Current issues of ACP Journal Club are published in Annals of Internal Medicine


Fibrinogen as a cardiovascular risk factor: a meta-analysis

ACP J Club. 1993 Sept-Oct;119:55. doi:10.7326/ACPJC-1993-119-2-055

Source Citation

Ernst E, Resch KL. Fibrinogen as a cardiovascular risk factor: a meta-analysis and review of the literature. Ann Intern Med. Jun 1993



To assess, using meta-analysis, whether high fibrinogen levels are associated with myocardial infarction, stroke, and ischemic heart disease.

Data sources

MEDLINE searches from 1980 to 1992 and a review of relevant bibliographies to identify epidemiologic studies as well as basic clinical and research studies. Search terms included fibrinogen and cardiovascular diseases.

Study selection

Epidemiologic studies were selected if they were prospective and included a description of the patients. 7 studies were selected, and 1 of these was excluded because of a small nonrepresentative group of patients and nonuniform follow-up. Not all studies excluded patients with known cardiovascular disease.

Data extraction

Sample size and composition; length of follow-up; end points (mortality, ischemic heart disease, myocardial infarction, and stroke); and method of measuring fibrinogen levels. Levels of fibrinogen were divided into tertiles. Odds ratios (ORs) for cardiovascular events were calculated for the upper tertile for fibrinogen levels compared with the lowest tertile for each study, and a summary OR was calculated.

Main results

6 studies with 14 630 men and 761 women (only the Framingham study included women) representing 92 147 person-years were included in the meta-analysis. Follow-up was from 2 to 13.5 years. 4 studies assessed ischemic heart disease; 2 assessed myocardial infarction; 2 assessed stroke; and 1 assessed ischemic heart deaths. ORs for cardiovascular events, comparing the highest tertile of fibrinogen levels with the lowest tertile, ranged from 1.8 (95% CI 1.2 to 2.5) to 4.1 (CI 2.3 to 6.9) with a summary OR of 2.3 (CI 1.9 to 2.8). Each study used multivariate analyses, adjusting for age, smoking, cholesterol level, and blood pressure, to show that fibrinogen was independently associated with cardiovascular disease. 5 of the studies, however, did not consider low-density lipoprotein in their analysis or measured it by inadequate methods. The mechanism behind the association is unclear but may be mediated through one or more of plasma viscosity, platelet aggregation, or vessel-wall inflammation.


High fibrinogen levels were associated with an increased risk for cardiovascular events (myocardial infarction, stroke, and ischemic heart disease), but causality was not proved.

Source of funding: {Deutsche Herzhilfe}.(Information supplied by author.)

For article reprint: Dr. E. Ernst, University of Vienna, AKH, Waehringer Guertel 18-20, 1090 Vienna, Austria. FAX 43-1-40400-5281.


This excellent, comprehensive review and analysis of 6 major prospective epidemiologic trials (including the Framingham Study) reminds us that fibrinogen is an important variable associated with coronary heart disease and stroke. All 6 studies showed an increased univariate risk for either coronary heart disease or stroke for those in the highest tertile of fibrinogen level. The overall OR of 2.3 is comparable with other commonly accepted risk factors, such as low-density lipoprotein cholesterol, smoking, and male gender.

High levels of fibrinogen have also been associated with other risk factors: male gender, increased weight, postmenopausal state, physical inactivity, diabetes, hypertension, and, most notably, smoking. Multivariate analysis controlling for these associations showed that the relation of fibrinogen to cardiovascular events remained statistically significant, although weakened in predictive strength.

Fibrinogen could certainly be associated with cardiovascular disease. It is a major determinant of blood viscosity because it interacts with platelet membrane receptors necessary for platelet aggregation. Fibrinogen is incorporated into vascular lesions and can stimulate smooth muscle cell proliferation and migration.

Why, then, is fibrinogen a weak element in the pantheon of cardiovascular risk factors? Probably because we do not know what to do about it. Fibrinogen is a risk marker begging for some preventive or clinical role. Part of the confusion rests with the uncertainty of the association with atherosclerotic disease; is it a causal one, a mediating one for other known risk factors, or simply a marker for a yet-to-be-elucidated risk factor? Although no current way exists to lower fibrinogen levels independent of other risk factor modification, the consistent strength of its association suggests that it should be measured in future epidemiologic studies and clinical trials of antithrombotic agents. Does it retain or lose its risk association in the presence of aspirin? How do other potentially prothrombotic risk factors, such as lipoprotein(a), interact with it? Such studies may slowly help to elucidate the clinical importance and mechanisms of this interesting risk association.

Donald A. Smith, MD
The Mount Sinai Medical Center New York, New York, USA