Lipoprotein(a): involved in events, but not burden of atherosclerotic disease?

See related article, pages 1407–1412 In this issue of Stroke, the Atherosclerosis Risk in Communities (ARIC) study provides evidence that lipoprotein(a) [Lp(a)] is a significant predictor of stroke.1 This is an important finding, because ARIC is both large and prospective. Over 14 000 patients were enrolled in 1987 to 1989, and during 13.5 years of follow-up, they experienced 496 ischemic strokes. Lp(a) remained a significant predictor of stroke after adjustment for age, race, systolic blood pressure, antihypertensive medication, smoking status, use of postmenopausal hormone therapy, diabetes, LDL cholesterol, HDL cholesterol, fibrinogen, and von Willebrand factor. It thus seems likely that Lp(a) is potentially an important treatable risk factor. But what is Lp(a), and how does it relate to atherosclerosis and cardiovascular events? A meta-analysis of prospective studies of coronary events2 seems to establish that Lp(a) is also a predictor of myocardial infarction, but numerous studies indicate that it does not seem to predict the burden of atherosclerosis as assessed by intima-media thickness,3,4 except perhaps in patients with renal failure.5 Rossl et al6 found that Lp(a) was not associated with IMT or plaque but was associated with stenosis of carotid arteries. We have previously shown that carotid plaque area more strongly predicts vascular outcomes than does stenosis assessed by carotid Doppler velocities7,8; an analysis of that database prompted by this commentary (previously unpublished) showed that in multiple regression adjusted for age, sex, systolic blood pressure and pack-years of smoking, Lp(a) was not associated with baseline plaque area (P 0.59), but was associated with baseline carotid stenosis (P 0.049). How can this be? It happens because phenotypes of atherosclerosis assessed by ultrasound, angiography and coronary calcification by computed tomography are biologically and genetically distinct.9,10 Because of compensatory enlargement,11,12 stenosis is probably not the consequence of plaque progression but the consequence of plaque rupture. Plaque area, which represents the burden of atherosclerosis at a point in time, reflects the effects over a lifetime of the traditional risk factors of atherosclerosis, but stenosis probably reflects the effects of factors affecting plaque rupture, such as inflammatory cytokines, matrix metalloproteinases and possibly infection, as well as the effects of factors affecting thrombosis and fibrinolysis.9 It seems likely that Lp(a) is involved in thrombosis and fibrinolysis. Two recent reviews help in the understanding of its mechanism of action and its role in atherosclerosis. Koschinsky13 emphasized the mechanism of action of Lp(a):