Physicians' Academy for Cardiovascular Education

JUPITER criteria identify more individuals in EPIC-Norfolk population at high CV risk than other guidelines

Literature - Sondermeijer BM, Boekholdt SM, Rana JS, Kastelein JJ et al. - Eur Heart J. 2013 Feb 28.

Clinical implications of JUPITER in a contemporary European population: the EPIC-Norfolk prospective population study.

Sondermeijer BM, Boekholdt SM, Rana JS, Kastelein JJ et al.
Eur Heart J. 2013 Feb 28. [Epub ahead of print]


A substantial number of patients who have reached the recommended range of LDL-C levels with statin treatment still develop cardiovascular (CV) events [1,2]. C-reactive protein (CRP) appears to have predictive value for residual cardiovascular risk. The Justification for the Use of statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER)-trial tested the efficacy of rosuvastatin as primary prevention among individuals with low LDL-C (<3.4 mmol/L) but with slightly elevated CRP (≥2 mg/L)[3]. 20 mg of rosuvastatin gave a significant reduction in CV events. This result has led to considerable debate on the value of use of rosuvastatin as primary prevention, for instance about the characteristics and risk profile of the JUPITER study population [4-8].
Based on questions on how many individuals in a European population are eligible for primary prevention according to the JUPITER criteria, this study aims to assess the proportion of individuals fulfilling the JUPITER criteria in the EPIC-Norfolk population study. Additional objectives included determining CV risk profile and risk of future coronary heart disease (CHD) among JUPITER-eligible individuals and comparison with other risk criteria.
The European Prospective investigation into Cancer and Nutrition (EPIC)-Norfolk study is a prospective population study of 25 639 persons aged 45-79 years old. Although set up to investigate determinants of cancer, additional data have been collected to enable studies into determinants of other diseases. 8397 patients meeting the JUPITER exclusion criteria were omitted from the study population.

Main results

  • 10.1% of the total study population were JUPITER eligible (LDL-C <3.4 mmol/L and CRP >2 mg/L).
  • In the JUPITER-eligible group, 33.8% had an intermediate Framingham risk score (FRS) and 25.9% had a high FRS. Men had higher FRS than women, with 78% of women having a low FRS.
  • In the JUPITER-eligible group, the rate of CHD events was 15.6% over a mean follow-up of 11.4 years (1.37%/year). HR for future CHD was 1.70 (95%CI: 1.31-2.21) as compared to group 1 (LDL-C <3.4 mmol/L and CRP <2 mg/L) and the HR for group 3 vs group 1 (LDL-C ≥3.4 mmol/L and CRP <2 mg/L) was 1.29 (95% CI:1.01–1.64).
  • In the group eligible for primary prevention, around 45% had an indication for statin therapy according to SCORE or US ATP III guidelines, whereas ~55% did not. In the latter groups, ~18% fulfilled JUPITER criteria, and would thus qualify for statin therapy independent of SCORE and US ATP III guidelines.


Around 10% of participants without prevalent CV disease in the EPIC-Norfolk population met JUPITER criteria. These people showed a considerable risk profile (almost 60% having intermediate or high FRS), although the JUPITER population has previously been described as an apparently healthy population with slightly increased C-reactive protein levels only.
Interestingly, individuals with high CRP and low LDL-C had higher absolute event rates than individuals with low CRP and high LDL-C. These data thus underscore previous evidence indicating a role for CRP in risk assessment.
Application of JUPITER criteria yielded an additional ~18% of people qualifying for statin therapy for primary prevention. Considering the high risk profile of JUPITER eligible individuals, the JUPITER data may be of interest for future European guidelines on CV disease prevention in clinical practice.

Editorial comment [9]

A challenge of risk evaluation in primary prevention is that it must integrate cumulative risk exposure and a ‘point’ estimate at any given moment in a lifetime. The use of circulating biomarkers such as C-reactive protein might enhance predictive power of such risk assessment.
Although the authors highlighted some limitations in their study design, it is unlikely that these would have affected the interpretations of their findings. However, the failure to include stroke in the composite endpoint analysis, might be a more important limitation, since evidence exists that CRP level might be superior to LDL-C concentration as a predictor of stroke.
The JUPITER trial has been surrounded by much debate and disagreement. A central question concerns the factors which drive chronic, moderately elevated systemic inflammation, and in turn CRP level. Socio-economic, environmental, lifestyle and genetic factors may contribute to an inflammatory state, of which CRP is an indicator.
A remaining issue to resolve is whether inclusion of CRP measurement may enhance identification of a high-risk group within the moderate risk individuals in the general European population. The simulation of Sondermeyer et al should be confirmed in an independent prospective intervention trial. These results should help determine whether CRP measurement can help efficiency of primary prevention with statins, by shifting a proportion of moderate risk people to eligibility for statin therapy.