACEBO 0.86 0.1493 0.71 1.05 185 / 5168 212 / 5137 ATOZ LOW DOSE 1.25 0.1036 0.96 1.62 130 / 2265 104 / 2232 AURORA PLACEBO 0.93 0.3162 0.80 1.08 636 / 1389 660 / 1384 CARE PLACEBO 0.91 0.3791 0.74 1.12 180 / 2081 196 / 2078 4S Upper limit Active Control 182 / 2221 256 / 2223 CORONA PLACEBO 0.93 0.2650 0.83 1.05 728 / 2514 759 / 2497 GISSI-P PLACEBO 0.81 0.1928 0.59 1.11 72 / 2138 88 / 2133 GREACE PLACEBO 0.56 0.0309 0.33 0.95 23 / 800 40 / 800 HPS PLACEBO 0.86 0.0003 0.80 0.93 1328 / 10269 1507 / 10267 Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 JUPITER PLACEBO 0.80 0.0189 0.66 0.96 198 / 8901 247 / 8901 LIPID PLACEBO 0.77 0.0000 0.69 0.86 717 / 4512 888 / 4502 MEGA PLACEBO 0.71 0.0532 0.50 1.00 55 / 3866 79 / 3966 PROSPER PLACEBO 0.98 0.8393 0.83 1.16 298 / 2891 305 / 2913 PROVE-IT LOW DOSE 0.68 0.0458 0.46 0.99 46 / 2099 66 / 2063 TNT LOW DOSE 1.01 0.9096 0.85 1.20 284 / 4995 282 / 5006 SEARCH LOW DOSE 0.99 0.8879 0.90 1.09 964 / 6031 970 / 6033 SPARCL PLACEBO 1.03 0.7962 0.84 1.25 216 / 2365 211 / 2366 CARDS PLACEBO 0.72 0.0617 0.51 1.02 61 / 1429 82 / 1412 ASPEN PLACEBO 1.02 0.9084 0.72 1.44 70 / 1211 68 / 1199 GISSI-HF PLACEBO 1.03 0.6431 0.91 1.17 657 / 2285 644 / 2289 4D PLACEBO 0.90 0.3628 0.72 1.13 297 / 619 320 / 633 IDEAL LOW DOSE 0.98 0.7832 0.84 1.14 366 / 4439 374 / 4449 MIRACL PLACEBO 0.95 0.7507 0.67 1.34 64 / 1538 68 / 1548 PATE LOW DOSE 0.69 0.3056 0.34 1.40 14 / 331 20 / 334 ACAPS PLACEBO 0.12 0.0485 0.02 0.99 1 / 460 8 / 459 ALERT PLACEBO 1.04 0.7357 0.81 1.34 143 / 1050 138 / 1052 CLAPT PLACEBO 0.20 0.3006 0.01 4.21 0 / 112 2 / 114 SHARP PLACEBO 1.02 0.6336 0.93 1.13 1142 / 4650 1115 / 4620 0.92 0.0007 0.87 0.96 Odds ratio and 95% CI 0.5 1 Favors Active 2 Favors Control All studies Figure 4. Forrest plot of random-effects meta-analysis of randomized trials of statins and all-cause mortality. therapy (relative risk, 1.03; 95% CI, 0.75–1.41). Studying intensive LDL reduction with statins, the CTT collaboration performed a meta-analysis of 26 RCTs and reported a nonsignificant trend toward an increased risk of ICH (relative risk, 1.15; 95% CI, 0.93–1.41).30 A more recent meta-analysis including 23 RCTs found no evidence of increased risk of ICH.53 We performed a comprehensive meta-analysis of previously conducted randomized clinical trials of statin therapy for which data on ICH or hemorrhagic stroke were available. We included 31 RCTs with a total of 182 803 subjects. In our analysis, active therapy was associated with a nonsignificant increase in the risk of ICH (OR, 1.08; 95% CI, 0.88 –1.32). The small nonsignificant observed excess of ICH was not related to the LDL effects of statin therapy. We found no relationship between the achieved LDL level or the degree of LDL reduction and the risk of hemorrhagic stroke in this analysis. This suggests that any potential increase in ICH risk could be attributable to other non-LDL effects of statin therapy. This is in agreement with a large meta-analysis of observational cohort studies that found no correlation with fatal stroke and baseline total cholesterol in almost 900 000 subjects.27 In addition to their lipid-lowering properties, statins may have antithrombotic properties by inhibiting platelet aggregation and enhancing fibrinolysis.54,55 The antithrombotic affects of statins could account for a theoretically increased risk of bleeding complications. When stratifying our analysis by type of prevention, there was a nonsignificant trend toward increased odds of ICH in secondary 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 of statin therapy for secondary stroke prevention. Post hoc analyses of HPS and SPARCL suggest that patients with prior stroke may be at particularly increased risk of hemorrhagic stroke.28,51 However, a recent large retrospective cohort study found no evidence of increased risk of ICH in patients treated with statins after an ischemic stroke.52 Westover et al,56 using a 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), was favored over a wide range of clinical parameters. Our analysis did not evaluate the risk of statin therapy in this subgroup of patients with prior ICH. Based on current data, caution should be used in treating these patients with statins, and further research is warranted. McKinney and Kostis Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 There are several limitations to this analysis. First, we did not have access to patient-level data and there may be unaccounted patient variables that influenced the rates of hemorrhagic stroke. Second, hemorrhagic stroke is a nebulous term that may include ICH, subarachnoid hemorrhage, subdural or epidural hemorrhages, or hemorrhagic transformation of an ischemic stroke. Although ICH is the condition of interest, it is possible that prior RCTs that reported hemorrhagic stroke rates included some of these other causes of intracranial bleeding. Lastly, although we performed the most comprehensive analysis yet published of statin therapy and ICH risk in randomized trials, it is possible that there are other trials, particularly those published in languages other than English, that were excluded. Although we included all pertinent trials, regardless of size, no publication bias or heterogeneity was found in our analysis. Statin therapy was not associated with a significant increased risk of ICH. There was no effect on ICH risk related to the degree of decline in LDL or to the achieved level. The significant reduction in total stroke and all-cause mortality more than offset any slight increase in ICH risk. These findings support the current recommendations to prescribe statins in otherwise appropriate patients. 9. 10. 11. 12. 13. 14. 15. Conclusions 16. In this meta-analysis of 31 RCTs, statin therapy was not associated with a significant increase in odds of ICH. However, statin therapy is associated with significant reductions in total stroke and death, thereby negating any potential hemorrhage risk in an unselected patient population. 17. Sources of Funding This study was funded in part by the Robert Wood Johnson Foundation and the Schering-Plough Foundation. 18. Disclosures 19. None. References 1. Iso H, Jacobs DR Jr, Wentworth D, Neaton JD, Cohen JD. Serum cholesterol levels and six-year mortality from stroke in 350 977 men screened 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. Different risk factors for different stroke subtypes: association of blood pressure, cholesterol, and antioxidants. Stroke. 1999;30:2535–2540. 3. Horenstein RB, Smith DE, Mosca L. Cholesterol predicts stroke mortality in 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. Cholesterol, coronary heart disease, and stroke in the Asia Pacific region. Int J Epidemiol. 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 coronary heart 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 myocardial infarction in patients with average cholesterol levels. Cholesterol and recurrent events trial investigators. N Engl J Med. 1996;335:1001–1009. 8. 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/ 20. 21. 22. 23. 24. 25. 26. 27. Statin Therapy and Risk of ICH 2155 TEXCAPS. Air Force/Texas Coronary Atherosclerosis Prevention Study. JAMA. 1998;279:1615–1622. 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. 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. 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 AntiAtherosclerosis Trial in the Elderly (PATE). J Atheroscler Thromb. 2001;8:33– 44. Major outcomes in moderately hypercholesterolemic, hypertensive patients randomized to pravastatin vs usual care: the Antihypertensive and LipidLowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT). JAMA. 2002;288:2998–3007. 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. 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 CoronaryHeart-Disease Evaluation (GREACE) study. Curr Med Res Opin. 2002; 18:220 –228. 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. 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. 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. 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. 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. 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. 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. Amarenco P, Bogousslavsky J, Callahan A III, 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. 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. 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. Amarenco P, Labreuche J. Lipid management in the prevention of stroke: review and updated meta-analysis of statins for stroke prevention. Lancet Neurol. 2009;8:453– 463. Cholesterol, diastolic blood pressure, and stroke: 13 000 strokes in 450 000 people in 45 prospective cohorts. Prospective studies collaboration. Lancet. 1995;346:1647–1653. Lewington S, Whitlock G, Clarke R, Sherliker P, Emberson J, Halsey J, et al. Blood cholesterol and vascular mortality by age, sex, and blood 2156 28. 29. 30. 31. 32. 33. 34. Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 35. 36. 37. 38. 39. 40. 41. 42. Stroke August 2012 pressure: a meta-analysis of individual data from 61 prospective studies with 55 000 vascular deaths. Lancet. 2007;370:1829 –1839. Goldstein MR, Mascitelli L, Pezzetta F. Hemorrhagic stroke in the stroke prevention by Aggressive Reduction in Cholesterol Levels Study. Neurology. 2009;72:1448; author reply 1448 –1449. Manktelow BN, Potter JF. Interventions in the management of serum lipids for preventing stroke recurrence. Cochrane Database Syst Rev. 2009;3:CD002091. Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170 000 participants in 26 randomised trials. Lancet. 2010;376:1670 –1681. Kostis WJ, Moreyra AE, Cheng JQ, Dobrzynski JM, Kostis JB. Continuation of mortality reduction after the end of randomized therapy in clinical trials of lipid-lowering therapy. J Clin Lipidol. 2011;5:97–104. Borenstein M. Introduction to Meta-Analysis. Chichester, UK: John Wiley & Sons; 2009. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56:455– 463. Rosenthal R. The ’file drawer problem’ and tolerance for null results. Psychol Bull. 1979;86:638 – 641. Orwin RG. A fail-safe N for effect size in meta-analysis. J Educ Stat. 1983;8:157–159. 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. 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. 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. 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. 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. 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. 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. 43. 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. 44. 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. 45. 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. 46. 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. 47. 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. 48. Sturgeon JD, Folsom AR, Longstreth WT Jr, Shahar E, Rosamond WD, Cushman M. Risk factors for intracerebral hemorrhage in a pooled prospective study. Stroke. 2007;38:2718 –2725. 49. Iribarren C, Reed DM, Burchfiel CM, Dwyer JH. Serum total cholesterol and mortality. Confounding factors and risk modification in JapaneseAmerican 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 cholesterol-lowering with simvastatin on stroke and other major vascular events in 20 536 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, Paterson JM, et al. Statins and intracerebral hemorrhage: a retrospective cohort study. 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 systematic review and meta-analysis. Circulation. 2011;124:2233–2242. 54. Serebruany VL, Miller M, Pokov AN, Malinin AI, Lowry DR, Tanguay JF, et al. Effect of statins on platelet par-1 thrombin receptor in patients with 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 of hemorrhagic stroke by inhibition of the PAR-1 receptor. Cerebrovasc Dis. 2007;24:477– 479. 56. Westover MB, Bianchi MT, Eckman MH, Greenberg SM. Statin use following intracerebral hemorrhage: a decision analysis. Arch Neurol. 2010;68:573–579. Statin Therapy and the Risk of Intracerebral Hemorrhage: A Meta-Analysis of 31 Randomized Controlled Trials James S. McKinney and William J. Kostis Downloaded from http://stroke.ahajournals.org/ by guest on October 1, 2016 Stroke. 2012;43:2149-2156; originally published online May 15, 2012; doi: 10.1161/STROKEAHA.112.655894 Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2012 American Heart Association, Inc. All rights reserved. Print ISSN: 0039-2499. Online ISSN: 1524-4628 The online version of this article, along with updated information and services, is located on the World Wide Web at: http://stroke.ahajournals.org/content/43/8/2149 Data Supplement (unedited) at: http://stroke.ahajournals.org/content/suppl/2012/05/15/STROKEAHA.112.655894.DC1.html http://stroke.ahajournals.org/content/suppl/2013/10/02/STROKEAHA.112.655894.DC2.html http://stroke.ahajournals.org/content/suppl/2016/04/10/STROKEAHA.112.655894.DC3.html Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Stroke is online at: http://stroke.ahajournals.org//subscriptions/ SUPPLEMENTAL MATERIAL Supplemental Figure 1. Literature search criteria for Medline (A), Web of Science (B), and the Cochrane Library (C), January 25, 2012. A. B. C. 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). A. B. ICH Meta-Analysis: ICH OR by Type of Prevention Group by Subgroup within study PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY PRIMARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY Overall Subgroup within study Statistics for each study Odds Lower ratio p-Value limit AF-TEXCAPS PRIMARY ALLHAT-LLTPRIMARY ASCOT-LLAPRIMARY AURORA PRIMARY HPS PRIMARY JUPITER PRIMARY MEGA PRIMARY PROSPER PRIMARY ASPEN PRIMARY PATE PRIMARY ACAPS PRIMARY ALERT PRIMARY BONE PRIMARY SHARP PRIMARY 4S ATOZ CARE CORONA GISSI-P GREACE LIPID PROVE-IT TNT SEARCH SPARCL GISSI-HF 4D IDEAL MIRACL CLAPT SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY SECONDARY 3.00 3.42 0.55 1.19 0.96 0.67 1.17 0.81 1.98 0.60 0.14 0.59 0.74 1.21 0.96 0.20 12.84 0.33 1.66 2.99 1.00 1.89 3.94 0.94 0.96 1.68 3.69 0.64 1.00 0.14 0.34 1.26 1.06 0.5014 0.0158 0.1071 0.5608 0.8434 0.4415 0.6632 0.6490 0.4296 0.1963 0.1965 0.1817 0.8540 0.3916 0.7663 0.2989 0.0820 0.1774 0.2308 0.5019 1.0000 0.1236 0.2205 0.8666 0.8871 0.0191 0.0454 0.4299 0.9969 0.1992 0.5060 0.1638 0.5409 Odds ratio and 95% CI Upper limit 0.12 73.62 1.26 9.27 0.26 1.14 0.66 2.14 0.65 1.41 0.24 1.87 0.57 2.41 0.32 2.04 0.36 10.85 0.28 1.30 0.01 2.75 0.27 1.28 0.03 18.28 0.78 1.87 0.75 1.23 0.01 4.17 0.72 228.14 0.07 1.65 0.72 3.80 0.12 73.55 0.06 16.02 0.84 4.24 0.44 35.25 0.48 1.87 0.55 1.68 1.09 2.60 1.03 13.23 0.21 1.96 0.32 3.11 0.01 2.78 0.01 8.34 0.91 1.73 0.87 1.29 0.01 0.1 Favors Active 1 10 Favors Control 100 Supplemental Table 1. Description of Trials, Patient Demographics Trial, Year Active Drug Control Total Subjects 4444 Age (y ears) 63 Gender (% male) 81 Race (% white) --- Follow-up (months) 64.8 DM (%) 4S, 19941 Simvastatin Placebo 4.5 AF-TEXCAPS, Lovastatin Placebo 6605 58 85 89 62.4 12.2 2 1998 ALLHAT-LLT, Pravastatin Usual care 10355 66 51 41 57.6 35.2 20023 ASCOT-LLA, Atorvastatin Placebo 10305 63 81 95 39.6 24.6 20034 5 A-to-Z, 2004 Simvastatin Low-dose 4497 61 76 --24 23.5 AURORA, Rosuvastatin Placebo 2773 64 62 85 45.6 26.0 6 2009 7 CARE, 1996 Pravastatin Placebo 4159 59 86 93 60 15.5 CORONA, Rosuvastatin Placebo 5011 73 76 --32.8 29.5 8 2007 GISSI-P, 20009 Pravastatin Usual care 4271 60 86 --23 13.7 GREACE, Atorvastatin Usual care 1600 59 79 --36 19.5 200210 11 HPS, 2002 Simvastatin Placebo 20536 65 75 --60 19.0 JUPITER, Rosuvastatin Placebo 17802 66 62 71 22.8 0.0 200812 LIPID, 199813 Pravastatin Placebo 9014 62 83 --72 9.0 14 MEGA, 2006 Pravastatin Placebo 7832 58 32 0 63.6 21.0 PROSPER, Pravastatin Placebo 5804 75 48 --38.4 11.8 200215 PROVE-IT, Atorvastatin Low dose 4162 58 78 91 24 17.6 200416 17 TNT, 2005 Atorvastatin Low dose 10001 61 81 94 58.8 15.0 SEARCH, Simvastatin Low dose 12064 64 83 --80.4 11.0 201018 SPARCL, 19 Atorvastatin Placebo 4731 63 60 --58.8 17.0 2006 20 CARDS, 2004 Atorvastatin Placebo 2841 62 68 95 46.8 100.0 21 ASPEN, 2006 Atorvastatin Placebo 2410 61 66 84 48 100.0 GISSI-HF, Rosuvastatin Placebo 4574 68 77 --46.8 26 22 2008 23 4D, 2005 Atorvastatin Placebo 1252 66 54 --46.8 100.0 IDEAL, 200524 Atorvastatin Low dose 8888 62 81 --57.6 12 MIRACL, Atorvastatin Placebo 3086 65 65 85 4 23 200125 PATE, 200126 Pravastatin Low dose 665 73 21 --46.8 30 ACAPS, 199427 Lovastatin Placebo 919 62 52 92 34.1 2.3 28 ALERT, 2003 Fluvastatin Placebo 2102 50 66 --61.2 19 BONE, 200729 Atorvastatin Placebo 604 59 0 84 13 0 30 CLAPT, 1999 Lovastatin Usual care 226 54 100 --24 5.7 31 SHARP, 2011 Simvastatin Placebo 9720 62 63 72 58.8 23 All Studies 63±5 67±21 78±26 46±18 25±27 (mean±SD) DM indicated diabetes mellitus; HTN, hy pertension; CVD, cardiovascular disease; ASA, aspirin; OA, oral anticoagulant. 26.0 Stroke (%) --- 100 Smoking (%) 26 22.0 --- 0 13 17 100.0 --- 14 23 31 100.0 9.7 19 33 17 HTN (%) CVD (%) ASA/OA (%) 37 50.0 --- 100 41 98 39.6 --- 40 15 42 42.5 --- 100 21 83 63.0 12.5 100 9 60 36.5 --- 100 14 50 43.0 --- 100 5 88 41.0 --- 87 14 63 56.7 N/A 0 16 17 41.5 42.0 3.6 0 100 0 63 21 83 39 61.9 11 44 27 36 50.0 --- 100 37 100 54.3 5.2 100 13 88 42.0 7 100 30 98 62.0 69 100 19 87 84.0 55.0 0 --- 0 21 23 12 15 --- 54.0 4.5 100 14 75 88.0 33 18 8.2 84 39 9 29 52 79 55 8.6 100 28 91 50 28.8 75 --41.2 --- 12.9 0 5.8 0 ----- 72 0 3.1 --50.9 --- 29 11.9 19 --10.2 13 --100 43 ------- 53±20 11±16 59±43 21±12 61±30 Supplemental Table 2. Description of Trials, LDL Characteristics Trial, Year 1 Baseline LDLActive 188 4S, 1994 AF-TEXCAPS, 2 150 1998 ALLHAT-LLT, 146 20023 ASCOT-LLA, 133 20034 5 A-to-Z, 2004 112 AURORA, 20096 100 7 CARE, 1996 139 CORONA, 20078 137 9 GISSI-P, 2000 152 10 GREACE, 2002 180 11 HPS, 2002 131 JUPITER, 200812 108 13 LIPID, 1998 150 MEGA, 200614 157 PROSPER, 20021 5 147 16 PROVE-IT, 2004 106 17 TNT, 2005 97 18 SEARCH, 2010 97 SPARCL, 200619 133 20 CARDS, 2004 118 ASPEN, 200621 113 22 GISSI-HF, 2008 122 23 4D, 2005 121 IDEAL, 200524 122 25 MIRACL, 2001 124 26 PATE, 2001 163 27 ACAPS, 1994 157 28 ALERT, 2003 159 29 BONE, 2007 156 30 CLAPT, 1999 181 31 SHARP, 2011 108 All Studies 137±25 (mean±SD) LDL indicated low density lipoprotein. Mean Baseline LDL -2 ΔLDL-Active – ΔLDLControl 68 188 ΔLDL-Acitve ΔLDL-Control (%) 36 -8 43 150 29 134 12 23 146 16 133 130 3 43 133 32 49 42 41 61 22 83 42 54 45 30 50 44 20 16 61 46 20 39 49 40 52 40 44 51 67 55 33 111 99 139 136 152 179 131 108 150 157 147 106 98 97 134 117 114 121 125 121 124 163 155 159 159 183 108 77 97 136 138 147 169 128 108 148 148 147 95 101 93 128 120 112 130 120 104 135 133 155 146 159 162 105 34 2 3 -2 5 10 3 0 2 9 0 11 -3 4 6 -3 2 -9 5 17 -11 30 0 13 0 21 3 15 40 38 63 17 73 39 54 43 21 50 33 23 12 55 49 18 48 44 23 63 10 44 38 67 34 30 112 100 139 137 152 180 131 108 150 157 147 106 98 97 134 118 114 122 123 122 124 163 156 159 157 182 108 13 40 27 46 11 41 30 50 29 13 34 31 24 12 41 42 16 40 36 19 51 6 28 24 43 19 28 45±14 137±25 132±25 5±10 40±17 137±25 30±12 Follow-up LDL-Acive ΔLDL-Active Baseline LDLControl Follow-up LDL-Control ΔLDL-Control 122 66 188 190 115 35 150 158 111 35 146 87 46 63 58 98 76 130 97 89 54 105 127 97 62 77 81 72 72 93 83 72 82 72 123 113 108 89 126 75 91±22 Supplemental References. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: The scandinavian simvastatin survival study (4s). Lancet. 1994;344:1383-1389. 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. 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. 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 angloscandinavian cardiac outcomes trial--lipid lowering arm (ascot-lla): A multicentre randomised controlled trial. Lancet. 2003;361:1149-1158. 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. 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. 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. 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. 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. 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;18:220228. 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. 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. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 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. 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. 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. 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. 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. 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. 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. 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. 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 noninsulin-dependent diabetes mellitus (aspen). Diabetes Care. 2006;29:1478-1485. 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. 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. 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. 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. 26. 27. 28. 29. 30. 31. 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. 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. 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. 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. 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. 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 placebocontrolled trial. Lancet. 2011;377:2181-2192. 30 Stroke 日本語版 Vol. 7, No. 3 Abstract 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 Hemorrhage A 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. 結果:合計 91,588 例の被験者が実薬群に含まれ,91,215 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 例が対照群に含まれた。実薬投与群と対照群の比較では 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 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 が報告されたスタチンを用いた無作為化比較試験のメ 。ICH のリスクは,低 タ解析を行った。 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 方法:2012 年 1 月 25 日までの Medline,Web of Science, 少の達成程度と関連していなかった。全脳卒中(OR,0.84; および Cochrane Library の文献検索を行った。また,抽 95 % CI,0.78 ∼ 0.91;p < 0.0001 )お よ び 全 死 因 死 亡 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 ( OR,0.92;CI,0.87 ∼ 0.96;p = 0.0007 )は実薬治療群 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 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 化比較試験で,ICH または出血性脳卒中が報告された全試 れなかった。 験を対象とした。主要評価項目は ICH とした。31 件の無 結論:スタチン療法の無作為化比較試験 31 件のメタ解析 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 では,スタチンの実薬治療は ICH の有意な増加と関連し 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 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 かった。 亡の有意な減少が認められた。 Stroke 2012; 43: 2149-2156 ICH- スタチンのメタ解析:ICH 試験名 4S AF-TEXCAPS ALLHAT-LLT ASCOT-LLA ATOZ AURORA CARE CORONA GISSI-P GREACE HPS JUPITER LIPID MEGA PROSPER PROVE-IT TNT SEARCH SPARCL ASPEN GISSI-HF 4D IDEAL MIRACL PATE ACAPS ALERT BONE CLAPT SHARP 試験内の サブグループ プラセボ プラセボ プラセボ プラセボ 低用量 プラセボ プラセボ プラセボ プラセボ プラセボ プラセボ プラセボ プラセボ プラセボ プラセボ 低用量 低用量 低用量 プラセボ プラセボ プラセボ プラセボ 低用量 プラセボ 低用量 プラセボ プラセボ プラセボ プラセボ プラセボ 各試験の統計値 オッズ比 0.20 3.00 3.42 0.55 12.84 1.19 0.33 1.66 2.99 1.00 0.96 0.67 1.89 1.17 0.81 3.94 0.94 0.96 1.68 1.98 3.69 0.64 1.00 0.14 0.60 0.14 0.59 0.74 0.34 1.21 1.08 p値 0.2989 0.5014 0.0158 0.1071 0.0820 0.5608 0.1774 0.2308 0.5019 1.0000 0.8434 0.4415 0.1236 0.6632 0.6490 0.2205 0.8666 0.8871 0.0191 0.4296 0.0454 0.4299 0.9969 0.1992 0.1963 0.1965 0.1817 0.8540 0.5060 0.3916 0.4687 下限 上限 0.01 4.17 0.12 73.62 1.26 9.27 0.26 1.14 0.72 228.14 0.66 2.14 0.07 1.65 0.72 3.80 0.12 73.55 0.06 16.02 0.65 1.41 0.24 1.87 0.84 4.24 0.57 2.41 0.32 2.04 0.44 35.25 0.48 1.87 0.55 1.68 1.09 2.60 0.36 10.85 1.03 13.23 0.21 1.96 0.32 3.11 0.01 2.78 0.28 1.30 0.01 2.75 0.27 1.28 0.03 18.28 0.01 8.34 0.78 1.87 0.88 1.32 脳卒中/合計 実薬 0 / 2,221 1 / 3,304 17 / 5,170 11 / 5,168 6 / 2,265 25 / 1,389 2 / 2,081 15 / 2,514 1 / 2,138 1 / 800 51 / 10,269 6 / 8,901 17 / 4,512 16 / 3,866 8 / 2,891 4 / 2,099 16 / 4,995 24 / 6,031 55 / 2,365 4 / 1,211 11 / 2,285 5 / 619 6 / 4,439 0 / 1,538 11 / 331 0 / 460 10 / 1,050 1 / 485 0 / 112 45 / 4,650 オッズ比および 95% CI 対照薬 2 / 2,223 0 / 3,301 5 / 5,185 20 / 5,137 0 / 2,232 21 / 1,384 6 / 2,078 9 / 2,497 0 / 2,133 1 / 800 53 / 10,267 9 / 8,901 9 / 4,502 14 / 3,966 10 / 2,913 1 / 2,063 17 / 5,006 25 / 6,033 33 / 2,366 2 / 1,199 3 / 2,289 8 / 633 6 / 4,449 3 / 1,548 18 / 334 3 / 459 17 / 1,052 0 / 119 1 / 114 37 / 4,620 0.01 全試験 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。 0.1 実薬が優位 1 10 対照薬が優位 100 Tratamiento con estatinas y riesgo de hemorragia intracerebral Un metanálisis de 31 ensayos controlados y aleatorizados James S. McKinney, MD; William J. Kostis, PhD, MD Antecedentes y objetivo—El tratamiento con estatinas reduce el riesgo de ictus isquémico. En algunos estudios se ha observado 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 Therapy 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 L 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 recurrencias de ictus, pero no una diferencia en los ictus mortales22. En un análisis post hoc del ensayo SPARCL se observó que el tratamiento con atorvastatina se asociaba de manera independiente con un aumento del riesgo de ictus hemorrágico (razón de riesgos, 1,68; IC del 95%, 1,09–2,59)28. Esto era especialmente 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 hipolipemiante y el ictus indicó un aumento significativo de las a 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-coenzima A reductasa (estatinas) es eficaz para reducir la mortalidad cardiovascular mediante la prevención del infarto agudo de miocardio y del ictus isquémico6–24. Es posible que el beneficio clínico no se limite a las propiedades hipolipemiantes de las estatinas sino que sea consecuencia también de otros efectos “pleiotrópicos”, incluidos los antiinflamatorios y antitrombóticos25. 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. 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 113 DOI: 10.1161/STROKEAHA.112.655894 114 Stroke Noviembre 2012 4.098 registros identificados a través de la base de datos 34 registros identificados a través de otras fuentes 31 títulos duplicados 4.101 registros examinados 3.930 excluidos Figura 1. Perfil de la búsqueda bibliográfica. 171 artículos completos examinados en cuanto a la elegibilidad 140 estudios excluidos 50: ausencia de datos sobre el subtipo de ictus 53: informe duplicado de un ensayo 37: revisión/estudio no aleatorizado 31 estudios incluidos en el metanálisis probabilidades de sufrir un ictus hemorrágico con el tratamiento con estatinas (OR, 1,72; IC del 95%, 1,20–2,46)29. Sin embargo, este análisis incluyó tan solo 2 ensayos, el Heart Protection 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 sobre 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 mortalidad 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 seguimiento < 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 alcanzado 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 disponibles del tratamiento con estatinas en los que se presentaban datos sobre la frecuencia del ictus hemorrágico. Métodos Los métodos utilizados en este metanálisis son similares a los descritos 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 metanálisis 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 criterios 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 calidad Los autores extrajeron los datos relativos a la estatina utilizada en el grupo de tratamiento activo, tipo de tratamiento (placebo, dosis bajas de estatina o asistencia habitual) utilizado en el grupo control, criterios de inclusión, 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, investigadores, evaluación de la situación vital), el porcentaje de pérdida del seguimiento y el uso de un análisis por intención de tratar. McKinney y Kostis Mal Tratamiento con estatinas y riesgo de hemorragia intracerebral 115 Tabla. Episodios cerebrovasculares y muerte Ensayo, año Participantes Incluidos (N = 182.803) Todos los ictus Ictus isquémico– Ictus –Tratamiento Todos los ictus Tratamiento isquémico– HIC–Tratamiento activo –Control activo Control activo HIC– Control Otros ictus– Tratamiento activo Ictus mortal– Mortalidad total– Mortalidad Otros ictus– Tratamiento Ictus mortal– Tratamiento total– Control activo Control activo Control 4S, 19946 4.444 44 64 29 49 0 2 15 13 14 12 182 256 AF-TEXCAPS, 6.605 14 17 1 1 1 0 12 16 ... ... 80 77 10.355 209 231 71 83 17 5 121 143 53 56 631 641 10.305 89 121 74 95 11 20 4 6 ... ... 185 212 4.497 28 35 22 31 6 0 0 4 ... ... 130 104 2.773 94 81 57 55 25 21 12 5 40 36 636 660 19988 ALLHAT-LLT, 200212 ASCOT-LLA, 200316 A-to-Z, 200417 AURORA, 200936 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, 4.271 20 19 15 13 1 0 4 6 4 4 72 88 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, 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, 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, 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, 12.064 255 279 233 255 24 25 0 0 57 67 964 970 4.731 265 311 218 274 55 33 7 12 24 41 216 211 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 1.115 358 (0,39) 318 (0,35) 9.768 (10,7) 10.427 (11,4) 200038 GREACE, 200214 200823 200621 200418 201024 SPARCL, 200622 CARDS, 200419 Total de todos los estudios (%) 2.866 (3,13) 3.396 (3,72) 2.039 (2,23) 2.518 (2,76) 462 (0,50) 554 (0,61) 525 (0,57) 562 (0,62) HIC indica hemorragia intracerebral. Síntesis y análisis de los datos Se calcularon las tasas de HIC, total de ictus y mortalidad por todas las causas en los grupos de tratamiento activo y de control. Los análisis 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, utilizando 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 116 Stroke Noviembre 2012 Metanálisis de HIC - Estatinas: HIC Nombre del estudio Análisis estadístico para cada estudio Subgrupo en el estudio 4S PLACEBO Ictus / Total Odds ratio Valor de p Límite inferior Límite Tratamiento superior activo 0,20 0,2989 0,01 4,17 Odds ratio e IC del 95% Control 0 / 2221 2 / 2223 0 / 3301 AF-TEXCAPS PLACEBO 3,00 0,5014 0,12 73,62 1 / 3304 ALLHAT-LLT PLACEBO 3,42 0,0158 1,26 9,27 17 / 5170 5 / 5185 ASCOT-LLA PLACEBO 0,55 0,1071 0,26 1,14 11 / 5168 20 / 5137 ATOZ DOSIS BAJA 12,84 0,0820 0,72 228,14 6 / 2265 0 / 2232 AURORA PLACEBO 1,19 0,5608 0,66 2,14 25 / 1389 21 / 1384 CARE PLACEBO 0,33 0,1774 0,07 1,65 2 / 2081 6 / 2078 CORONA PLACEBO 1,66 0,2308 0,72 3,80 15 / 2514 9 / 2497 0 / 2133 GISSI-P PLACEBO 2,99 0,5019 0,12 73,55 1 / 2138 GREACE PLACEBO 1,00 1,0000 0,06 16,02 1 / 800 1 / 800 HPS PLACEBO 0,96 0,8434 0,65 1,41 51 / 10269 53 / 10267 JUPITER PLACEBO 0,67 0,4415 0,24 1,87 6 / 8901 9 / 8901 LIPID PLACEBO 1,89 0,1236 0,84 4,24 17 / 4512 9 / 4502 MEGA PLACEBO 1,17 0,6632 0,57 2,41 16 / 3866 14 / 3966 PROSPER PLACEBO 0,81 0,6490 0,32 2,04 8 / 2891 10 / 2913 PROVE-IT DOSIS BAJA 3,94 0,2205 0,44 35,25 4 / 2099 1 / 2063 TNT DOSIS BAJA 0,94 0,8666 0,48 1,87 16 / 4995 17 / 5006 SEARCH DOSIS BAJA 0,96 0,8871 0,55 1,68 24 / 6031 25 / 6033 SPARCL PLACEBO 1,68 0,0191 1,09 2,60 55 / 2365 33 / 2366 ASPEN PLACEBO 1,98 0,4296 0,36 10,85 4 / 1211 2 / 1199 GISSI-HF PLACEBO 3,69 0,0454 1,03 13,23 11 / 2285 3 / 2289 4D PLACEBO 0,64 0,4299 0,21 1,96 5 / 619 8 / 633 IDEAL DOSIS BAJA 1,00 0,9969 0,32 3,11 6 / 4439 6 / 4449 MIRACL PLACEBO 0,14 0,1992 0,01 2,78 0 / 1538 3 / 1548 PATE DOSIS BAJA 0,60 0,1963 0,28 1,30 11 / 331 18 / 334 ACAPS PLACEBO 0,14 0,1965 0,01 2,75 0 / 460 3 / 459 ALERT PLACEBO 0,59 0,1817 0,27 1,28 10 / 1050 17 / 1052 0 / 119 BONE PLACEBO 0,74 0,8540 0,03 18,28 1 / 485 CLAPT PLACEBO 0,34 0,5060 0,01 8,34 0 / 112 1 / 114 SHARP 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 Favorable al tratamiento activo 10 100 Favorable al control Todos los estudios Figura 2. Gráfico de Forrest del metanálisis con efectos aleatorios de ensayos aleatorizados de las estatinas y la hemorragia intracerebral. 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 sensibilidad Se 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 consiguiente, con objeto de evitar la inclusión de una OR no definida, 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 influido 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 hemorrágico convergía con la adición iterativa de estudios progresivamente más grandes. Resultados Se identificaron 31 ensayos del tratamiento con estatinas en los que se incluyó en la asignación aleatoria del tratamiento a un total de 182.803 participantes, y se incluyeron dichos ensayos en el análisis6–24,36–47. De los 31 ECA, 6 estudios11,17,18,20,24,42 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 placebo o con la “asistencia habitual”. A un total de 91.588 participantes 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 McKinney y Kostis Mal Tratamiento con estatinas y riesgo de hemorragia intracerebral 117 Metanálisis de HIC - Estatinas: Total de ictus Nombre del estudio Análisis estadístico para cada estudio Subgrupo en el estudio Odds ratio 4S PLACEBO 0,68 Valor de p 0,0533 Límite inferior 0,46 Ictus / Total Límite Tratamiento superior activo 1,01 44 / 2221 Odds ratio e IC del 95% Control 64 / 2223 AF-TEXCAPS PLACEBO 0,82 0,5880 0,40 1,67 14 / 3304 17 / 3301 ALLHAT-LLT PLACEBO 0,90 0,2982 0,75 1,09 209 / 5170 231 / 5185 121 / 5137 ASCOT-LLA PLACEBO 0,73 0,0234 0,55 0,96 89 / 5168 ATOZ DOSIS BAJA 0,79 0,3448 0,48 1,30 28 / 2265 35 / 2232 AURORA PLACEBO 1,17 0,3223 0,86 1,59 94 / 1389 81 / 1384 CARE PLACEBO 0,68 0,0340 0,48 0,97 54 / 2081 78 / 2078 CORONA PLACEBO 0,86 0,2138 0,67 1,09 126 / 2514 145 / 2497 19 / 2133 GISSI-P PLACEBO 1,05 0,8780 0,56 1,97 20 / 2138 GREACE PLACEBO 0,52 0,1197 0,23 1,18 9 / 800 17 / 800 HPS PLACEBO 0,75 0,0000 0,66 0,85 444 / 10269 585 / 10267 JUPITER PLACEBO 0,51 0,0019 0,34 0,78 33 / 8901 64 / 8901 LIPID PLACEBO 0,81 0,0251 0,68 0,97 224 / 4512 272 / 4502 MEGA PLACEBO 0,83 0,3151 0,57 1,20 50 / 3866 62 / 3966 PROSPER PLACEBO 1,04 0,7533 0,81 1,33 135 / 2891 131 / 2913 PROVE-IT DOSIS BAJA 1,09 0,7928 0,58 2,03 21 / 2099 19 / 2063 TNT DOSIS BAJA 0,75 0,0209 0,59 0,96 117 / 4995 155 / 5006 SEARCH DOSIS BAJA 0,91 0,2900 0,77 1,08 255 / 6031 279 / 6033 SPARCL PLACEBO 0,83 0,0416 0,70 0,99 265 / 2365 311 / 2366 CARDS PLACEBO 0,53 0,0184 0,31 0,90 21 / 1429 39 / 1412 ASPEN PLACEBO 0,88 0,6022 0,55 1,41 34 / 1211 38 / 1199 66 / 2289 GISSI-HF PLACEBO 1,25 0,1786 0,90 1,74 82 / 2285 4D PLACEBO 1,41 0,0979 0,94 2,12 59 / 619 44 / 633 IDEAL DOSIS BAJA 0,87 0,2012 0,69 1,08 151 / 4439 174 / 4449 MIRACL PLACEBO 0,50 0,0507 0,25 1,00 12 / 1538 24 / 1548 PATE DOSIS BAJA 0,60 0,1963 0,28 1,30 11 / 331 18 / 334 ACAPS PLACEBO 0,09 0,1030 0,00 1,63 0 / 460 5 / 459 ALERT PLACEBO 1,03 0,8667 0,76 1,39 93 / 1050 91 / 1052 0 / 119 BONE PLACEBO 0,74 0,8540 0,03 18,28 1 / 485 CLAPT PLACEBO 0,34 0,5060 0,01 8,34 0 / 112 1 / 114 SHARP PLACEBO 0,80 0,0356 0,65 0,99 171 / 4650 210 / 4620 0,84 0,0000 0,78 0,91 0,5 1 Favorable al tratamiento activo 2 Favorable al control Todos los estudios Figura 3. Gráfico de Forrest del metanálisis con efectos aleatorios de ensayos aleatorizados de las estatinas y el total de ictus. 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 momento 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 intracerebral Se 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 heterogeneidad (interacción) fue de 0,38, Q = 30,705 y grados de libertad = 29. No hubo diferencias estadísticamente significativas 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 resultados 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 iteraciones 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 activo 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 118 Stroke Noviembre 2012 Metanálisis de HIC - Estatinas: Muerte Nombre del estudio 4S Análisis estadístico para cada estudio Subgrupo en el estudio PLACEBO Odds ratio Valor de p 0,69 0,0002 Límite inferior 0,56 Ictus / Total Límite Tratamiento superior activo 0,84 182 / 2221 Odds ratio e IC del 95% Control 256 / 2223 AF-TEXCAPS PLACEBO 1,04 0,8130 0,76 1,43 80 / 3304 77 / 3301 ALLHAT-LLT PLACEBO 0,99 0,8071 0,88 1,11 631 / 5170 641 / 5185 212 / 5137 ASCOT-LLA PLACEBO 0,86 0,1493 0,71 1,05 185 / 5168 ATOZ DOSIS BAJA 1,25 0,1036 0,96 1,62 130 / 2265 104 / 2232 AURORA PLACEBO 0,93 0,3162 0,80 1,08 636 / 1389 660 / 1384 CARE PLACEBO 0,91 0,3791 0,74 1,12 180 / 2081 196 / 2078 CORONA PLACEBO 0,93 0,2650 0,83 1,05 728 / 2514 759 / 2497 GISSI-P PLACEBO 0,81 0,1928 0,59 1,11 72 / 2138 88 / 2133 GREACE PLACEBO 0,56 0,0309 0,33 0,95 23 / 800 40 / 800 HPS PLACEBO 0,86 0,0003 0,80 0,93 1328 / 10269 1507 / 10267 JUPITER PLACEBO 0,80 0,0189 0,66 0,96 198 / 8901 247 / 8901 LIPID PLACEBO 0,77 0,0000 0,69 0,86 717 / 4512 888 / 4502 MEGA PLACEBO 0,71 0,0532 0,50 1,00 55 / 3866 79 / 3966 PROSPER PLACEBO 0,98 0,8393 0,83 1,16 298 / 2891 305 / 2913 PROVE-IT DOSIS BAJA 0,68 0,0458 0,46 0,99 46 / 2099 66 / 2063 TNT DOSIS BAJA 1,01 0,9096 0,85 1,20 284 / 4995 282 / 5006 SEARCH DOSIS BAJA 0,99 0,8879 0,90 1,09 964 / 6031 970 / 6033 SPARCL PLACEBO 1,03 0,7962 0,84 1,25 216 / 2365 211 / 2366 82 / 1412 CARDS PLACEBO 0,72 0,0617 0,51 1,02 61 / 1429 ASPEN PLACEBO 1,02 0,9084 0,72 1,44 70 / 1211 68 / 1199 GISSI-HF PLACEBO 1,03 0,6431 0,91 1,17 657 / 2285 644 / 2289 4D PLACEBO 0,90 0,3628 0,72 1,13 297 / 619 320 / 633 IDEAL DOSIS BAJA 0,98 0,7832 0,84 1,14 366 / 4439 374 / 4449 MIRACL PLACEBO 0,95 0,7507 0,67 1,34 64 / 1538 68 / 1548 PATE DOSIS BAJA 0,69 0,3056 0,34 1,40 14 / 331 20 / 334 ACAPS PLACEBO 0,12 0,0485 0,02 0,99 1 / 460 8 / 459 ALERT PLACEBO 1,04 0,7357 0,81 1,34 143 / 1050 138 / 1052 CLAPT PLACEBO 0,20 0,3006 0,01 4,21 0 / 112 2 / 114 SHARP PLACEBO 1,02 0,6336 0,93 1,13 1142 / 4650 1115 / 4620 0,92 0,0007 0,87 0,96 0,5 1 Favorable al tratamiento activo 2 Favorable al control 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. 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/control) 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 ictus Se produjeron en total 6.262 ictus en este metanálisis. La tasa global de ictus fue del 3,13% en el grupo de tratamiento activo frente al 3,72% en el grupo control. El tratamiento activo produjo una reducción significativa en el número total de ictus (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 causas Se 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). Discusión Los estudios epidemiológicos han descrito un aumento de las tasas de ictus hemorrágicos y de mortalidad debida a HIC 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 importante 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 tratados con simvastatina51. En tres metanálisis previos del tratamiento 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 describió 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 estatinas 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 sugiere 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. Además 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 estatinas podrían explicar un teórico aumento de las complicaciones hemorrágicas. Al estratificar nuestro análisis según el tipo de prevención, hubo una tendencia no significativa al aumento de las probabilidades 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 análisis 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 isquémico52. 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 hemorragia 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 estatinas, 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 posible que haya variables de estos no tenidas en cuenta que influyeran 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. Aunque 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 publicados en lenguas distintas del inglés, que quedaran excluidos. Aunque incluimos todos los ensayos pertinentes, con independencia 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 significativo 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 pudiera haber. Estos resultados respaldan las recomendaciones actuales para la prescripción de estatinas en los pacientes en los que, por lo demás, son apropiados. Conclusiones En este metanálisis de 31 ECA, el tratamiento con estatinas no se asoció a un aumento significativo de las probabilidades 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 financiación Este estudio fue financiado en parte por la Robert Wood Johnson Foundation y la Schering-Plough Foundation. Ninguna. Declaraciones Bibliografía 1. Iso H, Jacobs DR Jr, Wentworth D, Neaton JD, Cohen JD. Serum cholesterol levels and six-year mortality from stroke in 350 977 men screened 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. Different risk factors for different stroke subtypes: association of blood pressure, cholesterol, and antioxidants. Stroke. 1999;30:2535–2540. 3. Horenstein RB, Smith DE, Mosca L. Cholesterol predicts stroke mortality in 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- 1. Iso H, Jacobs DR Jr, Wentworth D, Neaton 120 Stroke Noviembre 2012 20. 2. 21. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. JD, Cohen JD. Serum cholesterol levels and six-year mortality from stroke in 350 977 men screened for the multiple risk factor intervention trial. N Engl J Med. 1989;320:904 –910. Leppala JM, Virtamo J, Fogelholm R, Albanes D, Heinonen OP. Different risk factors for different stroke subtypes: association of blood pressure, cholesterol, and antioxidants. Stroke. 1999;30:2535–2540. Horenstein RB, Smith DE, Mosca L. Cholesterol predicts stroke mortality in the women’s pooling project. Stroke. 2002;33:1863–1868. Zhang X, Patel A, Horibe H, Wu Z, Barzi F, Rodgers A, et al. Cholesterol, coronary heart disease, and stroke in the Asia Pacific region. Int J Epidemiol. 2003;32:563–572. 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. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344:1383–1389. 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. 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. 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. 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. 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 AntiAtherosclerosis Trial in the Elderly (PATE). J Atheroscler Thromb. 2001;8:33– 44. Major outcomes in moderately hypercholesterolemic, hypertensive patients randomized to pravastatin vs usual care: the Antihypertensive and LipidLowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT). JAMA. 2002;288:2998 –3007. 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. 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 CoronaryHeart-Disease Evaluation (GREACE) study. Curr Med Res Opin. 2002; 18:220 –228. 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. 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. 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. 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. 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. 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. 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. Amarenco P, Bogousslavsky J, Callahan A III, 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. Ridker PM, Danielson E, Fonseca FA, Genest J, Gotto AM Jr, Kastelein 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): multicentre randomised placebo-controlled trial. Lancet. 2004;364:685– 696. 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. 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. Amarenco P, Bogousslavsky J, Callahan A III, 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. 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. 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. Amarenco P, Labreuche J. Lipid management in the prevention of stroke: review and updated meta-analysis of statins for stroke prevention. Lancet Neurol. 2009;8:453– 463. Cholesterol, diastolic blood pressure, and stroke: 13 000 strokes in 450 000 people in 45 prospective cohorts. Prospective studies collaboration. Lancet. 1995;346:1647–1653. 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 studies with 55 000 vascular deaths. Lancet. 2007;370:1829 –1839. Goldstein MR, Mascitelli L, Pezzetta F. Hemorrhagic stroke in the stroke prevention by Aggressive Reduction in Cholesterol Levels Study. Neurology. 2009;72:1448; author reply 1448 –1449. Manktelow BN, Potter JF. Interventions in the management of serum lipids for preventing stroke recurrence. Cochrane Database Syst Rev. 2009;3:CD002091. Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170 000 participants in 26 randomised trials. Lancet. 2010;376:1670 –1681. Kostis WJ, Moreyra AE, Cheng JQ, Dobrzynski JM, Kostis JB. Continuation of mortality reduction after the end of randomized therapy in clinical trials of lipid-lowering therapy. J Clin Lipidol. 2011;5:97–104. Borenstein M. Introduction to Meta-Analysis. Chichester, UK: John Wiley & Sons; 2009. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56:455– 463. Rosenthal R. The ’file drawer problem’ and tolerance for null results. Psychol Bull. 1979;86:638 – 641. Orwin RG. A fail-safe N for effect size in meta-analysis. J Educ Stat. 1983;8:157–159. 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. 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. 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. 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. 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. 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. 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. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. ezetimibe in patients with chronic kidney disease (study of heart and renal protection): a randomised placebo-controlled trial. Lancet. 2011;377: McKinney y Kostis Mal Tratamiento con2181–2192. estatinas y riesgo de hemorragia intracerebral 121 48. Sturgeon JD, Folsom AR, Longstreth WT Jr, Shahar E, Rosamond WD, Cushman M. Risk factors for intracerebral hemorrhage in a pooled prospective study. Stroke. 2007;38:2718 –2725. 49. Iribarren C, Reed DM, Burchfiel CM, Dwyer JH. Serum total cholesterol Furberg CD, Adams HP Jr, Applegate WB, Byington RP, Espeland MA, and mortality. Confounding factors and risk modification in JapaneseHartwell T, et al. Effect of lovastatin on early carotid atherosclerosis and American men. JAMA. 1995;273:1926 –1932. cardiovascular events. Asymptomatic Carotid Artery Progression Study 50. Bjorkhem I, Meaney S. Brain cholesterol: long secret life behind a barrier. (ACAPS) research group. Circulation. 1994;90:1679 –1687. Arterioscler Thromb Vasc Biol. 2004;24:806 – 815. Holdaas H, Fellstrom B, Jardine AG, Holme I, Nyberg G, Fauchald P, et 51. Collins R, Armitage J, Parish S, Sleight P, Peto R. Effects of cholesteral. Effect of fluvastatin on cardiac outcomes in renal transplant recipients: ol-lowering with simvastatin on stroke and other major vascular events in a multicentre, randomised, placebo-controlled trial. Lancet. 2003;361: 20 536 people with cerebrovascular disease or other high-risk conditions. 2024 –2031. Lancet. 2004;363:757–767. Bone HG, Kiel DP, Lindsay RS, Lewiecki EM, Bolognese MA, Leary 52. Hackam DG, Austin PC, Huang A, Juurlink DN, Mamdani MM, Paterson ET, et al. Effects of atorvastatin on bone in postmenopausal women with JM, et al. Statins and intracerebral hemorrhage: a retrospective cohort dyslipidemia: a double-blind, placebo-controlled, dose-ranging trial. study. Arch Neurol. 2012;69:39 – 45. J Clin Endocrinol Metab. 2007;92:4671– 4677. 53. Hackam DG, Woodward M, Newby LK, Bhatt DL, Shao M, Smith EE, Kleemann A, Eckert S, von Eckardstein A, Lepper W, Schernikau U, et al. Statins and intracerebral hemorrhage: collaborative systematic Gleichmann U, et al. Effects of lovastatin on progression of non-dilated review and meta-analysis. Circulation. 2011;124:2233–2242. and dilated coronary segments and on restenosis in patients after PTCA. 54. Serebruany VL, Miller M, Pokov AN, Malinin AI, Lowry DR, Tanguay The Cholesterol Lowering Atherosclerosis PTCA Trial (CLAPT). Eur JF, et al. Effect of statins on platelet par-1 thrombin receptor in patients Heart J. 1999;20:1393–1406. with the metabolic syndrome (from the PAR-1 Inhibition by Statins Baigent C, Landray MJ, Reith C, Emberson J, Wheeler DC, Tomson C, [PARIS] study). Am J Cardiol. 2006;97:1332–1336. et al. The effects of lowering LDL cholesterol with simvastatin plus 55. Serebruany VL, Malinin AI, Hennekens CH. Statins increase risk of ezetimibe in patients with chronic kidney disease (study of heart and renal hemorrhagic stroke by inhibition of the PAR-1 receptor. Cerebrovasc protection): a randomised placebo-controlled trial. Lancet. 2011;377: Dis. 2007;24:477– 479. 2181–2192. 56. Westover MB, Bianchi MT, Eckman MH, Greenberg SM. Statin use Sturgeon JD, Folsom AR, Longstreth WT Jr, Shahar E, Rosamond WD, following intracerebral hemorrhage: a decision analysis. Arch Neurol. Cushman M. Risk factors for intracerebral hemorrhage in a pooled pro2010;68:573–579. spective study. Stroke. 2007;38:2718 –2725. Iribarren C, Reed DM, Burchfiel CM, Dwyer JH. Serum total cholesterol and mortality. Confounding factors and risk modification in JapaneseAmerican men. JAMA. 1995;273:1926 –1932. Bjorkhem I, Meaney S. Brain cholesterol: long secret life behind a barrier. Arterioscler Thromb Vasc Biol. 2004;24:806 – 815. Collins R, Armitage J, Parish S, Sleight P, Peto R. Effects of cholesterol-lowering with simvastatin on stroke and other major vascular events in 20 536 people with cerebrovascular disease or other high-risk conditions. Lancet. 2004;363:757–767. Hackam DG, Austin PC, Huang A, Juurlink DN, Mamdani MM, Paterson JM, et al. Statins and intracerebral hemorrhage: a retrospective cohort study. Arch Neurol. 2012;69:39 – 45. Hackam DG, Woodward M, Newby LK, Bhatt DL, Shao M, Smith EE, et al. Statins and intracerebral hemorrhage: collaborative systematic review and meta-analysis. Circulation. 2011;124:2233–2242. Serebruany VL, Miller M, Pokov AN, Malinin AI, Lowry DR, Tanguay JF, et al. Effect of statins on platelet par-1 thrombin receptor in patients with the metabolic syndrome (from the PAR-1 Inhibition by Statins [PARIS] study). Am J Cardiol. 2006;97:1332–1336. Serebruany VL, Malinin AI, Hennekens CH. Statins increase risk of hemorrhagic stroke by inhibition of the PAR-1 receptor. Cerebrovasc Dis. 2007;24:477– 479. Westover MB, Bianchi MT, Eckman MH, Greenberg SM. Statin use following intracerebral hemorrhage: a decision analysis. Arch Neurol. 2010;68:573–579.