· The New England journal of medicine 2012 · meta-analysis of prospective cohort studies · n=246,669 participants across 52 studies

C-reactive protein, fibrinogen, and cardiovascular disease prediction.

Cited 1170 times in the scientific literature.

Level 1 - systematic review of randomized trials

Individual-participant data meta-analysis of prospective cohort studies for risk prediction

PubMed 23034020 · doi:10.1056/NEJMoa1107477 · record verified 2026-08-29

What was done

Researchers analyzed individual-level data from 52 prospective studies comprising 246,669 participants without a history of cardiovascular disease. They assessed whether adding C-reactive protein (CRP) or fibrinogen levels to conventional risk factors (age, sex, smoking status, blood pressure, diabetes, total cholesterol, and high-density lipoprotein cholesterol) improved cardiovascular disease prediction using the C-index and net reclassification improvement across 10-year risk categories (<10%, 10% to <20%, and ≥20%). They also modeled the clinical impact of initiating statin therapy based on Adult Treatment Panel III guidelines following biomarker screening.

What was found

Adding high-density lipoprotein cholesterol to conventional risk factors increased the C-index by 0.0050. Further addition of CRP or fibrinogen increased the C-index by 0.0039 and 0.0027, respectively (P<0.001 for both). Net reclassification improvement was 1.52% for CRP and 0.83% for fibrinogen across 10-year risk tiers (P<0.02 for both). In a modeled population of 100,000 adults aged 40 or older with 15,025 at intermediate risk, screening the 13,199 eligible intermediate-risk individuals with CRP or fibrinogen was estimated to prevent roughly 30 additional cardiovascular events over 10 years, or one event for every 400 to 500 people screened.

Why it matters

Measuring CRP or fibrinogen provides statistically significant but clinically small improvements in cardiovascular risk discrimination and reclassification. In primary prevention, the clinical yield of targeted screening in intermediate-risk populations is modest.

Limits

The clinical benefit estimates were derived from decision modeling and guideline assumptions rather than direct outcomes from a randomized screening trial. The abstract reports data strictly for primary prevention in individuals without known cardiovascular disease, and biomarker cost-effectiveness or assay variability was not evaluated.

Cited by