(Stroke. 1997;28:946-950.)
© 1997 American Heart Association, Inc.
Articles |
From the Departments of General Internal Medicine and Clinical Epidemiology (R.G.J.W.), Leiden (Netherlands) University Medical Centre.
Correspondence to Dr G.J. Blauw, MD, PhD, Department of General Internal Medicine, Section of Gerontology and Geriatrics, C1-R41, Leiden University Medical Centre, Albinusdreef 2, PO Box 9600, 2300 RC Leiden, Netherlands.
| Abstract |
|---|
|
|
|---|
Methods The studies were identified using the Medline CD+ and Current Contents databases from January 1980 through May 1996, inclusive. All studies were evaluated on the use of a placebo control, monotherapy, and double blindness. When the type of stroke or the occurrence of clinical events or adverse effects were incompletely or not reported, the investigators were contacted personally. For each trial, the number of strokes in the treatment arm was compared with the number of strokes expected on all observations under the assumption that drug treatment had no effect.
Results A total of 462 strokes among 20 438 participants in 13 trials could be analyzed. A total of 181 strokes were observed in patients randomized to treatment with an HMG-CoA reductase inhibitor and 261 strokes in patients randomized to placebo. A lower than expected number of strokes was observed in the treatment groups of all but one trial (P=.001). Treatment with an HMG-CoA reductase inhibitor led to an overall risk reduction of 31% (odds ratio, 0.69; 95% confidence interval, 0.57 to 0.83).
Conclusions The combined data suggest that treatment with HMG-CoA reductase inhibitors prevents stroke in middle-aged persons. Because stroke is especially common in older age, these data reinforce the need for clinical trials to evaluate the effect of HMG-CoA reductase inhibitors in preventing stroke in the elderly.
Key Words: cholesterol HMG-CoA reductase inhibitors meta-analysis stroke prevention
| Introduction |
|---|
|
|
|---|
In contrast to the data on the reduction of ischemic heart disease, little is known about the effects of these drugs on other manifestations of atherosclerotic disease, such as stroke. To date, only in the 4S and CARE studies was a reduction of stroke observed.15 17 In other trials, such a beneficial effect could not be detected, most likely because the incidence of stroke in the studied groups was too low. Most of the participants were middle-aged men with signs and symptoms of coronary heart disease. Pooling the data from single trials may overcome this problem, as was done in the pravastatin atherosclerosis intervention program.6 This analysis suggested that pravastatin may reduce stroke risk.
Because stroke is a major cause of morbidity and mortality in older age and data from the single studies on the effect of the HMG-CoA reductase inhibitors are not conclusive yet, we combined the data on the occurrence of stroke in the randomized, placebo-controlled, double-blind trials with HMG-CoA reductase inhibitors published so far.
| Materials and Methods |
|---|
|
|
|---|
For each trial, the number of strokes in the treatment arm was compared with the number of strokes expected on all observations under the assumption that drug treatment had no effect.17 The number of expected strokes can be calculated from the marginals of a 2x2 table of active- and placebo-treated subjects. For each trial, the number of observed strokes (O) minus the number of expected strokes (E), as well as the variance, can then be calculated and summed over all trials. The formal test statistic whether the number of O-E strokes differs from zero is the total of O-E over all trials divided by the standard deviation (square root of the summed variances), which follows a Z distribution. The pooled OR for stroke in treated subjects was approximated by the exponent of Z divided by the standard deviation.18
Heterogeneity between studies was evaluated with
2 testing, subtracting Z2
from the sum of (O-E)2/variance of (O-E) for each
trial.18 Heterogeneity was also tested
after pooling the data of the small trials, since a large number of
small studies can obscure the true heterogeneity
between a few large trials.
| Results |
|---|
|
|
|---|
From the 114 studies included in the analysis, additional information was obtained for 16 trials. One investigator did not respond.21 The investigators of four trials could not provide additional information on strokes because the original data were no longer accessible.14 22 23 24 One of these was the EXCEL Study, an efficacy and safety trial including 8245 participants with a follow-up of 48 weeks.14 15 In this trial, one hemorrhagic stroke was observed but could not be traced to one of the treatment arms (personal communication, R.H. Bradford and A. Langendörfer, 1996). The three other studies were small trials that included some 212 participants.22 23 24
In 101 trials with a total of 5670 person-years, no strokes were
observed (Table 1
). These trials were not included in
the analysis because they contribute no information on the
possible effect of HMG-CoA reductase inhibitors on stroke
risk. The participants in these trials were for the greater part
healthy volunteers and asymptomatic subjects with
dyslipidemia aged 20 to 65 years.
|
A total of 442 strokes in 20 438 subjects with an observation period
of 87 481 person-years in 13 trials could be
analyzed.2 3 4 5 6 7 8 9 10 11 12 15 16 17 Among these trials reporting
on stroke, the Efficacy and Safety of Pravastatin
(ESP)2 represents part of a larger study,
investigating the efficacy and safety of pravastatin. Nine
trials were designed to determine the effect on the regression of
atherosclerosis.3 4 5 6 7 8 9 10 11 12 The
4S,15 WOSCOP,16 and CARE
studies17 were principally designed to determine the
effect on clinical end points. The baseline characteristics of the
trials reporting on stroke are shown in Table 2
.
|
Overall, a total of 181 strokes were observed in the patients
randomized to treatment with an HMG-CoA reductase inhibitor
and 261 strokes in the patients randomized to placebo (Table 3
). A lower than expected number of strokes was observed
in all but one trial. Before pooling, we tested for
heterogeneity between the treatment effects in the
different trials. The test statistic was not significant
(
212=13.1, P=.3),
indicating that the separate studies estimate a common underlying risk
reduction. When the three large clinical intervention
trials15 16 17 were analyzed together with the
combined data of the 10 smaller trials,2 3 4 5 6 7 8 9 10 11 12 the test
statistic became significant (
23=8.3,
P<.05). Heterogeneity between the 10
smaller trials was absent (
29=5.0,
P>.5). Thus, heterogeneity was present
between the clinical intervention trials; the 4S, CARE, and WOSCOP
studies; and the other trials,2 3 4 5 6 7 8 9 10 11 12 which is also reflected
by the effect of the HMG-CoA reductase inhibitors on stroke
risk (see below).
|
The reduction in total strokes in the subjects treated with HMG-CoA reductase inhibitors was highly significant (Z=3.9, P<.001). The pooled data amounted to an overall risk reduction of 31% during treatment with an HMG-CoA reductase inhibitor (OR, 0.69; 95% CI, 0.57 to 0.83). The risk reduction in the 10 smaller trials was 72% (OR, 0.28; 95% CI, 0.14 to 0.56). A similar result was obtained from the six smaller trials using pravastatin (OR, 0.36; 95% CI, 0.16 to 0.81). The 4S15 and CARE studies17 showed a significant reduction of 30% and 31% in stroke risk, whereas in the WOSCOP study16 treatment decreased stroke risk by only 11%.
The total number of fatal strokes was considerably smaller and not different between active- and placebo-treated subjects (30 and 27 patients, respectively; OR, 1.09; 95% CI, 0.65 to 1.84). When the data were analyzed after the exclusion of transient ischemic attacks in four studies,2 4 9 15 the risk reduction was 30% (OR, 0.70; 95% CI, 0.57 to 0.85).
| Discussion |
|---|
|
|
|---|
A statistically significant risk reduction was observed in two of the three large clinical trials, 4S15 and CARE,17 as well as in the combination of the smaller trials.2 3 4 5 6 7 8 9 10 11 12 In contrast, only a modest effect of pravastatin was observed in the WOSCOP study.16 Formal heterogeneity testing confirmed that this effect of pravastatin in the WOSCOP study might be at odds with the overall estimate of 31% reduction in stroke risk. The fact that the CARE study, as well as the combined data from all other trials with pravastatin, shows a significant reduction of stroke risk makes it unlikely that the lack of effect in WOSCOP is drug specific. A possible explanation for the observed difference is that the WOSCOP study was basically designed as a primary prevention trial, whereas the other trials were secondary prevention trials. The subjects enrolled in WOSCOP differ significantly from those in the 4S,15 CARE,17 and other studies.2 3 4 5 6 7 8 9 10 11 12 The participants in WOSCOP were not selected on clinical end points of atherosclerosis but on risk factors. It may thus be hypothesized that other pathophysiological mechanisms underlie the occurrence of strokes in the participants of the 4S, CARE, and WOSCOP studies.
The present findings can be interpreted as evidence for cholesterol being a causal determinant of stroke. In the recent analysis of cohort studies, an overall association between total blood cholesterol and stroke was absent.25 Also noteworthy, most of the observations in these cohort studies concerned fatal stroke. In accordance with this, we did not find a risk reduction by statins on fatal stroke.
Alternatively, in view of the findings of the above-mentioned analysis of cohort studies,25 the presented data can also be interpreted as evidence for a cholesterol-independent effect of statins on cerebrovascular disease. Experimental data indicate that the statins may inhibit cell proliferation and G proteins.26 27 28 In animal experiments, it has been shown that lovastatin has a direct vasodilating effect.29 Several clinical observations corroborate these experimental data in the published clinical intervention trials. The reduction of coronary events starts almost immediately during treatment,15 16 and the risk reduction is independent of the cholesterol level at baseline18 30 and is not correlated with the regression of coronary atherosclerosis.8
The presented data suggest that some 40 strokes may be prevented when treating 10 000 patients with coronary artery disease with statins for a considerable length of time. These patients are likely a subset of the population susceptible for stroke. Among this subset of patients, a large proportion of the strokes is caused by ischemia, likely due to ongoing atherosclerotic disease. As such, the effect of lipid control is likely to be beneficial. Age is a powerful risk factor for atherosclerotic disease, and the incidence of ischemic stroke increases progressively after the age of 70 years. Therefore, a large number of elderly persons with atherosclerosis may benefit from treatment with statins. In contrast to expectation, however, hypercholesterolemia is not an independent risk factor for cardiovascular disease beyond the age of 70 years. These data reinforce the need for clinical trials to evaluate the effect of statins to prevent stroke and coronary artery disease in the elderly with clinical signs and symptoms of atherosclerosis.
| Selected Abbreviations and Acronyms |
|---|
|
Received November 21, 1996; revision received January 31, 1997; accepted February 19, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. R. Caplan A 70-Year-Old Man With a Transient Ischemic Attack: Review of Internal Carotid Artery Stenosis JAMA, July 2, 2008; 300(1): 81 - 90. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Rodriguez-Yanez, J. Agulla, R. Rodriguez-Gonzalez, T. Sobrino, and J. Castillo Review: Statins and stroke Therapeutic Advances in Cardiovascular Disease, June 1, 2008; 2(3): 157 - 166. [Abstract] [PDF] |
||||
![]() |
B. A. Golomb, J. E. Dimsdale, H. L. White, J. B. Ritchie, and M. H. Criqui Reduction in Blood Pressure With Statins: Results From the UCSD Statin Study, a Randomized Trial Arch Intern Med, April 14, 2008; 168(7): 721 - 727. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. L. Air and B. M. Kissela Diabetes, the Metabolic Syndrome, and Ischemic Stroke: Epidemiology and possible mechanisms Diabetes Care, December 1, 2007; 30(12): 3131 - 3140. [Full Text] [PDF] |
||||
![]() |
S. Khella and M. B. Bleicher Stroke and Its Prevention in Chronic Kidney Disease Clin. J. Am. Soc. Nephrol., November 1, 2007; 2(6): 1343 - 1351. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Gnavi, A. Migliardi, M. Demaria, A. Petrelli, A. Caprioglio, and G. Costa Statins prescribing for the secondary prevention of ischaemic heart disease in Torino, Italy. A case of ageism and social inequalities Eur J Public Health, October 1, 2007; 17(5): 492 - 496. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. K. Sampson, M. A. Pfeffer, J. J.V. McMurray, Y. Lokhnygina, H. D. White, S. D. Solomon, and for the Valsartan in Acute Myocardial iNfarcTion ( Predictors of stroke in high-risk patients after acute myocardial infarction: insights from the VALIANT trial Eur. Heart J., March 2, 2007; 28(6): 685 - 691. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Aboyans, L. Labrousse, P. Lacroix, J. Guilloux, S. Sekkal, A. Le Guyader, E. Cornu, and M. Laskar Predictive factors of stroke in patients undergoing coronary bypass grafting: statins are protective. Eur. J. Cardiothorac. Surg., August 1, 2006; 30(2): 300 - 304. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Stoll and M. Bendszus Inflammation and Atherosclerosis: Novel Insights Into Plaque Formation and Destabilization Stroke, July 1, 2006; 37(7): 1923 - 1932. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Amarenco, J. Labreuche, P. Lavallee, and P.-J. Touboul Statins in Stroke Prevention and Carotid Atherosclerosis: Systematic Review and Up-to-Date Meta-Analysis Stroke, December 1, 2004; 35(12): 2902 - 2909. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Li, R. Higdon, W. A. Kukull, E. Peskind, K. Van Valen Moore, D. Tsuang, G. van Belle, W. McCormick, J. D. Bowen, L. Teri, et al. Statin therapy and risk of dementia in the elderly: A community-based prospective cohort study Neurology, November 9, 2004; 63(9): 1624 - 1628. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Ovbiagele, J. L. Saver, A. Fredieu, S. Suzuki, N. McNair, A. Dandekar, T. Razinia, and C. S. Kidwell PROTECT: A coordinated stroke treatment program to prevent recurrent thromboembolic events Neurology, October 12, 2004; 63(7): 1217 - 1222. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. Schonbeck and P. Libby Inflammation, Immunity, and HMG-CoA Reductase Inhibitors: Statins as Antiinflammatory Agents? Circulation, June 1, 2004; 109(21_suppl_1): II-18 - II-26. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Marti-Fabregas, M. Gomis, A. Arboix, A. Aleu, J. Pagonabarraga, R. Belvis, D. Cocho, J. Roquer, A. Rodriguez, M. D. Garcia, et al. Favorable Outcome of Ischemic Stroke in Patients Pretreated with Statins Stroke, May 1, 2004; 35(5): 1117 - 1121. [Abstract] [Full Text] [PDF] |
||||
![]() |
W.-Y. ONG and B. HALLIWELL Iron, Atherosclerosis, and Neurodegeneration: A Key Role for Cholesterol in Promoting Iron-Dependent Oxidative Damage? Ann. N.Y. Acad. Sci., March 1, 2004; 1012(1): 51 - 64. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Zuliani, A. Cherubini, A. R. Atti, A. Ble, C. Vavalle, F. Di Todaro, C. Benedetti, S. Volpato, M. G. Marinescu, U. Senin, et al. Low Cholesterol Levels Are Associated With Short-Term Mortality in Older Patients With Ischemic Stroke J. Gerontol. A Biol. Sci. Med. Sci., March 1, 2004; 59(3): M293 - M297. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. H. Danton and W. D. Dietrich The Search for Neuroprotective Strategies in Stroke AJNR Am. J. Neuroradiol., February 1, 2004; 25(2): 181 - 194. [Full Text] [PDF] |
||||
![]() |
V. G. Athyros, A. A. Papageorgiou, A. N. Symeonidis, T. P. Didangelos, A. N. Pehlivanidis, V. I. Bouloukos, and D. P. Mikhailidis Early Benefit from Structured Care with Atorvastatin in Patients with Coronary Heart Disease and Diabetes Mellitus: A Subgroup Analysis of the GREek Atorvastatin and Coronary Heart Disease Evaluation (GREACE) Study Angiology, November 1, 2003; 54(6): 679 - 690. [Abstract] [PDF] |
||||
![]() |
J.-N. Trochu, S. Mital, X.-p. Zhang, X. Xu, M. Ochoa, J. K Liao, F. A Recchia, and T. H Hintze Preservation of NO production by statins in the treatment of heart failure Cardiovasc Res, November 1, 2003; 60(2): 250 - 258. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Pierre-Paul and V. Gahtan Noncholesterol-Lowering Effects of Statins Vascular and Endovascular Surgery, September 1, 2003; 37(5): 301 - 313. [Abstract] [PDF] |
||||
![]() |
W. Balduini, E. Mazzoni, S. Carloni, M. G. De Simoni, C. Perego, L. Sironi, and M. Cimino Prophylactic but Not Delayed Administration of Simvastatin Protects Against Long-Lasting Cognitive and Morphological Consequences of Neonatal Hypoxic-Ischemic Brain Injury, Reduces Interleukin-1{beta} and Tumor Necrosis Factor-{alpha} mRNA Induction, and Does Not Affect Endothelial Nitric Oxide Synthase Expression Stroke, August 1, 2003; 34(8): 2007 - 2012. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. W. E. Weverling-Rijnsburger, I. J. A. M. Jonkers, E. van Exel, J. Gussekloo, and R. G. J. Westendorp High-Density vs Low-Density Lipoprotein Cholesterol as the Risk Factor for Coronary Artery Disease and Stroke in Old Age Arch Intern Med, July 14, 2003; 163(13): 1549 - 1554. [Abstract] [Full Text] [PDF] |
||||
![]() |
M R Law, N J Wald, and A R Rudnicka Quantifying effect of statins on low density lipoprotein cholesterol, ischaemic heart disease, and stroke: systematic review and meta-analysis BMJ, June 25, 2003; 326(7404): 1423. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. M Buckley Review: Lipids and stroke The British Journal of Diabetes & Vascular Disease, May 1, 2003; 3(3): 170 - 176. [Abstract] [PDF] |
||||
![]() |
J.-C. Corvol, A. Bouzamondo, M. Sirol, J.-S. Hulot, P. Sanchez, and P. Lechat Differential Effects of Lipid-Lowering Therapies on Stroke Prevention: A Meta-analysis of Randomized Trials Arch Intern Med, March 24, 2003; 163(6): 669 - 676. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Shahar, L. E. Chambless, W. D. Rosamond, L. L. Boland, C. M. Ballantyne, P. G. McGovern, and A. R. Sharrett Plasma Lipid Profile and Incident Ischemic Stroke: The Atherosclerosis Risk in Communities (ARIC) Study Stroke, March 1, 2003; 34(3): 623 - 631. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Sironi, M. Cimino, U. Guerrini, A. M. Calvio, B. Lodetti, M. Asdente, W. Balduini, R. Paoletti, and E. Tremoli Treatment With Statins After Induction of Focal Ischemia in Rats Reduces the Extent of Brain Damage Arterioscler. Thromb. Vasc. Biol., February 1, 2003; 23(2): 322 - 327. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Lalouschek, W. Lang, S. Greisenegger, and M. Mullner Determination of Lipid Profiles and Use of Statins in Patients With Ischemic Stroke or Transient Ischemic Attack Stroke, January 1, 2003; 34(1): 105 - 110. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Di Napoli, A.A. Taccardi, M. Oliver, and R. De Caterina Statins and stroke: evidence for cholesterol-independent effects Eur. Heart J., December 2, 2002; 23(24): 1908 - 1921. [PDF] |
||||
![]() |
T. Marshall and A. Rouse Resource implications and health benefits of primary prevention strategies for cardiovascular disease in people aged 30 to 74: mathematical modelling study BMJ, July 27, 2002; 325(7357): 197 - 197. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Koren-Morag, D. Tanne, E. Graff, U. Goldbourt, and for the Bezafibrate Infarction Prevention Study Gr Low- and High-Density Lipoprotein Cholesterol and Ischemic Cerebrovascular Disease: The Bezafibrate Infarction Prevention Registry Arch Intern Med, May 13, 2002; 162(9): 993 - 999. [Abstract] [Full Text] [PD |