Association between lipid levels and cardiovascular risk in older adults is controversial. Some studies have found a direct association between elevated cholesterol and adverse events, but in general, the association is not nearly as firm as it is for middle-aged adults. Recommendations for cholesterol screening in older adults largely rely on what is known about the risk association in younger age groups. In the Cardiovascular Health Study, Psaty and associates attempted to determine the association between lipid levels and cardiovascular risk in patients 65 years and older.

Participants were recruited for this multicenter, prospective cohort study in two phases. The first group consisted of 5,201 participants, and three years later, 687 black patients were recruited. All participants underwent a complete baseline history and physical examination. Fasting lipid levels, including high-density lipoprotein (HDL) cholesterol, low-density lipoprotein cholesterol, and triglycerides, collected at baseline, as were other physical and laboratory findings. Participants also underwent carotid sonography. Patients were excluded if they had prebaseline myocardial infarction (MI), stroke, or congestive heart failure.

As follow-up, participants alternated between telephone interviews and clinic visits at six-month intervals. At each contact, assessments were made to identify cardiovascular events, hospitalization, and mortality. Participants were followed for an average of 7.5 years.

Lipid categories were defined according to the Adult Treatment Panel III guidelines. Lipid-lowering therapy was uncommon at baseline and increased only slightly during the study. The most pronounced change in risk for each standard deviation of change in lipid measure (also called hazard ratio) was for the association between HDL levels and MI, with a hazard ratio of 0.85. This association applied more strongly to persons younger than 75 years. Other associations between lipids and cardiovascular events were marginal. One analysis did show a relationship between ischemic stroke and HDL levels in men. Another showed a slight association between ischemic stroke in persons older than 75 years. Two-lipid models corroborated the single model findings associating HDL levels with MI risk, but otherwise, an analysis based on multiple lipids revealed no striking associational pattern.

The authors conclude that in older adults there is an inverse association between HDL levels and MI risk. There was a similar association between HDL levels and ischemic stroke, but only in men. Otherwise, lipids did not predict cardiovascular events particularly well and were not associated with total mortality. The weak association between lipids and cardiovascular events found in this epidemiologic study could be because of early mortality of persons with high lipids who would have had the highest risk. Because statins have been shown to reduce vascular risk in older adults, it could be that their protective effect is caused by some mechanism other than lipid lowering.

editor’s note: This study indicates that more evidence about the role of lipid lowering for primary prevention in older adults is needed. Applying Adult Treatment Panel III guidelines to older adults is based on a prediction model rather than on direct evidence. This study adds to the growing evidence that cholesterol levels in older adults do not necessarily have the same prognostic value that they have in younger adults. One earlier study evaluated total cholesterol levels in older adults after acute MI. The authors found that cholesterol levels did not correlate with all-cause mortality after a follow-up of six years.¹ Another study showed that low cholesterol levels in a mixed cohort of older adults was associated with increased mortality.² Although these studies do not distinguish between cause and effect, they do suggest that a nuanced approach to cholesterol levels—with attention to fractionated lipid levels—will be required before making treatment recommendations. This is more important because of the possibility that statins have protective benefits other than lowering lipid levels.—c.w.