The Association Between Low-Density Lipoprotein Cholesterol and Incident Atherosclerotic Cardiovascular Disease in Older Adults

Results From the National Institutes of Health Pooled Cohorts

Michael G. Nanna, MD; Ann Marie Navar, MD, PhD; Daniel Wojdyla, MSc; Eric D. Peterson, MD, MPH


J Am Geriatr Soc. 2019;67(12):2560-2567. 

In This Article

Abstract and Introduction


Background/Objectives: Elevated low-density lipoprotein cholesterol (LDL-C) in early adulthood is associated with increased risk of atherosclerotic cardiovascular disease (ASCVD). The strength of the association between LDL-C and ASCVD among older adults, however, is less understood.

Design: We examined individual-level cohort data from the National Institutes of Health Pooled Cohorts (Framingham Study, Framingham Offspring Study, Multi-Ethnic Study of Atherosclerosis, and Cardiovascular Health Study), which prospectively measured CVD risk factors and incident disease.

Setting: Prospective cohort study.

Participants: Adults, aged 75 years or older, free of ASCVD.

Measurements: We evaluated the associations between LDL-C and incident ASCVD (stroke, myocardial infarction, and cardiovascular death) in unadjusted analysis and in multivariable-adjusted Cox proportional hazards models. We assessed 5-year Kaplan-Meier ASCVD event rates in patients with and without hyperlipidemia (LDL-C ≥130 mg/dL or on lipid-lowering medications), stratified by the number of other risk factors, including smoking, diabetes, and hypertension.

Results: We included 2667 adults, aged 75 years or older (59% female), free of ASCVD; median age was 78 years, with median LDL-C of 117 mg/dL. In both unadjusted and adjusted analyses, there was no association between LDL-C and ASCVD (adjusted hazard ratio = 1.022; 95% confidence interval = 0.998–1.046; P = .07). Among adults without other risk factors (free of smoking, diabetes, and hypertension), event rates were similar between those with and without hyperlipidemia (Kaplan-Meier rates = 5.8% and 7.0%, respectively). Among adults with one or two or more other risk factors, the presence of hyperlipidemia was also not associated with 5-year CVD event rates (Kaplan-Meier rates = 12.8% vs 15.0% [P = .44] for one other risk factor and 21.9% vs 24.0% [P = .59] for two or more other risk factors).

Conclusion: Among a well-characterized cohort, LDL-C was not associated with CVD risk among adults aged 75 years or older, even in the presence of other risk factors.


Hyperlipidemia is a well-established modifiable risk factor for cardiovascular events among adults.[1–3] In population-wide studies, for every 1-mmol/L increase in non–high-density lipoprotein cholesterol, ischemic heart disease mortality increases by approximately one-third;[4] however, much of this evidence comes from research on younger populations. The association between other lipoprotein laboratory parameters, such as total cholesterol and mortality, has been previously evaluated in older adults.[4–6] A few epidemiological studies have assessed the association between low-density lipoprotein cholesterol (LDL-C) and mortality, but those including older adults have found conflicting results.[4,6–11] Even fewer studies have examined the association of LDL-C and atherosclerotic cardiovascular disease (ASCVD) risk, including cardiovascular events in addition to cardiovascular mortality, in those with advanced age; those that have were limited by small sample sizes taken from a single city.[6,8] As a result, whether LDL-C levels can be used to risk stratify older adults who survive to older ages without actually developing ASCVD is uncertain.

In addition to few epidemiological studies, there are also few randomized primary prevention trials of statin therapy for ASCVD that have included older adults (aged ≥75 years).[12–15] The Prospective Study of Pravastatin in the Elderly at Risk trial included 5804 patients, aged 70 to 82 years, with known vascular disease or with risk factors for vascular disease, and found that pravastatin therapy reduced the risk of cardiovascular events compared with placebo.[14] However, a subgroup analysis of the primary prevention cohort (N = 3239) found no significant improvement in the primary end point of coronary heart disease death, nonfatal myocardial infarction (MI), and fatal or nonfatal stroke.[14] A subgroup analysis of the Justification for Use of Statins in Prevention: An Intervention Trial Evaluating Rosuvastatin trial included 5595 patients, aged 70 years or older, and did find a 39% reduction in the hazard of the primary end point of MI, unstable angina, stroke, arterial revascularization, or cardiovascular death with rosuvastatin vs placebo.[15] More recently, the results of a randomized trial investigating the efficacy of ezetimibe in high-risk primary prevention Japanese older adults (aged ≥75 years) were presented as a late-breaking clinical trial at the American Heart Association 2018 Scientific Sessions, demonstrating a significant reduction in cardiovascular events at 5 years with ezetimibe vs control;[16] the results of this study have yet to be published. As a result of the relatively limited evidence base, cholesterol guidelines are less definitive in treatment algorithms for this particular population, and suggest a shared decision-making approach when considering initiation of statin treatment in individuals aged 75 years or older.[12] This lack of evidence in older individuals is important since clinicians will increasingly face treatment decisions in this quickly expanding demographic group.[17,18] A particularly challenging subgroup is those older adult patients (aged ≥75 years) who survive to older age without any traditional cardiovascular risk factors beyond age, sex, or race, but whose predicted risk using existing risk stratification tools remains elevated.

Using data from the National Heart, Lung, and Blood Institute (NHLBI) Pooled Cohorts, containing prospectively measured CVD risk factor profiles and closely adjudicated CVD events, we evaluated the risk of new-onset ASCVD among a large cohort of individuals who were free of CVD by the age of 75 years. Using these data, we set out to: (1) evaluate the overall unadjusted and risk-adjusted association between LDL-C and CVD events among older adults (aged ≥75 years) without known ASCVD; and (2) assess the impact of hyperlipidemia on the risk of new ASCVD among older adults with varying risk factor profiles.