Reduction in Stroke with Statin Therapy

Sources:

Hebert PR, et al. JAMA 1997;278:313-321; Bucher HC, et al. Ann Intern Med 1998;128:89-95.

the question as to whether hypercholesterolemia is a risk factor for stroke has been controversial. Previous data have been inconclusive, and a large meta-analysis published recently, consisting of an overview of 45 trials and 450,000 subjects, concluded that there is no independent association between baseline cholesterol and stroke risk. Two more recent overviews or meta-analyses of randomized, clinical trials using cholesterol reduction therapy, especially the statin drugs, have been published and clearly indicate a significant benefit from statins on fatal and non-fatal stroke. The earlier study was performed by the Division of Preventive Medicine at Harvard and analyzed only on statin trials published between 1985 and October 1996. A total of 16 trials were included in this analysis of 29,000 subjects who were followed for an average of 3.3 years, although there was substantial variation in the follow-up period. Some of the studies were regression trials, which have rather small cohorts and short follow-up. The majority of the event data was derived from three large, well-known statin trials: 4-S, Western Scotland, and CARE, which represent approximately 90% of all strokes and deaths in this analysis. Mean baseline cholesterol was highly variable; in general, it was above normal except for the CARE trial. The average reduction in total cholesterol ranged from 17-32% and LDL cholesterol from 26-45%. Both fatal and non-fatal strokes were counted, as well as all deaths.

The results indicate a significant 29% reduction in total stroke rate, with a greater proportional reduction in secondary prevention trials than in the Western Scotland primary prevention trial. When all secondary prevention studies were included, there was a 32% reduction in risk vs. an approximately 20% non significant reduction in the primary prevention cohort. All deaths, including cardiovascular and coronary, were substantially decreased by 22%, 28%, and 31%, respectively. Again, there were greater proportional reductions in the secondary prevention cohorts. There was a robust one-third reduction in any coronary artery disease events in the 16 trials. Of importance, cancer and non cardiovascular death were not increased with statin therapy. However, Hebert and colleagues point out that the confidence intervals regarding cancer are wide, and that a small increase in cancer risk cannot be excluded by the data. Hebert et al conclude that statin therapy reduces ischemic stroke through suppression of atherogenic progression, and that the benefits of cholesterol lowering appear to be proportional to the degree of reduction in baseline lipid levels. Total mortality, but not stroke mortality, was decreased significantly.

The second study is a meta-analysis of all randomized, clinical trials reporting stroke data through October 1996 that employed cholesterol-lowering interventions, including diet and non-statin drugs. A different methodology was used, with the primary end point being non fatal and fatal stroke in the randomized data base. Four categories of drug therapy were individually analyzed: statins (8 trials), fibrates (5 trials), resins, and diet. Miscellaneous therapies, including ileal bypass and niacin, were also assessed. The total database in the intervention cohort consisted of approximately 50,000 subjects, and the control group consisted of 57,000. Twenty-eight trials in all were included in the analysis.

The overall stroke-risk ratio for all cholesterol-lowering interventions was not significantly reduced (5% lower in treated patients). However, the statins reduced total stroke by 24%, and this was different from the other lipid-lowering drugs, which did not significantly reduce stroke. Overall, death rates and coronary artery disease mortality were reduced in the entire 28 trials by 13%, with the statins having a risk ratio of 0.69 or a 31% reduction. Total mortality was reduced by 20% by the statins and 6% when all studies were included (not significant).

The authors’ comment that there was a consistent effect of the statins across all studies and that the summary estimates for the statins is clinically important and statistically significant, whereas this is not true for other lipid-lowering trials. Bucher and colleagues caution that because these data were comparisons between study trials, and that the statin trials include older individuals with higher event rates, and that there may be some artificial difference between the robust statin potency compared to other drug therapies. Bucher et al conclude that their data indicate that elevated cholesterol is a risk factor for stroke and may be causal. They point out that it is possible that lipid lowering may have effects other than reduction in cholesterol, resulting in benefits in stroke. The issue of hemorrhagic vs. ischemic stroke cannot be resolved by the limited available data on hemorrhagic stroke due to classification problems. Finally, Bucher et al suggest that risk reduction with stroke is comparable to that shown by antiplatelet agents. In an unpublished analysis, they report that statins have a favorable effect on both coronary artery disease and overall mortality, reflecting all 64 randomized trials. Bucher et al suggest that statins, but not other lipid lowering agents, are effective in stroke prevention.

Comment by Jonathan Abrams, MD

These two databases are important, as they offer a new preventive strategy for stroke. Whereas it has been previously believed that only hypertension and atrial fibrillation were major conditions that predispose to stroke, and that elevated cholesterol did not have a causal role, we now know that hypercholesterolemia is another reason to consider statin therapy in high-risk primary prevention subjects or in patients who have already had cardiovascular disease—particularly a coronary event. As with fatal and non fatal myocardial infarction and overall coronary mortality, the risk reduction with statins is powerful and consistent, approaching 25-30%. Thus, it is likely that widespread and appropriate use of statins will not only decrease coronary artery disease, clinical events, and death, but the benefits will translate into cerebrovascular disease.

The mechanisms whereby the statins induce such positive benefits are presumed to be related to a major lowering of LDL cholesterol, perhaps including a minor elevation of HDL cholesterol and improvement in the total cholesterol HDL ratio. However, a number of other parameters may be influenced by these potent drugs, including hemostatic factors, the thrombolytic milieu, decreased LDL oxidation, and alterations in various aspects in vascular biology. Furthermore, assuming that more strokes than previously recognized are due to aortic atheromatous disease, it is also conceivable that statin therapy will reduce the atheroma burden in the aorta over time, and this would be another mechanism for reducing stroke rates (in addition to less left ventricular damage from fewer myocardial infarctions).

Now, it can be stated categorically that statin therapy in appropriate individuals will decrease the risk of death, heart attack, angina pectoris, need for revascularization, and stroke. This is good news indeed, and the statin story is continuing to unfold. n

Dr. Abrams is Professor of Medicine, Division of Cardiology, University of New Mexico, Albuquerque.