Prevalence of High C-Reactive Protein in Persons with Serum Lipid Concentrations within Recommended Values

Papers in Press. First published June 17, 2004 as doi:10.1373/clinchem.2004.036004 Clinical Chemistry 50:9 000 000 (2004) Lipids, Lipoproteins, and Cardiovascular Risk Factors Prevalence of High C-Reactive Protein in Persons with Serum Lipid Concentrations within Recommended Values Umed A. Ajani, * Earl S. Ford, and Ali H. Mokdad Background: C-Reactive protein (CRP) has been shown to be a strong predictor of coronary heart disease (CHD) and is being considered in cardiovascular risk disease assessment. The number of normolipidemic individuals who are eligible for evaluation of CRP in overall CHD risk assessment is not known. Methods: We analyzed data from the National Health and Nutrition Examination Survey 1999 2000 and computed the prevalence of high CRP (>3 mg/l) among normolipidemic adults. We also computed the prevalence among individuals free of CHD and diabetes. In addition, we examined the prevalence stratified by body mass index. Results: The prevalence of high CRP among those with lipid concentrations within recommended values ranged from 28.8% to 35.3% depending on the lipid fraction examined. Exclusion of individuals with CHD or diabetes and those with CRP concentrations >10 mg/l reduced the prevalence range (23.1 27.1%). Prevalence increased with increasing body mass index. Conclusions: In 2000, 12 million adults in the United States considered normolipidemic had high CRP concentrations. Additional studies to explore the role of CRP in cardiovascular disease risk assessment are needed. 2004 American Association for Clinical Chemistry Department of Adult and Community Health, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA. *Address correspondence to this author at: Centers for Disease Control and Prevention, 4770 Buford Hwy. NE, Mailstop K-66, Atlanta, GA 30341. Fax 770-488-8150; e-mail uajani@cdc.gov. Received April 22, 2004; accepted May 27, 2004. Previously published online at DOI: 10.1373/clinchem.2004.036004 Coronary heart disease (CHD) 1 is one of the major causes of morbidity and mortality in the United States, causing 1 in every 5 deaths in the year 2000 (1). One approach to reducing the incidence of and mortality from CHD is to identify individuals who are at high risk of developing CHD and administer preventive measures in a timely fashion. The goal is to be able to predict future coronary events so that preventive measures can be implemented. Several risk assessment algorithms for CHD have been developed. These generally incorporate such risk factors as age, smoking status, blood pressure, and serum lipid concentrations, including LDL-cholesterol (LDL-C), HDLcholesterol (HDL-C), and total cholesterol (TC) (2, 3). More recently, however, markers of inflammation, including cytokines and adhesion molecules, have been studied for their association with CHD (4). Of these, C-reactive protein (CRP), a nonspecific marker of inflammation, has consistently been shown to be a strong predictor of coronary events (5 8). Consequently, measuring CRP may help in identifying persons at high risk of a future coronary event (9). Of particular interest is the observation that increased concentrations of CRP have been found to predict increased risk of cardiovascular disease in people without overt hyperlipidemia (10, 11). Incorporating the measurement of CRP in clinical practice could have a critical long-term impact, but this requires further exploration in terms of costs and benefits. To better understand the ramifications of using CRP to identify people at high risk of cardiovascular disease, an estimate of the number of people with increased CRP concentrations is needed. We therefore used data from the National Health and Nutrition Examination Survey (NHANES) to study the prevalence of high concentrations of CRP among normolipidemic individuals. 1 Nonstandard abbreviations: CHD, coronary heart disease; LDL-C and HDL-C, LDL- and HDL-cholesterol, respectively; TC, total cholesterol; CRP, C-reactive protein; and NHANES, National Health and Nutrition Examination Survey. 1 Copyright 2004 by The American Association for Clinical Chemistry

2 Ajani et al.: High CRP in Normolipidemic Individuals Materials and Methods A representative sample of the noninstitutionalized US civilian population included in NHANES 1999 2000 was used for this study (12). A stratified multistage sampling design was used, and African-American and Mexican- American participants, those 60 years of age, and those with low income were oversampled. The participants were interviewed by trained interviewers using a computer-assisted personal interview system. The interviews were conducted at each participant s home, and the participants were requested to attend a mobile examination center. Additional questionnaires were completed, various examinations were conducted, and blood samples were collected at the mobile examination center. Greater details of the collection and processing of blood samples are provided in the NHANES Laboratory/Medical Technologists Procedures Manual (13). Specimens were stored at 20 C until shipment to the University of Washington for CRP measurement or to the Lipoprotein Analytical Laboratory at Johns Hopkins University for lipid measurements. The CRP concentrations were measured by latex-enhanced nephelometry on a Behring Nephelometer. Cholesterol was measured by enzymatic assays on a Hitachi 717 Analyzer (Roche Diagnostics) with commercial reagents. Serum control pools were obtained from Solomon Park Research Laboratories. Information regarding existing medical conditions was collected by questionnaires. Participants were asked, Has a doctor or other health professional ever told you that you had coronary heart disease? Similar questions were asked for angina and heart attack (also called myocardial infarction). In addition, the participants were asked, Have you ever been told by a doctor or other health professional that you have diabetes or sugar diabetes? Those responding affirmatively were considered to have these conditions. Participants were also asked whether a doctor or other health professional had advised them to take prescribed medicine to lower blood cholesterol and whether they were following that advice. The answers to these questions were used to identify participants currently taking cholesterol-lowering medication. Demographic data were also collected by questionnaire and used as follows: age at screening (continuous), gender, race or ethnicity (white, African-American, Mexican- American, other). Body mass index, defined as body weight in kilograms divided by height (in meters) squared (kg/m 2 ), was categorized into three groups: normal ( 25 kg/m 2 ), overweight (25 to 30 kg/m 2 ), and obese ( 30 kg/m 2 ). analysis A total of 2188 participants 20 years and older attended the morning clinic and provided fasting blood samples. We limited our analyses to participants with no missing information for age, gender, race/ethnicity, body mass index, and concentrations of CRP, HDL-C, LDL-C, and TC. This gave 1774 participants available for data analyses. We used the American Heart Association s recommended values to categorize serum lipid concentrations (14). Participants with serum HDL-C concentrations of 400 mg/l and TC concentrations 2000 mg/l were considered at high risk. For LDL-C, participants with serum concentrations 1300 mg/l were considered at high risk. We also used a combined category called the normal lipid profile, which included participants with all three serum lipid concentrations (HDL-C, LDL-C, and TC) within the recommended intervals. For CRP, serum concentrations 3.0 mg/l were considered high, as recommended by the American Heart Association and the CDC (4). We computed crude and adjusted prevalence of high CRP for those with recommended concentrations of serum HDL-C, LDL-C, and TC and for those with a normal lipid profile. The prevalence estimates were adjusted for age, sex, and race/ethnicity. The recommendations regarding the serum LDL-C concentration depend on the presence of other risk factors; therefore, we also conducted our analyses by excluding participants with existing CHD and diabetes. Several reports have shown a strong association between obesity and serum concentrations of CRP (15 17); we therefore computed the prevalence of CRP among those with serum lipid concentrations within by categories of body mass index. We used SUDAAN (Software for the Statistical Analysis of Correlated Data) (18) to account for the complex sampling design and to calculate the weighted estimates. Results The crude prevalence of high CRP concentration among those with an overall normal lipid profile was 28.8% (Table 1). The prevalence of high CRP concentration increased with age and was higher among women and Mexican-American participants. The adjusted prevalence of high CRP value by age, gender, and race/ethnicity is presented in Fig. 1. Adjustment for these factors did not alter the results. When examined by each lipid separately, the prevalence of high CRP concentrations ranged from 31.2% to 35.3% for crude values and from 32.8% to 34.7% after adjustment. Exclusion of individuals with existing CHD and diabetes led to slightly lower estimates. The adjusted prevalence of high CRP among individuals with a normal lipid profile was 28.6%. The range by separate lipid concentrations was 31.8 33.4%. After we excluded those with heart disease and diabetes, approximately one-fifth (19.7%) of those with normal lipid profiles were obese, approximately one in four persons (25.9%) were overweight, and the remaining 54.4% were within the normal body mass index range ( 25 kg/m 2 ). The prevalence of high CRP concentration varied widely by body mass index. Those within the normal body mass index range had a crude prevalence of

Clinical Chemistry 50, No. 9, 2004 3 Table 1. Prevalence of high CRP concentrations (>3.0 mg/l) among those with serum lipid concentrations within, NHANES 1999 2000. Normal LDL-C a Normal HDL-C a Normal TC a Normal lipid profile a All participants 57.4% 75.5% 50.6% 33.5% Median CRP concentrations, mg/l 2.0 2.0 1.7 1.5 Crude (SE) prevalence of high CRP, % 34.3 (1.9) 35.3 (1.9) 31.2 (2.1) 28.8 (2.5) Adjusted (SE) b prevalence of high CRP, % 34.7 (1.9) 34.1 (1.9) 32.8 (2.2) 29.1 (2.6) Estimated no. with high CRP 39 614 994 53 556 554 31 725 125 19 404 367 Persons without CHD and diabetes mellitus 56.8% 77.0% 50.3% 34.5% Median CRP concentrations, mg/l 1.9 1.9 1.7 1.4 Crude (SE) prevalence of high CRP, % 32.9 (2.0) 34.3 (2.0) 29.9 (2.2) 28.1 (2.7) Adjusted (SE) b prevalence of high CRP, % 33.4 (1.9) 33.0 (1.9) 31.8 (2.3) 28.6 (2.7) Estimated no. with high CRP 33 060 702 46 546 352 26 652 940 17 156 960 Stratified by body mass index (kg/m 2 ) Normal ( 25.0 kg/m 2 ) Crude (SE) prevalence of high CRP, % 20.5 (2.4) 19.8 (2.2) 16.0 (2.5) 16.7 (2.8) Adjusted (SE) b prevalence of high CRP, % 21.5 (2.2) 19.5 (2.0) 18.0 (2.6) 17.9 (2.8) Estimated no. with high CRP 9 686 302 12 525 522 6 754 469 5 520 589 Overweight (25.1 to 30.0 kg/m 2 ) Crude (SE) prevalence of high CRP, % 33.8 (4.0) 35.1 (3.2) 31.4 (4.2) 33.6 (5.6) Adjusted (SE) b prevalence of high CRP, % 34.9 (4.3) 34.7 (3.4) 33.8 (4.4) 34.5 (5.5) Estimated no. with high CRP 9 939 317 14 163 735 8 307 616 5 317 662 Obese ( 30 kg/m 2 ) Crude (SE) prevalence of high CRP, % 56.1 (4.1) 60.9 (3.3) 56.3 (4.5) 52.5 (6.1) Adjusted (SE) b prevalence of high CRP, % 54.3 (3.7) 57.1 (3.4) 55.3 (4.2) 49.3 (5.6) Estimated no. with high CRP 13 435 082 19 857 096 11 590 855 6 318 710 a See text; defined according to the American Heart Association (14). b Adjusted for age, sex, and race or ethnicity. high CRP ranging from 16.0% (for those with TC within the recommended range) to 20.5% (for those with LDL-C within the recommended range). The prevalence of high CRP concentration increased with increasing body mass Fig. 1. Adjusted prevalence of high CRP among people with serum lipid concentrations within by gender, age groups, and race/ethnicity. Each prevalence is adjusted by other two demographic characteristics. u, all individuals; f, individuals with CRP 10 mg/l excluded. yrs, years; Amer, American. index, with prevalences of 31.4 35.1% for overweight people and 52.5 60.9% for obese participants (Table 1). The possibility exists that CRP concentrations 10 mg/l among normolipidemic individuals may be attributable to other infections or inflammation (4). After we excluded participants with CRP concentrations 10 mg/l, the adjusted prevalence of high CRP ( 3 mg/l) was 23.1% among those with a normal lipid profile, but ranged between 25.9% and 27.1% when examined by each lipid separately (Table 2). Similarly, the adjusted prevalences of high CRP among normal, overweight, and obese participants with a normal lipid profile were 14.4%, 31.6%, and 36.0% respectively. Earlier studies reported that the use of lipid-lowering drugs may lower serum concentrations of CRP (19, 20). Exclusion of participants who reported current use of prescription medicines for high cholesterol did not change our results (data not shown). Discussion The data from NHANES 1999 2000 show that approximately one in every four persons 20 years of age and older without CHD and diabetes and with serum lipid concen-

4 Ajani et al.: High CRP in Normolipidemic Individuals Table 2. Prevalence of high CRP concentrations (>3.0 mg/l) among persons with serum lipid concentrations within after exclusion of those with CRP >10 mg/l, NHANES 1999 2000. Normal LDL-C a Normal HDL-C a Normal TC a Normal lipid profile a All participants 57.4% 75.9% 51.4% 34.2% Median CRP concentrations, mg/l 1.7 1.6 1.4 1.2 Crude (SE) prevalence of high CRP, % 26.7 (1.9) 28.1 (2.2) 24.3 (1.9) 22.2 (2.4) Adjusted (SE) b prevalence of high CRP, % 27.3 (1.9) 27.3 (2.2) 25.8 (2.0) 22.9 (2.5) Estimated no. with high CRP 26 900 632 37 538 753 21 915 520 13 363 660 Persons without CHD and diabetes mellitus 57.1% 77.2% 51.3% 35.1% Median CRP concentrations, mg/l 1.7 1.6 1.4 1.1 Crude (SE) prevalence of high CRP, % 26.2 (2.0) 27.5 (2.2) 24.0 (2.1) 22.1 (2.7) Adjusted (SE) b prevalence of high CRP, % 27.1 (2.0) 26.6 (2.2) 25.9 (2.2) 23.1 (2.7) Estimated no. with high CRP 23 577 949 33 359 076 19 336 181 12 243 745 Stratified by body mass index Normal ( 25.0 kg/m 2 ) Crude (SE) prevalence of high CRP, % 15.9 (2.4) 16.0 (2.3) 12.7 (2.5) 13.0 (2.7) Adjusted (SE) b prevalence of high CRP, % 17.2 (2.2) 16.0 (2.2) 14.7 (2.5) 14.4 (2.7) Estimated no. with high CRP 6 980 415 9 474 718 5 067 842 4 060 391 Overweight (25.1 to 30.0 kg/m 2 ) Crude (SE) prevalence of high CRP, % 30.3 (3.8) 30.9 (3.3) 28.4 (4.0) 30.9 (5.8) Adjusted (SE) b prevalence of high CRP, % 30.9 (4.1) 29.8 (3.4) 30.3 (4.3) 31.6 (5.8) Estimated no. with high CRP 8 291 472 11 480 752 7 070 541 4 619 161 Obese ( 30 kg/m 2 ) Crude (SE) prevalence of high CRP, % 44.6 (4.3) 49.7 (4.0) 44.9 (4.7) 38.9 (5.8) Adjusted (SE) b prevalence of high CRP, % 43.0 (3.9) 46.1 (4.1) 44.1 (4.3) 36.0 (5.1) Estimated no. with high CRP 8 306 062 12 403 606 7 197 798 3 564 194 a See text, defined according to the American Heart Association (14). b Adjusted for age, sex, and race or ethnicity. trations within had a high CRP concentration ( 3 mg/l). The prevalence of high CRP concentrations varied with body mass index, and at least one-third of the obese people with lipid concentrations within and without CHD or diabetes had a high CRP concentration. Higher serum concentrations of CRP have been shown to predict future cardiovascular events in apparently healthy individuals and among both men and women. At present, however, serum lipid concentrations are generally used, in addition to other behavioral risk factors, to identify people at high risk of coronary events. Although high cholesterol concentrations do predict future coronary events, many such events occur among normolipidemic individuals who appear to be at low clinical risk. This has led to the evaluation of new cardiovascular disease risk factors to better predict coronary events (21). Discussion about the role of CRP in the primary prevention of cardiovascular events, its comparison with serum lipid concentrations in the prediction of future events, its clinical applications, and the possibility of including CRP testing as a screening tool for cardiovascular disease is ongoing (10, 11, 22, 23). Our estimates are based on persons 20 years of age and older with no history of diagnosed CHD and diabetes and with a normal serum lipid profile. We used our sample proportions to compute population estimates. According to the 2000 census population of the US, an estimated 177 million people are 20 years of age and older and have no history of diabetes or heart disease. The estimate for the number of people with lipid concentrations within recommended ranges depends on the specific lipid fraction used: HDL-C, LDL-C, or TC. If the overall normal lipid profile is used, 34%, or 61 million people, have a normal lipid profile. Of these, 28%, or 17.2 million, have high CRP concentrations. This estimate is reduced to 12.2 million after people with CRP concentrations 10 mg/l are excluded. At least part of the high CRP concentrations may be explained by obesity and obesity-related conditions. On the basis of prevalence estimates among those with a normal lipid profile by body mass index, an estimated 5.5 million people with a body mass index 25 kg/m 2 have a high CRP concentration. This estimate is reduced to 4.1 million after the exclusion of people with CRP concentrations 10 mg/l. The population estimates are much higher if examined by each lipid separately. If the prevalence estimates for those with a body mass index

Clinical Chemistry 50, No. 9, 2004 5 25 kg/m 2 were applied to all eligible participants, a total of 7.1 million people would be expected to have a high CRP concentration instead of 12.2 million. This difference of 5.1 million people may be attributable to obesity or obesity-related conditions. We took a conservative approach and made our estimates after excluding people with diagnosed CHD or diabetes and CRP concentrations 10 mg/l. It is possible that some of those with a CRP concentration 10 mg/l may actually be at high risk of a coronary event. The increased risk of a coronary event among people with increased CRP concentrations and lipid concentrations within can be mitigated in at least two ways. The public health perspective is to promote and encourage behavioral changes that may lead to decreases in CRP concentrations. These include reducing overall obesity and waist circumference (24, 25), cessation of smoking (26), better diet with increased fiber intake (27), and physical activity (28). Behavioral changes in these modifiable risk factors have been shown to lower CRP concentrations and, hence, the risk of coronary events. Alternatively, the clinical approach to reduce serum concentrations of cholesterol and CRP with therapeutic and pharmacologic agents has also been discussed (20). In conclusion, our results provide important information about the size of the US population with lipid concentrations within who may be eligible for CRP evaluation. Additional studies are needed to further clarify the predictive value of CRP among normolipidemic persons and to establish the cost benefit ratio of adding the CRP test to the routine assessment of coronary risk as suggested by the American Heart Association (29). References 1. American Heart Association. Heart disease and stroke statistics, 2003 update. Dallas, TX: American Heart Association, 2002:1 42. 2. Wilson PW, D Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:1837 47. 3. Kinosian B, Glick H, Garland G. Cholesterol and coronary heart disease: predicting risks by levels and ratios. Ann Intern Med 1994;121:641 7. 4. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO III, Criqui M, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 2003;107:499 511. 5. Kuller LH, Tracy RP, Shaten J, Meilahn EN. Relation of C-reactive protein and coronary heart disease in the MRFIT nested casecontrol study. Multiple Risk Factor Intervention Trial. Am J Epidemiol 1996;144:537 47. 6. Koenig W, Sund M, Frohlich M, Fischer HG, Lowel H, Doring A, et al. C-Reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men: results from the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. Circulation 1999;99:237 42. 7. Danesh J, Whincup P, Walker M, Lennon L, Thomson A, Appleby P, et al. Low grade inflammation and coronary heart disease: prospective study and updated meta-analyses. BMJ 2000;321:199 204. 8. Ridker PM, Hennekens CH, Buring JE, Rifai N. C-Reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000;342:836 43. 9. Rifai N, Ridker PM. Proposed cardiovascular risk assessment algorithm using high-sensitivity C-reactive protein and lipid screening. Clin Chem 2001;47:28 30. 10. Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation 2003;107: 363 9. 11. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557 65. 12. Centers for Disease Control and Prevention. NHANES 1999 2000 data files. www.cdc.gov/nchs/about/major/nhanes/ NHANES99_00.htm (accessed July 30, 2003). 13. Centers for Disease Control and Prevention. NHANES laboratory/ medical technologists procedures manual.????,??: Centers for Disease Control and Prevention,????. 14. American Heart Association. American Heart Association. cholesterol levels: AHA recommendation. www.americanheart.org/ presenter.jhtml?identifier 4500 (accessed December 18, 2003). 15. Hak AE, Stehouwer CD, Bots ML, Polderman KH, Schalkwijk CG, Westendorp IC, et al. Associations of C-reactive protein with measures of obesity, insulin resistance, and subclinical atherosclerosis in healthy, middle-aged women. Arterioscler Thromb Vasc Biol 1999;19:1986 91. 16. Ford ES. Body mass index, diabetes, and C-reactive protein among U.S. adults. Diabetes Care 1999;22:1971 7. 17. Suzuki K, Ito Y, Ochiai J, Kusuhara Y, Hashimoto S, Tokudome S, et al. Relationship between obesity and serum markers of oxidative stress and inflammation in Japanese. Asian Pac J Cancer Prev 2003;4:259 66. 18. Research Triangle Institute. SUDAAN user s manual. Research Triangle Park, NC: Research Triangle Institute. 19. Sager PT, Melani L, Lipka L, Strony J, Yang B, Suresh R, Veltri E. Effect of coadministration of ezetimibe and simvastatin on highsensitivity C-reactive protein. Am J Cardiol 2003;92:1414 8. 20. Ridker PM, Rifai N, Clearfield M, Downs JR, Weis SE, Miles JS, et al. Measurement of C-reactive protein for the targeting of statin therapy in the primary prevention of acute coronary events. N Engl J Med 2001;344:1959 65. 21. Ridker PM. Evaluating novel cardiovascular risk factors: can we better predict heart attacks? Ann Intern Med 1999;130:933 7. 22. Ridker PM. High-sensitivity C-reactive protein and cardiovascular risk: rationale for screening and primary prevention. Am J Cardiol 2003;92:17K 22K. 23. Koenig W. C-Reactive protein and cardiovascular risk: has the time come for screening the general population? Clin Chem 2001;47: 9 10. 24. Tchernof A, Nolan A, Sites CK, Ades PA, Poehlman ET. Weight loss reduces C-reactive protein levels in obese postmenopausal women. Circulation 2002;105:564 9. 25. Valsamakis G, McTernan PG, Chetty R, Al Daghri N, Field A, Hanif W, et al. Modest weight loss and reduction in waist circumference after medical treatment are associated with favorable changes in serum adipocytokines. Metabolism 2004;53:430 4. 26. Frohlich M, Sund M, Lowel H, Imhof A, Hoffmeister A, Koenig W. Independent association of various smoking characteristics with

6 Ajani et al.: High CRP in Normolipidemic Individuals markers of systemic inflammation in men. Results from a representative sample of the general population (MONICA Augsburg Survey 1994/95). Eur Heart J 2003;24:1365 72. 27. Ajani UA, Ford ES, Mokdad AH. Dietary fiber and C-reactive protein: findings from national health and nutrition examination survey data. J Nutr 2004;134:1181 5. 28. Ford ES. Does exercise reduce inflammation? Physical activity and C-reactive protein among U.S. adults. Epidemiology 2002;13: 561 8. 29. Smith SC Jr., Greenland P, Grundy SM. AHA Conference Proceedings. Prevention conference V: beyond secondary prevention: identifying the high-risk patient for primary prevention: executive summary. American Heart Association. Circulation 2000;101: 111 6.