The article by Bucher and coworkers [1] confirms previous contributions that have identified an effect of statins to reduce stroke [2, 3]. However, it also raises several questions.

First, the article reviewed only 8 of 15 published trials of statins that provide information on stroke [2, 3]. The criteria that led to the exclusion of these other available clinical trials would be of interest. This is important because 7 of the 8 statin trials reviewed by Bucher and coworkers are studies of the secondary prevention of cardiovascular disease. Our own work [2] and that of others [3] suggests that secondary prevention statin studies show a stronger effect of cholesterol reduction on stroke prevention than do stain studies that either include a mixture of persons with and without coronary disease or focus attention on primary prevention. In fact, for statins, there is a preponderance of trials of secondary (as opposed to primary) prevention. This may be one of the important differences between the new meta-analyses of statin trials and two previous negative meta-analyses of clinical trials that did not include 3-hydroxy-3-methylglutaryl coenzyme A (HMGcoA) reductase inhibitors. In the latter, a large proportion of the pharmacologic trials were trials of the primary prevention of coronary artery disease [4, 5].

In addition, Bucher and colleagues conclude that "... in hyperlipidemic patients who have not previously had stroke, HMGcoA reductase inhibitors reduce the incidence of stroke." It is not clear that patients who previously had stroke were, in fact, systematically excluded from the studies reviewed.

Finally, we believe that it is now important to begin to better understand the mechanisms whereby cholesterol lowering reduces stroke. My colleagues and I have previously suggested three possibilities: plaque regression, improvement in endothelial stability, and reduction of coronary heart disease [2]. Does this most recent meta-analysis favor one of these mechanisms?