Gender Differences in Cardiovascular Mortality by C-Reactive Protein Level in the United States: Evidence From the National Health and Nutrition Examination Survey III

Aug 01, 2013
Font Size
A
A
A
Authors:
Doran B, Zhu W, Muennig P.
Citation:
Am Heart J 2013;166:45-51.
Summary By:
Elizabeth A. Jackson, MD, FACC

Study Questions:

Does the relationship between C-reactive protein (CRP) levels and mortality differ for women and men?

Methods:

Data from the National Health and Nutrition Examination Survey III 1988-1994 were used for the present analysis. All models controlled for race, age, smoking, high-density lipoprotein (HDL), hypertension, diabetes mellitus, waist circumference, and total cholesterol. The primary outcomes of interest were cardiovascular (CV) mortality and all-cause mortality identified through the National Death Index through 2006.

Results:

A total of 13,878 individuals (7,364 women and 6,514 men) with a median follow-up of 18.2 years were included in this cohort. Over a total follow-up of 18.2 years, there were 940 deaths from CV causes and 1,721 deaths from all other causes. Women were more likely to have elevated levels of CRP than men. In women and men, older age, low HDL, hypertension, diabetes, an enlarged waist circumference, and elevated total cholesterol were all associated with elevated levels of CRP. Smoking was not associated with elevated CRP in women, but was in men. Men with a CRP >3.0 mg/L had elevated CV mortality hazards (hazard ratio [HR], 1.79; 95% confidence interval [CI], 1.23-2.60) and all-cause mortality hazards (HR, 1.57; 95% CI, 1.29-1.90), compared to men with a CRP ≤3.0 mg/L. In contrast, among women, elevated CRP was not significantly associated with either increased CV (HR, 1.20; 95% CI, 0.90-1.59) or all-cause mortality hazards (HR, 1.09; 95% CI, 0.93-1.29).

Conclusions:

The investigators concluded that CRP in relation to CV risk may differ by gender.

Perspective:

This analysis provides data on the association of CRP and CV and all-cause mortality. In particular, CRP does not appear to add additional information on mortality risk beyond traditional risk factors.

Keywords: Follow-Up Studies, Continental Population Groups, Risk Factors, Smoking, Cholesterol, Incidence, Waist Circumference, Dyslipidemias, C-Reactive Protein, Body Mass Index, Biomarkers, Confidence Intervals, Hypertension, Diabetes Mellitus


< Back to Listings