Race and Gender Differences in C-Reactive ProteinLevels

Khera A, McGuire DK, Murphy SA, et al. J Am Coll Cardiol2005;46:464 –9.

Study Question: Are there race and gender differences in thedistribution of C-reactive protein (CRP) levels?Methods: CRP was measured in 2749 white and black subjectsaged 30 to 65 years participating in the Dallas Heart Study, amultiethnic, population-based probability sample. Levels ofCRP were compared between different race and gendergroups. The high-sensitivity CRP assay minimal detectablerange was 0.1 mg/L, and the upper limit was 20 mg/L.Results: Mean age in all subjects was about 45 years. Com-pared to white subjects, black men and women had morediabetes and hypertension. Black women had a higher BMIand less often used estrogens, and black men had the highesttobacco use. Subjects were not equally distributed between theusual CRP risk tertile (in mg/L �1�20%; 1–3�31%;�3�49%). Hypertension, diabetes, lipid risk factors, BMI andestrogens were each significantly associated with increasingCRP. Black subjects had higher CRP levels than did whites(median, 3.0 vs. 2.3 mg/L; p�0.001) and women had higherCRP levels than did men (median, 3.3 vs. 1.8 mg/L; p�0.001).The sample-weight adjusted proportion of subjects with CRPlevels �3 mg/L was 31%, 40%, 51% and 58% in white men,black men, white women, and black women, respectively(p�0.05 for each group vs. white men). After adjustment fortraditional cardiovascular (CV) risk factors, estrogen and statinuse, and BMI, a CRP level �3 mg/L remained more commonin white women (odds ratio [OR] 1.6; 95% CI 1.1–2.5) andblack women (OR 1.7; 95% CI 1.2–2.6) but not in black men(OR 1.3; 95% CI 0.8–1.9) when compared with white men.Conclusions: Significant race and gender differences exist inthe population distribution of CRP. Further research isneeded to determine whether race and gender differences inCRP levels contribute to differences in CV outcomes, andwhether thresholds for CV risk assessment should be ad-justed for different race and gender groups.Perspective: Considering the potential value of using CRPfor risk stratification, it is very important to ensure thecut-points for risk stratification are properly adjusted forgender, race, ethnicity and other potentially confoundingvariables. Melvyn Rubenfire

C-Reactive Protein and the 10-Year Incidence ofCoronary Heart Disease in Older Men and WomenCushman M, Arnold AM, Psaty BM, et al. Circulation2005;112:25–31.

Study Question: Is a high level of C-reactive protein (CRP)

associated with long-term coronary heart disease (CHD)risk in the elderly?Methods: Baseline CRP was measured in 3971 men andwomen �65 years of age without prior vascular diseasesparticipating in the Cardiovascular Health Study en-rolled between 1989 and 1993. CRP was stratified as low(�1 mg/L), intermediate (1–3 mg/L), and high (�3mg/L). The presence of subclinical cardiovascular dis-ease (CVD) was assessed by carotid IMT �80th percen-tile or �25% stenosis, ECG abnormalities, Rose ques-tionnaire, angina or claudication, and ankle/brachialindex (ABI) �0.9. Primary outcome was the relationshipbetween CRP group and 10-year risk of CHD risk ad-justed for classic risk factors and subclinical CVD. Re-searchers defined CHD as myocardial infarction (MI) orcoronary death obtained by hospitalization dischargecodes and subjected to verification.Results: Mean age was 72.5 years, 42% were male, meanLDL-C was 134 mg/dL, 33% were ever smokers, 20%used aspirin regularly, 64% had subclinical vasculardisease, 29% a CRP �1 mg/L, 45% 1–3 mg/L, and 26%had elevated concentrations �3 mg/L. With 10 years offollow-up, 547 participants developed CHD. With ele-vated CRP, the 10-year cumulative CHD incidences were33% in men and 17% in women. The age-, ethnicity-,and gender-adjusted relative risk (RR) of CHD for CRP�3 mg/L compared with �1 mg/L was 1.82 (95% CI1.46 –2.28). Adjusting for conventional risk factors re-duced the RR to 1.45 (95% CI 1.14 –1.86). The popula-tion-attributable risk of CHD for elevated CRP was 11%.Risk relationships did not differ in subgroups defined bybaseline risk factors. When assessed whether CRP im-proved prediction by the Framingham Risk Score (FRS)among men with a 10-year FRS of 10 –20%, the observedCHD incidence was 32% for elevated CRP. Amongwomen, CRP discriminated best among those with a10-year predicted risk �20%; the incidences were 31%and 10% for elevated and normal CRP levels, respec-tively.Conclusions: In older men and women, elevated CRP wasassociated with increased 10-year risk of CHD, regardless ofthe presence or absence of cardiac risk factors. A single CRPmeasurement provided information beyond conventionalrisk assessment, especially in intermediate-Framingham-risk men and high-Framingham-risk women.Perspective: There is good evidence to add CRP to standardrisk stratification in middle-aged and elderly men andwomen with a FRS of �1% for deciding whether to useaspirin and statins. Somewhat lost in the study is the findingthat subclinical CVD was associated with �2% annual riskof death or MI in men with a low CRP. The risk for eventswas 70% greater when the carotid IMT was �80th percen-tile with a CRP �1 mg/L and similar for major ECG abnor-malities, abnormal ABI, and carotid stenosis. MelvynRubenfire

ACC CURRENT JOURNAL REVIEW November 2005

16

Preventive CardiologyAbstracts