| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Lipids, Lipoproteins, and Cardiovascular Risk Factors |
1 Division of Adult and Community Health, National Center for Chronic Disease Prevention and Health Promotion, and 2 Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA 30341.
aAddress correspondence to this author at: Centers for Disease Control and Prevention, 4770 Buford Hwy., MS K66, Atlanta, GA 30341. Fax 770-488-8150; e-mail eford{at}cdc.gov.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We examined the distribution and correlates of CRP, using data from 2205 women 
20 years of age from the National Health and Nutrition Examination Survey 1999–2000. CRP was measured with a high-sensitivity latex-enhanced turbidimetric assay.
Results: CRP concentration ranged from 0.1 to 296.0 mg/L (median, 2.7 mg/L). After exclusion of women with a CRP concentration >10 mg/L, the median was 2.2 mg/L. Approximately 25.7% of women, representing 
26.8 million women, did not report using hormone replacement therapy and had a CRP concentration >3.0 to 10 mg/L, a category considered to indicate high risk for cardiovascular disease. Multiple linear regression analysis included age, race or ethnicity, education, smoking status, total cholesterol concentration, triglyceride concentration, systolic blood pressure, waist circumference, and concentrations of glucose, insulin, c-peptide, and glycated hemoglobin. CRP concentration varied by race or ethnicity (Mexican American > white) and hormone replacement therapy (users > nonusers). In addition, significant and independent associations existed between CRP and waist circumference, total cholesterol and triglyceride concentrations, and systolic blood pressure but not age, smoking status, alcohol use, insulin concentration, glycated hemoglobin, and c-peptide concentration.
Conclusion: Large numbers of US women have an increased concentration of CRP.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To date, the best data about the distribution of CRP concentrations among US women have come from the Women’s Health Study (WHS) (3). However, these women were not a representative sample of US women: those <45 years of age were not included, 95% of participants were white, and all participants were health professionals. To build on the previous knowledge of the distribution of CRP among US women, we analyzed data from the National Health and Nutrition Examination Survey (NHANES) 1999–2000 with the following three objectives: (a) to describe the distribution of CRP concentration among adult women representative of the US population, including different age and racial or ethnic groups; (b) to estimate the proportions and numbers of adult women at high risk for cardiovascular disease based on their CRP concentrations; and (c) to examine correlates of CRP concentrations among women.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
60 years of age, African Americans, and Mexican Americans were oversampled. The study received human subjects approval from the CDC, and participants signed an informed consent form. After processing of blood samples, serum samples were frozen, stored at -20 °C, and shipped to the University of Washington Medical Center (Seattle, WA), where CRP concentrations were measured by latex-enhanced nephelometry (N High Sensitivity CRP assay) on a BN II nephelometer (Dade Behring Inc.) (5). Two levels of control materials from Bio-Rad Laboratories, Inc. were used for quality control, and CV ranged from 4.9% to 7.8%. The means for the control materials were 1.67 and 3.82 mg/L for samples analyzed during an initial 9-month period and 1.84 and 3.95 mg/L for a subsequent 12-month period.
We examined the CRP concentration distribution by age and by race or ethnicity. In addition, we examined the relationships between CRP concentration and education, smoking status, total cholesterol concentration, systolic blood pressure, waist circumference, alcohol use, hormone replacement therapy, glucose concentration, insulin concentration, c-peptide concentration, and glycated hemoglobin concentration.
Although the focus of our study was on women, we did compare CRP concentrations between men and women. We also examined the differences between men and women, using data from NHANES III in addition to the data from NHANES 1999–20000. In NHANES III, CRP was measured by latex-enhanced nephelometry, a low-sensitivity test. A more detailed description of CRP measurements in NHANES III has been published previously (6).
We limited our analyses to women and men 20 years of age who attended the mobile examination center. We calculated a detailed percentile distribution of CRP concentration for all women, for six age groups (20–29, 30–39, 40–49, 50–59, 60–69, and 70 years), and for three race or ethnic groups (white, African American, and Mexican American). We also calculated percentiles for women who were not using hormone replacement therapy because its use may increase CRP concentrations. We recalculated the percentiles after excluding women with concentrations >10 mg/L because such increased concentrations may be attributable to conditions other than cardiovascular disease. We used SAS 8.02 to generate weighted percentiles of the CRP concentration distribution using the mobile examination center sampling weights. To estimate the numbers of women at high risk for cardiovascular disease based on their CRP concentrations, we applied the proportions of women with a CRP concentration >3 mg/L to the number of US women reported from the 2000 census.
To examine correlates of CRP, we performed linear regression analyses using log-transformed CRP concentration as the dependent variable to improve its skewed distribution. Three analyses were conducted. The first included all women with complete data for age, race or ethnicity, education, smoking status, total cholesterol concentration, systolic blood pressure, waist circumference, alcohol use, and hormone replacement therapy. The second was limited to nonpregnant women who participated in the morning examination and fasted 8 h and for whom measurements of glucose, insulin, c-peptide, and glycated hemoglobin concentrations were made. In the third analysis, we examined whether sex was an independent predictor of CRP concentration after adjusting for the variables listed included in the first analysis. We used SUDAAN for regression analyses of log-transformed CRP concentration among the four race or ethnic groups.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
20 years of age had a value for CRP concentration, and 375 women had a CRP concentration >10 mg/L. CRP ranged from 0.1 to 296.0 mg/L (Table 1
). The geometric mean (SE) concentration was 2.4 (0.1) mg/L, and the median was 2.7 mg/L.
|
We were able to determine hormone replacement therapy status for 2044 women, of whom 506 reported ever using hormone replacement therapy and 257 reported currently using such therapy. The unadjusted geometric mean CRP concentration was 3.8 mg/L (age-adjusted, 3.4 mg/L) among the 257 current users, 2.8 mg/L (age-adjusted, 2.5 mg/L) among the 249 former users, and 2.2 mg/L (age-adjusted, 2.3 mg/L) among the 1787 women who did not use such therapy (Wald F-test, P <0.001). Compared with women who currently used hormone replacement therapy, only women who had never used such hormones had a significantly lower CRP concentration (t-test, P <0.001 from linear regression).
Percentiles of CRP concentration for all women and women who did not use hormone replacement therapy are shown in Table 1
. The median CRP concentration was highest among women 50–59 years of age. Whites had a lower age-adjusted geometric mean (2.3 mg/L) than either African Americans (3.1 mg/L; t-test, P = 0.007) or Mexican Americans (3.5 mg/L; t-test, P <0.001). Unadjusted geometric mean concentrations were 2.3 mg/L among white women, 3.0 mg/L among African-American women, and 3.2 mg/L among Mexican-American women. Among women who did not report using hormone replacement therapy, the age-adjusted geometric means were 2.2 mg/L (unadjusted, 2.1 mg/L) among white women, 3.1 mg/L (unadjusted, 3.0 mg/L) among African-American women, and 3.4 mg/L (unadjusted, 3.2 mg/L) among Mexican-American women. We also present the distribution of CRP after excluding participants with a CRP value >10 mg/L. The median concentration decreased to 2.1 mg/L for white women, 2.3 mg/L for African-American women, and 2.8 mg/L for Mexican-American women (Table 2
).
|
Among all women, 24.5% had a CRP concentration <1 mg/L, 29.7% had a concentration of 1–3 mg/L, 31.8% had a concentration >3 to 10 mg/L, and 14.0% had a concentration >10 mg/L (Table 3
). When we excluded women who were currently taking hormone replacement therapy, the respective percentages were 26.7%, 30.1%, 30.2%, and 13.0%. The age-specific percentages of women with a CRP concentration >3 mg/L are shown in Fig. 1
.
|
|
In univariate linear regression analysis, log-transformed CRP was not significantly associated with fasting time (Wald F-test, P = 0.783). In our first multiple regression model, we included all women with complete data for CRP concentration and independent variables. Log-transformed CRP concentration was significantly associated with race or ethnicity and hormone replacement therapy use (Table 4
). Mexican-American women had a higher concentration than white women, and users of hormone replacement therapy had a higher concentration than those who did not use such therapy. Women who had not completed high school tended to have a higher concentration than those with more than a high school education. In addition, total cholesterol concentration, systolic blood pressure, and waist circumference were all significantly and positively associated with CRP concentration. Although the Wald F-statistic for alcohol use was not significant (P = 0.210), the t-test suggested that women who reported having more than seven drinks per month vs those who had less than one drink per month had a lower CRP concentration. In a model limited to women with a CRP concentration 
10 mg/L, results were similar (data not shown).
|
In our second set of analyses, we examined a model limited to women 20 years of age who were not pregnant, attended the morning examination, had fasted 8 h, and had measurements for triglyceride, glucose, insulin, c-peptide, and glycated hemoglobin concentrations. Several differences between this model and the previous one were present (Table 4
). Race or ethnicity (Wald F-test, P = 0.157), educational status (Wald F-test, P = 0.425), and total cholesterol concentration were not significantly associated with CRP concentration. Systolic blood pressure, waist circumference, and triglyceride concentration were significantly and positively associated with CRP concentration. Glucose concentration was not a significant predictor, however. When we used the same set of covariates except glucose concentration, the insulin (t-test, P = 0.122), c-peptide (t-test, P = 0.204), and glycated hemoglobin (t-test, P = 0.765) concentrations were not significantly associated with CRP concentration. In regression analyses that adjusted for age only, glucose (t-test, P = 0.006), insulin (t-test, P <0.001), c-peptide (t-test, P <0.001), and glycated hemoglobin (t-test, P <0.001) concentrations were all significantly associated with CRP concentration. After adjustment for age and waist circumference, glucose and glycated hemoglobin concentrations were no longer significantly associated with CRP concentration, but insulin and c-peptide concentrations were. When we added triglyceride concentration to the model, neither insulin nor c-peptide concentration was significantly associated with CRP concentration.
Women had a higher CRP concentration than did men in unadjusted linear regression analysis and after adjustment for age, race or ethnicity, education, smoking status, total cholesterol concentration, triglyceride concentration, systolic blood pressure, and waist circumference. After we excluded women who currently used hormone replacement therapy, the adjusted geometric mean was 1.4 mg/L for men and 2.5 mg/L for women (t-test, P <0.001). Among participants with a CRP concentration 10 mg/L, the adjusted geometric mean concentration was 1.2 mg/L among men and 1.9 mg/L among women (t-test, P <0.001). Among white participants with a CRP concentration 10 mg/L, the adjusted geometric mean concentration was 1.2 mg/L among men and 1.9 mg/L among women (t-test, P <0.001).
In addition, we compared the age-adjusted proportions of men and women who had a CRP concentration >3 to 10 mg/L in NHANES III and NHANES 1999–2000 after excluding women who reported using hormone replacement therapy (Fig. 2
). In NHANES III, 16.4% of men (n = 7334) and 21.9% of women (n = 7679) had a CRP concentration >3 to 10 mg/L (P <0.001). In a logistic regression model from which participants with a CRP concentration >10 mg/L and women who reported currently using hormone replacement therapy were excluded, the odds ratio for having a CRP concentration >3 mg/L was 2.07 [95% confidence interval (CI), 1.73–2.46] for women compared with men after adjustment for age, race or ethnicity, education, smoking status, total cholesterol concentration, systolic blood pressure, waist circumference, and alcohol use. In NHANES 1999–2000, 22.1% of men (n = 1940) and 30.6% of women (n = 1787) had a CRP concentration >3 to 10 mg/L (P <0.001; multiple-adjusted odds ratio, 2.59; 95% CI, 1.96–3.42).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
20 years of age, representing 26.8 million women, did not report currently using hormone replacement therapy and had a CRP concentration >3 to 10 mg/L. An additional 11.0%, representing 11.5 million women who did not report using hormone replacement therapy, had a CRP concentration >10 mg/L. The CRP concentrations representing the cut points for the percentiles that divide the distribution in the range 0.1–10 mg/L into thirds among women who did not report using hormone replacement therapy are consistent with the cut points proposed by the CDC/American Heart Association statement about inflammatory markers (2).
The literature about possible differences in CRP concentration distribution among men and women is inconclusive. In the largest study, the median concentrations were in close agreement between US men and women (3). CRP concentrations were higher among women than men in NHANES 1999–2000, however; previously, we reported that the median concentration among men 20 years of age was 1.6 mg/L (7). In our present analyses of the same survey, the median concentration was 2.5 mg/L among women 20 years of age who did not use hormone replacement therapy. We also showed that the proportion of women who had a CRP concentration >3 mg/L was significantly higher than that among men in NHANES III. Thus, in two nationally representative surveys separated by 8–9 years, a larger percentage of women had a high CRP concentration than men. Other US studies have also reported that CRP concentrations were higher among women than men (8)(9)(10). Among European populations, men and women had similar median CRP concentrations (11). Some of the inconsistency among studies may reflect the different sociodemographic composition of the samples, differences in the prevalence of CRP correlates such as smoking and obesity, and the inclusion and exclusion criteria for selecting study participants. For example, women of the WHS were health professionals, whereas women in NHANES represent all walks of life. Almost 95% of women in the WHS were white, in contrast to NHANES, which included substantial numbers of African-American and Mexican-American women because they were oversampled. In the WHS, 12.6% of women reported smoking, and the mean body mass index was 26 kg/m2 (12). In NHANES 1999–2000, 21.5% of women reported currently smoking, and the mean body mass index was 28.2 kg/m2.
Few data have been available about racial or ethnic differences among women in the US (6)(13)(14). In the present study, we showed that Mexican-American women had a higher adjusted CRP concentration measured with a high-sensitivity assay than did white women. African-American women had a higher age-adjusted geometric mean concentration than white women, but the difference was no longer significant after adjustment for potential confounders. The higher CRP concentrations among Mexican-American women than white women could indicate that Mexican-American women are at increased risk for cardiovascular disease, especially if their risk functions for cardiovascular disease are similar to those of other women.
The results from our analysis of NHANES 1999–2000 data also showed significant differences among CRP concentrations based on use of hormone replacement therapy and was suggestive of differences according to educational status. Significant associations also existed between CRP and body mass index, waist circumference, total cholesterol concentration, triglyceride concentration, and systolic blood pressure. We did not find significant independent associations with age, smoking status, alcohol use, insulin concentration, glycated hemoglobin concentration, and c-peptide concentration. Previous studies of correlates of CRP concentration, such as those we included in our study, among women have produced inconsistent results for many correlates (6)(8)(9)(10)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37)(38)(39)(40).
Perhaps the major limitation of our study is the cross-sectional study design, which limits the interpretation of cause and effect of the associations we found. Unfortunately, our sample size was not sufficient to report CRP concentration distributions for women of other Hispanic ethnicities or of other races or ethnicities. Although our study included some of the largest numbers of nonwhite participants to date, a larger sample size would have allowed us to perform more detailed analyses. Future studies with larger sample sizes of nonwhite women would be helpful to better examine racial or ethnic differences in distributions and determinants of CRP concentration.
In conclusion, our results should further the understanding of the distribution and correlates of CRP concentration among US women. Additional studies of CRP among US women of other races or ethnicities could help to improve understanding of the distribution and correlates of CRP among growing population groups.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  T. W. McDade, J. N. Rutherford, L. Adair, and C. Kuzawa Adiposity and Pathogen Exposure Predict C-Reactive Protein in Filipino Women J. Nutr., December 1, 2008; 138(12): 2442 - 2447. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kelley-Hedgepeth, D. M. Lloyd-Jones, A. Colvin, K. A. Matthews, J. Johnston, M. R. Sowers, B. Sternfeld, R. C. Pasternak, C. U. Chae, and for the SWAN Investigators Ethnic Differences in C-Reactive Protein Concentrations Clin. Chem., June 1, 2008; 54(6): 1027 - 1037. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S-S. Lee, S. Singh, L. Magder, and M. Petri Predictors of high sensitivity C-reactive protein levels in patients with systemic lupus erythematosus Lupus, February 1, 2008; 17(2): 114 - 123. [Abstract] [PDF]
|
|
|
|
 |
|
 E. M. Crimmins, J. K. Kim, D. E. Alley, A. Karlamangla, and T. Seeman Hispanic Paradox in Biological Risk Profiles Am J Public Health, July 1, 2007; 97(7): 1305 - 1310. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Ye, Z. Yu, H. Li, O. H. Franco, Y. Liu, and X. Lin Distributions of C-Reactive Protein and its Association With Metabolic Syndrome in Middle-Aged and Older Chinese People J. Am. Coll. Cardiol., May 1, 2007; 49(17): 1798 - 1805. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 NACB WRITING GROUP MEMBERS, D. A. Morrow, C. P. Cannon, R. L. Jesse, L. K. Newby, J. Ravkilde, A. B. Storrow, A. H.B. Wu, and R. H. Christenson National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: Clinical Characteristics and Utilization of Biochemical Markers in Acute Coronary Syndromes Circulation, April 3, 2007; 115(13): e356 - e375. [Full Text] [PDF]
|
|
|
|
|
|
NACB WRITING GROUP MEMBERS, D. A. Morrow, C. P. Cannon, R. L. Jesse, L. K. Newby, J. Ravkilde, A. B. Storrow, A. H.B. Wu, R. H. Christenson, NACB COMMITTEE MEMBERS, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: Clinical Characteristics and Utilization of Biochemical Markers in Acute Coronary Syndromes Clin. Chem., April 1, 2007; 53(4): 552 - 574. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ishikawa, Y. Tamura, S. Hoshide, K. Eguchi, S. Ishikawa, K. Shimada, and K. Kario Low-Grade Inflammation Is a Risk Factor for Clinical Stroke Events in Addition to Silent Cerebral Infarcts in Japanese Older Hypertensives: The Jichi Medical School ABPM Study, Wave 1 Stroke, March 1, 2007; 38(3): 911 - 917. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Blumenthal, E. D. Michos, and K. Nasir Further Improvements in CHD Risk Prediction for Women JAMA, February 14, 2007; 297(6): 641 - 643. [Full Text] [PDF]
|
|
|
|
|
|
M. A. Albert and P. M Ridker C-Reactive Protein as a Risk Predictor: Do Race/Ethnicity and Gender Make a Difference? Circulation, August 1, 2006; 114(5): e67 - e74. [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. Hogue Jr, T. Hershey, D. Dixon, R. Fucetola, A. Nassief, K. E. Freedland, B. Thomas, and K. Schechtman Preexisting cognitive impairment in women before cardiac surgery and its relationship with C-reactive protein concentrations. Anesth. Analg., June 1, 2006; 102(6): 1602 - 1608. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C. Kao, S.-C. Shiesh, and T.-J. Wu Serum C-reactive protein as a marker for wellness assessment. Ann. Clin. Lab. Sci., March 1, 2006; 36(2): 163 - 169. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Jiang, D. R. Jacobs Jr., E. Mayer-Davis, M. Szklo, D. Herrington, N. S. Jenny, R. Kronmal, and R. G. Barr Nut and Seed Consumption and Inflammatory Markers in the Multi-Ethnic Study of Atherosclerosis Am. J. Epidemiol., February 1, 2006; 163(3): 222 - 231. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Khera, D. K. McGuire, S. A. Murphy, H. G. Stanek, S. R. Das, W. Vongpatanasin, F. H. Wians Jr, S. M. Grundy, and J. A. de Lemos Race and Gender Differences in C-Reactive Protein Levels J. Am. Coll. Cardiol., August 2, 2005; 46(3): 464 - 469. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Nicklas, T. You, and M. Pahor Behavioural treatments for chronic systemic inflammation: effects of dietary weight loss and exercise training Can. Med. Assoc. J., April 26, 2005; 172(9): 1199 - 1209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Levinson, J. J. Miller, and R. J. Elin Poor Predictive Value of High-Sensitivity C-Reactive Protein Indicates Need for Reassessment Clin. Chem., October 1, 2004; 50(10): 1733 - 1735. [Full Text] [PDF]
|
|
|
|
|
|
M. Lambert, E. E. Delvin, G. Paradis, J. O'Loughlin, J. A. Hanley, and E. Levy C-Reactive Protein and Features of the Metabolic Syndrome in a Population-Based Sample of Children and Adolescents Clin. Chem., October 1, 2004; 50(10): 1762 - 1768. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
