Impressive Data on Prognostic Importance of Lipids May Lead to Earlier Prescription in Those with High Long-Term ASCVD Risk Estimates


The medical literature is replete with convincing evidence for a causal role of non-HDL cholesterol (non-HDL-C) and LDL-cholesterol (LDL-C) in atherosclerotic cardiovascular disease (ASCVD) events. Both the American and European guidelines recommend lipid-lowering therapy for primary and secondary prevention in high-risk individuals. However, as FJ Brunner et al. point out in their recent article published in Lancet: "Application of non-HDL cholesterol for population-based cardiovascular risk stratification: results from the Multinational Cardiovascular Risk Consortium"1 there have been limited data on the association between levels of non-HDL concentration in young adults and very long-term ASCVD outcomes.

Their article had three scientific goals: 1) evaluate long-term risk for ASCVD relative to non-HDL-C levels (on the basis of existing thresholds of lipid concentrations); 2) establish an easily applicable tool to assess the long-term risk for ASCVD events associated with non-HDL-C using a derivation and validation model; and 3) provide a model indicating the potential benefit of an early lipid-lowering strategy in individuals without prevalent cardiovascular disease.


In this retrospective observational study, the authors analyzed individual-level data from the Multinational Cardiovascular Risk Consortium, which had data on 524,444 individuals from 44 cohorts. The authors excluded individuals with prevalent cardiovascular disease (history of myocardial infarction, coronary artery revascularization or ischemic or hemorrhagic stroke). The authors analyzed almost 400,000 individuals (184,055 about half women with a median age of 51 years from 38 primarily European-ancestry cohorts); they were able to derive and validate a model to estimate the long-term probability for an ASCVD event by the age of 75 years.

The authors included nearly 200,000 individuals (48% women) in the derivation cohort and nearly 200,000 individuals were included in the validation cohort. Median follow-up was 13.5 years (IQR 7.0-20.1) and a maximum follow-up of 43.6 years between 1970 to 2013 during which 54,542 ASCVD events occurred.


This article emphasizes the prognostic impact of non-HDL-C on long-term ASCVD. The incidence curve analysis showed progressively higher 30-year ASCVD event rates for increasing non-HDL-C categories (from ~8% for non-HDL-C <2.6mmol/L or < 100 mg/dL to ~34% for non-HDL-C >5.7mmol/L or 220 mg/dL for women and from 13% to 44% for men). The highest hazard ratios (HR) for having a cardiovascular event by the age of 75 years were found in individuals <45 years of age with an elevated non-HDL-C at baseline.

The HR increased as the baseline non-HDL-C increased, but it was most dramatic in subjects <45 years of age at baseline. The authors speculated that this association is due to the following factors: 1) the effects of lifetime accumulation of proatherogenic lipids in the vulnerable period of the fourth and fifth decades of life; 2) younger individuals with elevated non-HDL-C will accumulate more years of proatherogenic lipid levels.

The study has an inherent selection bias since people with prevalent cardiovascular disease are not included in the study; thus, individuals who reach an older age without developing cardiovascular disease are a population possibly enriched with protective characteristics against proatherogenic effects of elevated non-HDL-Cl. Older individuals are also more likely to be started on preventive therapies than younger individuals that will decrease ASCVD risk.

Similar Analysis from the Johns Hopkins Precursors Study

A similar but smaller study published in 1993 by Klag et al.2 in the New England Journal of Medicine supports the authors' claims that an elevated non-HDL-C early in life increases the lifetime risk of ASCVD. Klag et al. performed a prospective study of 1,017 young male Hopkins Medical Students (mean age 22), who were followed for 27 to 42 years to assess the risk of cardiovascular disease and mortality associated with serum cholesterol levels during early adult life. The mean total cholesterol was 192 mg/dL.

Over a median follow-up of just over 30 years, there were 125 cardiovascular events, and 97 were due to coronary heart disease. The baseline serum cholesterol level was strongly associated with both cardiovascular and total mortality. The paper concluded that "it remains for future studies to demonstrate that interventions designed to lower the cholesterol level in young adulthood in the primary prevention of coronary heart disease." Two and half years later the West of Scotland trial was the first statin trial to show a significant decrease in cardiovascular events in the high-risk primary prevention setting. Since then multiple studies have demonstrated similar conclusions, statins decrease ASCVD events in both primary and secondary prevention.

Limitations of Traditional ASCVD Risk Assessment

The findings of this study by Brunner et al.1 highlight some of the limitations of the conventional primary prevention guideline recommendations, which are based on a person's individual 10-year ASCVD risk. This time frame underestimates or fails to take into proper account the cumulative lifetime risk for having a major cardiovascular event due to lipid-related atherogenesis, a time dependent process.

Furthermore, this study implies that younger individuals might be more adversely affected by modest elevation in non-HDL-C compared to older individuals, as they inevitably accumulate more years of proatherogenic lipid levels. The authors demonstrate this by showing women <45 years old with non-HDL of 2.6mmol to <3.7mmol (100 mg/dL to 143 mg/dL) with 0-1 modifiable cardiovascular risk factors have a similar probability of having a major cardiac event compared to women >60 with non-HDL-C >5.7mmol (220 mg/dL)with 0-1 modifiable cardiovascular risk factor (6.9% vs 8.2% respectively).

Importance of Lifetime Risk Assessment

The increased lifetime risk of moderately elevated non-HDL-C supports a potential future shift in the primary prevention treatment paradigm, and suggests that estimating both an individual's short-term (10 years) and lifetime risk of developing a major cardiovascular event is a better strategy for assessing the need to initiate cholesterol-lowering therapy. This will enable more informed shared decision making by the patient and clinician. Current guidelines recommend lifetime risk estimation, but this is geared toward motivating the patient to make sustained lifestyle changes rather than also adding a statin. This class of medication is generally very safe and lovastatin was FDA-approved more 32 years ago.

The Prognostic Importance of a Low Non-HDL-C

Using a derivation and validation model, the authors were able to derive a tool to estimate the probability of having a major cardiovascular event by the age of 75 based on the following factors: baseline non-HDL-C, age, sex, and the presences or absence of ≥2 classic modifiable cardiovascular risk factors (i.e. hypertension, diabetes, obesity and smoking). The derived tool estimation showed a high comparability between the derivation and validation cohorts as reflected by smooth calibration curves analyses and a root mean square error lower than 1% for the estimated probabilities of cardiovascular disease.

Using this tool, and assuming a 50% reduction in non-HDL-C, the authors were able to make predictions of the probability of reducing a major cardiovascular event by the age of 75 years with lipid lowering interventions. The authors created a user-friendly table that shows the theoretical probability reduction, which could potentially be used in patient discussion regarding the long-term benefit of lipid lowering therapies such as diet, exercise and medications.

Limitations of the Analysis by Brunner et al.

The authors did acknowledge the following limitations of their predictions, since only baseline data of blood lipids were available and no information about dynamic changes during the follow-up period or about the initiation of lipid-lowering therapy was available. Additionally, the therapeutic benefit of lipid-lowering intervention investigated is based on a hypothetical model that assumes a stable reduction of non-HDL-C. This assumption posits that treatment effects are sustained over a much longer term than has been studied in clinical trials.

In summary, there is clear and compelling data that maintaining lower LDL-cholesterol and non-HDL-cholesterol levels throughout life are associated with a lower risk of ASCVD. Optimizing patient dietary and exercise habits to lower their lipids, and consideration of early statin therapy, will be important to slow the progression of subclinical atherosclerosis. We have learned a great deal since Dr. Klag's landmark 1993 article, but there is still much more work to be done in the field of primary prevention. We believe that this paper will encourage more patients and providers to start statin therapy earlier based on their 30-year ASCVD risk rather than what is now recommended in the US and European guidelines.


  1. Brunner FJ, Waldeyer C, Ojeda F, et al. Application of non-HDL cholesterol for population-based cardiovascular risk stratification: results from the Multinational Cardiovascular Risk Consortium. Lancet 2019;394:2173-83.
  2. Klag MJ, Ford DE, Mead LA, et al. Serum cholesterol in young men and subsequent cardiovascular disease. N Engl J Med 1993;328:313-18.

Clinical Topics: Diabetes and Cardiometabolic Disease, Dyslipidemia, Prevention, Lipid Metabolism, Nonstatins, Novel Agents, Statins, Hypertension

Keywords: Cholesterol, LDL, Hydroxymethylglutaryl-CoA Reductase Inhibitors, Prospective Studies, Cardiovascular Diseases, Lovastatin, Secondary Prevention, Retrospective Studies, Risk Factors, Coronary Vessels, Calibration, Lipids, Goals, Atherosclerosis, Cholesterol, Risk Assessment, Cohort Studies, Myocardial Infarction, Coronary Disease, Primary Prevention, Hypertension, Life Style, Diabetes Mellitus, Obesity, Stroke, Dyslipidemias

< Back to Listings