Cholesterol, statins, and mortality.

摘要

In the most recent Prospective Studies Collaboration (PSC) meta-analysis,1 49% of stroke deaths were classified as either ischaernic or haemorrhagic. This seems high since most prospective studies are unable to distinguish the two types of stroke reliably.Misclassification of strokes through use of clinical features alone could help explain two surprising results. First, that the positive association between total cholesterol and ischaemic stroke in PSC was much weaker than in MRFIT (see webfigure 8)1 and other cohort studies2 in which CT scans or autopsy were used to classify stroke, and weaker than the reduction in ischaemic stroke in randomised trials of cholesterol reduction.2 Second, that the inverse relation between total cholesterol and haemorrhagic stroke was weaker in PSC than in MRFIT and other cohort studies.