Elevated Serum C-Reactive Protein Levels and Advanced Atherosclerosis in Youth

From the Louisiana State University Health Sciences Center (A.W.Z., S.H., G.T.M., J.P.S.), New Orleans; the University of Vermont, Burlington (R.P.T.); the University of Texas Health Science Center at San Antonio (C.A.M., H.C.M.); the Ohio State University (E.E.H.), Columbus; and the Southwest Foundation for Biomedical Research (H.C.M.), San Antonio, Tex.

Correspondence to Henry C. McGill, Jr, MD, Southwest Foundation for Biomedical Research, PO Box 760549, San Antonio, TX 78245-0549. E-mail hmcgill@icarus.sfbr.org

    Abstract

Objective— To determine the associations among serum C-reactive protein (CRP) concentration, age, sex, risk factors for coronary heart disease (CHD), and atherosclerosis in young people.

Methods and Results— In 1244 subjects 15 to 34 years of age, we measured gross atherosclerotic lesions in the right coronary artery (RCA) and abdominal aorta (AA) and American Heart Association (AHA) lesion grade in the left anterior descending (LAD) coronary artery; serum CRP, lipoprotein cholesterol, and thiocyanate (for smoking) concentrations; intimal thickness of renal arteries (for hypertension); glycohemoglobin (for hyperglycemia); and body mass index (for obesity). Serum CRP levels increased with age, were higher in women than in men, and were positively related to obesity and hyperglycemia. Serum CRP 10 mg/L was associated with more extensive gross raised lesions in the RCA after age 25 and in the AA after age 30. Serum CRP 3 was associated with a greater prevalence of AHA grade 5 lesions in the proximal LAD coronary artery after age 25. The associations of CRP with lesions were independent of the traditional CHD risk factors.

Conclusion— Serum CRP level is independently associated with advanced atherosclerosis in young persons.

In young subjects 15 to 34 years of age, serum CRP concentration was associated with raised lesions in the abdominal aorta and right coronary artery, and with American Heart Association grade 5 lesions in the proximal left anterior descending coronary artery, independently of traditional coronary heart disease risk factors.

Key Words: atherosclerosis ? youth ? risk factors ? C-reactive protein ? coronary heart disease

    Introduction

The serum concentration of C-reactive protein (CRP), an acute-phase reactant marker of inflammation, is an independent risk factor for coronary heart disease (CHD),1 but the value of CRP in CHD risk assessment is not firmly established.2 Evidence regarding the association of CRP with preclinical atherosclerosis is limited, and the available results are mixed. However, markers of inflammation are associated with the CHD risk factors in children;3 and elevated CRP levels are associated with surrogate markers of atherosclerosis, such as brachial artery flow-mediated responsiveness, in children.4

In this report, we describe the association of serum CRP levels with the gross and microscopic features of atherosclerosis in young men and women from the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) study.

    Methods

Study Subjects

Study subjects were persons 15 to 34 years of age who died of external causes (accidents, homicides, suicides) within 72 hours after injury and were autopsied within 48 hours after death in a cooperating forensic laboratory. The institutional review board of each participating center approved this study.

CRP Measurements

Serum CRP concentration was measured using a high-sensitivity ELISA method,5 which has a coefficient of variation of 5%. We grouped CRP levels into 3 categories selected before statistical analysis: low, 0CRP<3 mg/L (n=1024; 82.3%); intermediate, 3CRP<10 (n=155; 12.5%); and high, CRP10 (n=65; 5.2%). Analyses indicated no association of lesions with 1CRP<3 mg/L compared with CRP<1 mg/L, and we combined these levels into the 0CRP<3 mg/L category.

There was no association between CRP level and time interval between death and autopsy (r=0.018; P=0.5243; mean 20.4 hours; SD 10.9 hours), nor between CRP level and time interval between death and refrigeration of the body (r=0.013; P=0.6775; mean 4.4 hours; SD 5.1 hours).

CHD Risk Factor Measurements

Methods for measuring the CHD risk factors are described in previous publications.6–8 Briefly, we measured total serum cholesterol concentration and high-density lipoprotein (HDL) cholesterol concentration (after precipitation of other lipoproteins) by a cholesterol oxidase method and calculated non-HDL cholesterol by subtraction. A non-HDL cholesterol concentration 160 mg/dL (4.14 mmol/L) was considered elevated, and an HDL cholesterol concentration <35 mg/dL (0.91 mmol/L) was considered low. A serum thiocyanate level 90 μmol/L defined a smoker. Mean blood pressure 110 mm Hg, indicated by the intimal thickness of small renal arteries, was identified as hypertension. Body mass index (BMI) 30 kg/m2 indicated obesity, and red blood cell glycosylated hemoglobin 8% indicated hyperglycemia. CRP levels and all risk factor measurements were available for 1244 cases.

Arteries and Lesions

Three pathologists blindly and independently estimated the extent of fatty streaks and raised lesions in the abdominal aorta (AA) and right coronary artery (RCA) as described previously.9 The average of the 3 independent grades was used as the consensus grade. Gross assessment of lesions was available for 1223 AAs and 1192 RCAs.

Two pathologists blindly and independently evaluated left anterior descending (LAD) coronary artery sections10 using the American Heart Association (AHA) grading system.11 Differences were resolved by discussion and a consensus grade was determined. Grade 0 designated a normal artery with no intimal lipid and with or without adaptive intimal thickening. Grade 1 and 2 lesions corresponded to gross fatty streaks; grade 1 contained isolated macrophage foam cells, and grade 2 contained numerous macrophage foam cells and fine particles of extracellular lipid but no pools of extracellular lipid. Grade 3 lesions contained numerous macrophage foam cells and 1 pools of extracellular lipid, but no well-defined core of lipid, and represented intermediate or transitional lesions. Grade 4 lesions contained numerous macrophage foam cells plus a well-defined core of extracellular lipid covered by normal intima. Grade 5 lesions showed 1 cores of extracellular lipid, plus a thickened fibrous cap, vascularization, or calcification. Grade 4 and 5 lesions corresponded to gross raised lesions.11 Grade 6 lesions were not encountered. AHA grades were available for 884 LAD coronary arteries.

Topographical Maps

The morphometry laboratory digitized the image of each Sudan IV–stained AA and RCA and an outline of raised lesions on a black-and-white print and converted the images to a standard template.12 Composite images were assembled to indicate prevalence of each type of lesion at each location in the image of the artery.

Statistical Methods

The relation of CRP level to sex, race, age, and the CHD risk factors was analyzed by multiple regression analysis.13 The distribution of CRP levels was positively skewed, and we analyzed the logarithm of CRP levels. The relation of extent of lesions (percent intimal surface involved) to CRP, with adjustment for sex, race, age, and the CHD risk factors, also was analyzed by multiple regression analysis.13 A logit transformation, with a small constant added to avoid the logarithm of zero, was applied to the percent surface area involved. Overall association of AHA grade with CRP was analyzed using a 2 test for independence.14 Because expected frequencies were low, the StatXact software (3.0.2; Cytel Corp.) was used to calculate the P value. To examine the details of the association using the ordered nature of the AHA grade,14 the AHA grades were partitioned over a series of incremental cut points with the level of lesions for categorization as lesion positive rather than lesion negative becoming increasingly severe. These dichotomizations form the cumulative odds model.14 The proportional odds assumption was not valid for these data, and we report odds ratios (ORs) appropriate for each cut point. To analyze the association of CRP with AHA grade with adjustment for other variables, we used multivariable logistic regression analysis for the cumulative odds model.14

    Results

CRP and Sex, Race, and Age

Geometric mean CRP levels by sex, race, and 5-year age group are given in Table 1. CRP levels increased with age (P=0.0120), and women had higher CRP levels than men (P=0.0113). There were no significant differences between whites and blacks (P=0.4456) or interactions among sex, race, and age (P0.3466).

   TABLE 1. Geometric Mean CRP Level by Sex, Race, and 5-Year Age Group

CRP and CHD Risk Factors

Geometric mean CRP levels by risk factor groups are given in Table 2. CRP was not associated with non-HDL cholesterol, smoking status, or hypertension. CRP was higher in cases with low HDL after age 25 (age by HDL interaction P=0.0244). CRP was significantly associated with obesity (P=0.0187) and with hyperglycemia (P=0.0001).

   TABLE 2. Geometric Mean CRP Level by CHD Risk Factors, Adjusted for Sex, Race, and Age

CRP and Gross Lesions

The association of extent of gross lesions with CRP levels was similar whether adjusted for CHD risk factors or not adjusted, and only adjusted values are presented. The mean extent involvement with fatty streaks and raised lesions in the AA and RCA by 5-year age group and CRP level, adjusted for sex, race, and risk factors (non-HDL cholesterol, HDL cholesterol, smoking, hypertension, obesity, and hyperglycemia), is given in Figure 1. The associations of CRP with fatty streaks and raised lesions were similar for men and women (sex by CRP interaction P0.2792) and for whites and blacks (race by CRP interaction P0.6662) for both arteries.

   Figure 1. Mean percent intimal surface area (+SE) involved with fatty streaks (top panels) and raised lesions (bottom panels) in the AA (left panels) and RCA (right panels) by age and CRP level, adjusted for sex, race, and risk factors (high non-HDL cholesterol concentration, low HDL cholesterol concentration, smoking, hypertension, obesity, hyperglycemia). Open bars indicate CRP <3 mg/L; gray bars, 3CRP <10 mg/L; black bars, CRP 10 mg/L; y, years.

The extent of AA fatty streaks was not associated with CRP (P=0.9788). In 30- to 34-year-old subjects, the extent involvement with AA raised lesions was greater for those with CRP 10 compared with those with CRP <10 (P=0.0043).

The extent involvement with RCA fatty streaks was greater for those with 3CRP <10 compared with those with CRP <3 in every age group, although these differences were not statistically significant. The differences in RCA fatty streaks between those with CRP 10 and those with 3CRP <10 also were not significant. The decrease in RCA fatty streaks observed for cases with CRP 10 (Figure 1) was attributable to the conversion of fatty streaks to raised lesions.

The extent involvement with RCA raised lesions was greater for those with CRP 10 compared with those with CRP <10 in both 25- to 29-year-old subjects (P=0.0087) and 30- to 34-year-old subjects (P=0.0013). The extent involvement with RCA raised lesions did not differ significantly between those with 3 CRP <10 compared with those with CRP <3 (P=0.5808).

A total of 106 cases 25 to 34 years old had no CHD risk factors (Non-HDL cholesterol <160 mg/dL; HDL cholesterol 35 mg/dL; BMI <30 kg/m2; glycohemoglobin <8%; normotensive; and nonsmoker). Among these subjects, CRP was not significantly associated with AA fatty streaks, AA raised lesions, or RCA fatty streaks. Involvement with RCA raised lesions was greater for those with CRP 10 compared with cases with CRP <10 (0.4±0.1 for CRP <10; 1.1±0.4 for CRP 10; P=0.0092).

Topographical maps of raised lesion prevalence in the AA and RCA (Figure 2) show that after age 25, cases with 3CRP <10 have slightly greater lesion prevalence, and those with CRP 10 have substantially greater prevalence. The increased prevalence associated with CRP levels occurs in areas of the AA and RCA that are prone to raised lesions.12 Although unadjusted for CHD risk factors, these maps are consistent with the results of adjusted analyses of percent surface area involved with lesions (Figure 1).

   Figure 2. Prevalence maps of raised lesions in the AA (top panels) and RCA (bottom panels) by CRP level and 10-year age group. y indicates years.

CRP and LAD Lesions

The prevalence of AHA grade by 10-year age group and CRP level is shown in Table 3.

   TABLE 3. No. and Prevalence of Cases by 10-Year Age Group, AHA Grade, and CRP Level

There was an association of AHA grade and CRP level in 25- to 34-year-olds but not in 15- to 24-year-olds (15 to 24 years: 2=3.10, df=10, P=0.9783; 25 to 34 years: 2=16.61, df=10, P=0.0815). Partitioning indicated a significant association of AHA grade with CRP 3 compared with CRP <3 (2=14.517; df=5; P=0.0123) among 25- to 34-year-old persons but not among 15- to 24-year-old persons (2=0.986; df=5; P=0.9834). There was no association of AHA grade with CRP 10 compared with 3CRP <10 in either age group (15 to 24: 2=2.059, df=4, P=0.7517; 25 to 34: 2=2.742, df=5, P=0.7397). We combined the 3CRP <10 and CRP 10 categories for subsequent analyses.

To further examine the relation of CRP to AHA grade in individuals 25 to 34 years old, we partitioned the AHA grades into 5 contrasting levels of severity of lesions (Table 4). There was no association of CRP with AHA grades 1 through 5 versus grade 0 (OR, 1.15; P=0.5796). The association of CRP 3 with AHA grades 2 through 5 versus 0 and 1 (OR, 1.57; P=0.0599) was of borderline statistical significance. The ORs for AHA grades 3 through 5 versus 0 through 2 (OR, 1.51; P=0.1514) and grades 4 and 5 versus grades 0 through 3 (OR, 1.49; P=0.2835) were of similar magnitude to the OR for grades 2 through 5 versus 0 and 1, although not statistically significant. The association of CRP 3 with AHA grade 5 versus 0 through 4 was statistically significant (OR, 3.92; P=0.0054). The estimated OR of grade 5 versus grades 0 through 4 was the same whether comparing 3CRP <10 versus CRP <3 (OR, 3.92; P=0.0163) or CRP 10 versus CRP <3 (OR, 3.92; P=0.0685).

   TABLE 4. No. of Cases and Prevalence Among Sets of Cases Defined by Increasing Severity of AHA Lesion Grade, by CRP Category, 25- to 34-Year-Old Persons

In cases aged 25 to 34 years, the OR for CRP (CRP 3 versus CRP <3) and AHA grade 5 versus 0 through 4 in the LAD coronary artery, with multivariable adjustment for sex, race, age, and CHD risk factors, was 3.07 (95% CI, 1.08 to 8.75). ORs for other lesion comparisons were not statistically significant.

Seventy-seven cases 25 to 34 years old had no CHD risk factors. Three of the 77 cases, all with CRP <3, had grade 4 lesions, and none of the 77 cases had grade 5 lesions. The ORs for grades 2 through 4 versus grades 0 and 1 (OR, 5.75; 95% CI, 1.71 to 19.38) and grades 3 and 4 versus grades 0 through 2 (OR, 4.67; 95% CI, 1.19 to 18.26) were statistically significant.

To investigate the possibility that an underlying association of smoking with elevated CRP levels was responsible for the association of elevated CRP with greater prevalence of grade 5 lesions, in cases 25 to 34 years old, we examined the association of CRP 3 with grades 5 versus 0 through 4 by smoking status. In nonsmokers, OR was 4.33 (P=0.3450), and in smokers, OR was 3.90 (P=0.0107). In cases with BMI <30 kg/m2 and glycohemoglobin <8%, nonsmokers had an OR of 4.65 (P=0.1771), whereas smokers had an OR of 3.88 (P=0.0237). Although not all of these ORs were statistically significant because of the small numbers of cases, the consistency of the OR for CRP and lesions in nonsmokers and smokers suggested that the observed association with CRP was not attributable to an underlying association with smoking.

    Discussion

Summary of Results

Accounting for CHD risk factors, elevated serum CRP was associated with a greater extent of raised lesions in the RCA after age 25 and in the AA after age 30, and with a greater prevalence of grade 5 lesions after age 25. These findings indicate that elevated serum CRP is associated with accelerated progression of atherosclerosis in young adults independently of the traditional CHD risk factors.

CRP and CHD Risk Factors

The positive association of serum CRP with age and the higher serum CRP concentrations in women are consistent with previous reports.15,16 The positive relationships of CRP with obesity and hyperglycemia are also consistent with other reports of an association with the metabolic syndrome17 and with obesity in adults18,19 and children.20 The lack of an association of CRP with smoking in this study may be attributable to the shorter period of smoking in these young subjects compared with the longer exposure to smoking in middle-aged adults.21

CRP and Atherosclerotic Lesions

Two cross-sectional studies of adults showed no association between CRP levels and carotid intima-medial thickness (IMT) measured by ultrasound.21,22 Two other studies found no association between serum CRP and coronary artery calcification.23,24 The Framingham Study found an association between CRP and internal carotid IMT (but not common carotid IMT) only in women.25

CRP levels were higher in subjects dying with, but not resulting from, severe coronary atherosclerosis, and were also higher in subjects dying because of recent coronary thrombosis without infarction, than in subjects dying from noncardiovascular causes with minimal coronary atherosclerosis.26 Immune reactivity to heat shock proteins, but not CRP concentration, was associated with carotid and femoral IMT in young (17 to 18 years) men.27 Human transgene expression accelerated aortic atherosclerosis in apolipoprotein E–deficient mice.28

The AHA classification system is based on a temporal sequence that is deduced from lesions taken from standard locations in many persons of different ages.29 It is assumed that lesions of a given severity have passed through all lower levels of the lesion at an earlier age. Elevated serum CRP was associated with higher prevalence of grade 5 lesions after age 25. This result indicates that elevated serum CRP is associated with accelerated transition from grade 4 to more advanced grade 5 lesions. Grade 5 lesions have thicker and less orderly fibromuscular caps that have been disrupted by hematomas or thrombotic deposits. Thrombosis is likely to occur with an elevated CRP level, which is also associated with an elevated level of fibrin D-dimer.30 Whatever the mechanism, progression to a grade 5 lesion is clinically significant because such lesions are more likely to reduce lumen size.31

There was a difference between the results based on the gross assessment of raised lesions in the RCA and the microscopic assessment of lesions in the LAD coronary artery. Although AHA grades 4 and 5 correspond to gross raised lesions, these 2 lesion types cannot be separated by gross grading. Grade 5 lesions are more prevalent than grade 4 lesions in the 3CRP <10 and CRP 10 categories (Table 3). This result suggests that most of the gross raised lesions in the RCA in the elevated CRP categories are likely grade 5. Grade 5 lesions in the LAD coronary artery were associated with CRP 3, whereas the extent of RCA raised lesions was associated only with CRP >10. Possible reasons for this different association are that associations vary between different arteries and even between regions within arteries, or that microscopic assessments are simply more sensitive than the percent intimal surface involved with gross lesions.

In all of the studies relating CRP to atherosclerosis and CHD, including the present one, the analysis and interpretation of results are complicated by the association of CRP levels with obesity, hyperglycemia, and smoking, all of which are independently associated with atherosclerosis and CHD. Despite these possible sources of confounding, extensive analyses of data from PDAY subjects indicate that the association of CRP with advanced atherosclerosis is independent of the traditional CHD risk factors.

Most reports and reviews indicate that CRP >3 mg/L represents high risk,2 although some32 suggest that CRP >10 may be attributable to acute infections and that such values initially should be ignored. The association of CRP 10 with grade 5 lesions reported here suggests that the higher levels are also important, and 1 report finds that CRP 20 mg/L predicts cardiovascular events.33

Implications for Prevention

Numerous studies have reported that the statin drugs lower CRP levels as well as lower low-density lipoprotein (LDL) cholesterol levels, and statins have been suggested for subjects with high CRP levels34 as an additional measure to retard the progression of atherosclerosis. However, pharmacological intervention in young persons with high CRP levels may not be necessary because weight loss alone is associated with reduction in CRP levels.35 Exercise training and concomitant weight loss reduce CRP,36,37 regardless of CRP genotype,38 and cardiac rehabilitation and exercise training also are associated with substantial reductions in CRP levels independently of weight loss.39 Although these promising results of exercise and weight loss so far are limited to middle-aged and older adults, they suggest that similar effects are likely to occur in young adults. Thus, the same lifestyle practices recommended to control blood cholesterol, blood pressure, obesity, and hyperglycemia in youth40 are also likely to lower serum CRP levels.

Conclusion

The results of this study indicate that an elevated serum CRP concentration is independently associated with advanced coronary and abdominal aortic atherosclerosis in young adults. These results provide additional justification for recommending that young persons adopt a healthy lifestyle for the long-range prevention of CHD and related atherosclerotic diseases.

    Acknowledgments

Institutions cooperating in the PDAY study and supporting grants from the NHLBI and other sources: University of Alabama, Birmingham (HL-33733 and HL-33728); Albany Medical College, NY(HL-33765); Baylor College of Medicine, Houston, Tex (HL-33750); University of Chicago, Illinois (HL-33740 and HL-45715); the University of Illinois, Chicago (HL-33758); Louisiana State University Health Sciences Center, New Orleans (HL-33746 and HL-45720); Louisiana Cancer and Lung Trust Fund (2000-02-01); University of Maryland, Baltimore (HL-33752 and HL-45693); Medical College of Georgia, Augusta (HL-33772); University of Nebraska Medical Center, Omaha (HL-33778); the Ohio State University, Columbus (HL-33760 and HL-45694); Southwest Foundation for Biomedical Research, San Antonio, Tex (HL-39913); gift of Peter and Beth Dahlberg; the University of Texas Health Science Center at San Antonio (HL-33749 and HL-45719); Vanderbilt University, Nashville, Tenn (HL-33770 and HL-45718); and West Virginia University Health Sciences Center, Morgantown (HL-33748). Institution cooperating in this ancillary study and supporting grants: University of Vermont, Burlington (HL-46696 and HL-58329).

    Footnotes

Reprint requests to Jack P. Strong, MD, Department of Pathology, Louisiana State University Health Sciences Center, 1901 Perdido St, New Orleans, LA 70112-1393. E-mail jstron@lsuhsc.edu

References

Danesh J, Wheeler JG, Hirschfield GM, Eda S, Eiriksdottir G, Rumley A, Lowe GD, Pepys MB, Gudnason V. C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med. 2004; 350: 1387–1397.

Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO III, Criqui M, Fadl YY, Fortmann SP, Hong Y, Myers GL, Rifai N, Smith SC Jr, Taubert K, Tracy RP, Vinicor F. 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.

Ferguson MA, Gutin B, Owens S, Litaker M, Tracy RP, Allison J. Fat distribution and hemostatic measures in obese children. Am J Clin Nutr. 1998; 67: 1136–1140.

Jarvisalo MJ, Harmoinen A, Hakanen M, Paakkunainen U, Viikari J, Hartiala J, Lehtimaki T, Simell O, Raitakari OT. Elevated serum C-reactive protein levels and early arterial changes in healthy children. Arterioscler Thromb Vasc Biol. 2002; 22: 1323–1328.

Rifai N, Tracy RP, Ridker PM. Clinical efficacy of an automated high-sensitivity C-reactive protein assay. Clin Chem. 1999; 45: 2136–2141.

McGill HC Jr, McMahan CA, Malcom GT, Oalmann MC, Strong JP, for the PDAY Research Group. Effects of serum lipoproteins and smoking on atherosclerosis in young men and women. Arterioscler Thromb Vasc Biol. 1997; 17: 95–106.

McGill HC Jr, McMahan CA, Tracy RE, Oalmann MC, Cornhill JF, Herderick EE, Strong JP; for the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Research Group. Relation of a postmortem renal index of hypertension to atherosclerosis and coronary artery size in young men and women. Arterioscler Thromb Vasc Biol. 1998; 18: 1108–1118.

McGill HC Jr, McMahan CA, Malcom GT, Oalmann MC, Strong JP; Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Research Group. Relation of glycohemoglobin and adiposity to atherosclerosis in youth. Arterioscler Thromb Vasc Biol. 1995; 15: 431–440.

Strong JP, Malcom GT, McMahan CA, Tracy RE, Newman WP III, Herderick EE, Cornhill JF; for the Pathobiological Determinants of Atherosclerosis in Youth Research Group. Prevalence and extent of atherosclerosis in adolescents and young adults: implications for prevention from the Pathobiological Determinants of Atherosclerosis in Youth Study. J Am Med Assoc. 1999; 281: 727–735.

McGill HC Jr, McMahan CA, Zieske AW, Tracy RE, Malcom GT, Herderick EE, Strong JP, PDAY Research Group. Association of coronary heart disease risk factors with microscopic qualities of coronary atherosclerosis in youth. Circulation. 2000; 102: 374–379.

Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W Jr, Rosenfeld ME, Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Arterioscler Thromb Vasc Biol. 1995; 15: 1512–1531.

McGill HC Jr, McMahan CA, Herderick EE, Tracy RE, Malcom GT, Zieske AW, Strong JP; PDAY Research Group. Effects of coronary heart disease risk factors on atherosclerosis of selected regions of the aorta and right coronary artery. Arterioscler Thromb Vasc Biol. 2000; 20: 836–845.

Draper NR, Smith H. Applied Regression Analysis. New York, NY: John Wiley & Sons; 1998.

Agresti A. Categorical Data Analysis. 2nd ed. Hoboken, NJ: John Wiley & Sons Inc; 2002.

Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med. 1997; 336: 973–979.

Ridker PM, Buring JE, Shih J, Matias M, Hennekens CH. Prospective study of C-reactive protein and the risk of future cardiovascular events among apparently healthy women. Circulation. 1998; 98: 731–733.

Young TK, Gelskey DE. Is noncentral obesity metabolically benign? Implications for prevention from a population survey. J Am Med Assoc. 1995; 274: 1939–1941.

Rexrode KM, Buring JE, Manson JE. Abdominal and total adiposity and risk of coronary heart disease in men. Int J Obes. 2001; 25: 1047–1056.

Lemieux I, Pascot A, Prud’homme D, Almeras N, Bogaty P, Nadeau A, Bergeron J, Despres JP. Elevated C-reactive protein. Another component of the atherothrombotic profile of abdominal obesity. Arterioscler Thromb Vasc Biol. 2001; 21: 961–967.

Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med. 1997; 337: 869–873.

Tracy RP, Psaty BM, Macy E, Bovill EG, Cushman M, Cornell ES, Kuller LH. Lifetime smoking exposure affects the association of C-reactive protein with cardiovascular disease risk factors and subclinical disease in healthy elderly subjects. Arterioscler Thromb Vasc Biol. 1997; 17: 2167–2176.

Folsom AR, Pankow JS, Tracy RP, Arnett DK, Peacock JM, Hong Y, Djousse L, Eckfeldt JH; Investigators of the NHLBI Family Study. Association of C-reactive protein with markers of prevalent atherosclerotic disease. Am J Cardiol. 2001; 88: 112–117.

Hunt ME, O’Malley PG, Vernalis MN, Feuerstein IM, Taylor AJ. C-reactive protein is not associated with the presence or extent of calcified subclinical atherosclerosis. Am Heart J. 2001; 141: 206–210.

Redberg RF, Rifai N, Gee L, Ridker PM. Lack of association of C-reactive protein and coronary calcium by electron beam computed tomography in postmenopausal women: implications for coronary artery disease screening. J Am Coll Cardiol. 2000; 36: 39–43.

Wang TJ, Nam BH, Wilson PW, Wolf PA, Levy D, Polak JF, D’Agostino RB, O’Donnell CJ. Association of C-reactive protein with carotid atherosclerosis in men and women: the Framingham Heart Study. Arterioscler Thromb Vasc Biol. 2002; 22: 1662–1667.

Burke AP, Tracy RP, Kolodgie F, Malcom GT, Zieske A, Kutys R, Pestaner J, Smialek J, Virmani R. Elevated C-reactive protein values and atherosclerosis in sudden coronary death: association with different pathologies. Circulation. 2002; 105: 2019–2023.

Knoflach M, Kiechl S, Kind M, Said M, Sief R, Gisinger M, van der Zee R, Gaston H, Jarosch E, Willeit J, Wick G. Cardiovascular risk factors and atherosclerosis in young males: ARMY study (Atherosclerosis Risk-Factors in Male Youngsters). Circulation. 2003; 108: 1064–1069.

Paul A, Ko KW, Li L, Yechoor V, McCrory MA, Szalai AJ, Chan L. C-reactive protein accelerates the progression of atherosclerosis in apolipoprotein E-deficient mice. Circulation. 2004; 109: 647–655.

Stary HC. The histological classification of atherosclerotic lesions in human coronary arteries. In: Fuster V, Ross R, Topol EJ, eds. Atherosclerosis and Coronary Artery Disease. Philadelphia, Pa: Lippincott-Raven Publishers; 1996: 463–474.

Lowe GD, Sweetnam PM, Yarnell JWG, Rumley A, Rumley C, Bainton D, Ben-Shlomo Y. C-reactive protein, fibrin D-dimer, and risk of ischemic heart disease: the Caerphilly and Speedwell studies. Arterioscler Thromb Vasc Biol. 2004; 24: 1957–1962.

Stary HC. Natural history and histological classification of atherosclerotic lesions: an update. Arterioscler Thromb Vasc Biol. 2000; 20: 1177–1178.

Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation. 2003; 107: 363–369.

Ridker PM, Cook N. Clinical usefulness of very high and very low levels of C-reactive protein across the full range of Framingham Risk Scores. Circulation. 2004; 109: 1955–1959.

Ridker PM. High-sensitivity C-reactive protein and cardiovascular risk: rationale for screening and primary prevention. Am J Cardiol. 2003; 92: 17K–22K.

Tchernof A, Nolan A, Sites CK, Ades PA, Poehlman ET. Weight loss reduces C-reactive protein levels in obese postmenopausal women. Circulation. 2002; 105: 564–569.

Despres JP. CRP and risk of coronary heart disease. Can exercise training cool down the flames? Arterioscler Thromb Vasc Biol. 2004; 24: 1743–1745.

Okita k, Nishijima H, Murakami T, Nagai T, Morita N, Yonezawa K, Iizuka K, Kawaguchi H, Kitabatake A. Can exercise training with weight loss lower serum C-reactive protein levels? Arterioscler Thromb Vasc Biol. 2004; 24: 1868–1873.

Obisesan TO, Leeuwenburgh C, Phillips T, Ferrell RE, Phares DA, Prior SJ, Hagberg JM. C-reactive protein genotypes affect baseline, but not exercise training-induced changes, in C-reactive protein levels. Arterioscler Thromb Vasc Biol. 2004; 24: 1874–1879.

Milani RV, Lavie CJ, Mehra MR. Reduction in C-reactive protein through cardiac rehabilitation and exercise training. J Am Coll Cardiol. 2004; 43: 1056–1061.

McGill HC Jr, McMahan CA. Starting earlier to prevent heart disease. J Am Med Assoc. 2003; 290: 2320–2322.