C-Reactive Protein and Electrocardiographic ST-Segment Depression Additively Predict Mortality The Strong Heart Study

Transcription

1 Journal of the American College of Cardiology Vol. 45, No. 11, by the American College of Cardiology Foundation ISSN /05/$30.00 Published by Elsevier Inc. doi: /j.jacc C-Reactive Protein and Electrocardiographic ST-Segment Depression Additively Predict Mortality The Strong Heart Study Peter M. Okin, MD, FACC,* Mary J. Roman, MD, FACC,* Lyle G. Best, MD, Elisa T. Lee, PHD, James M. Galloway, MD, FACC, Barbara V. Howard, PHD, Richard B. Devereux, MD, FACC* New York, New York; Timber Lake, South Dakota; Oklahoma City, Oklahoma; and Washington, DC OBJECTIVES BACKGROUND METHODS RESULTS CONCLUSIONS This study was designed to examine whether high-sensitivity C-reactive protein (CRP) and electrocardiographic (ECG) ST-segment depression (STD) have additive utility for predicting cardiovascular disease (CVD) death and all-cause death (ACD). C-reactive protein, a marker of systemic inflammation, and ECG STD, an index of myocardial ischemia and hypertrophy, independently predict mortality. Electrocardiograms and CRP levels were examined in 2,155 American Indian participants in the second Strong Heart Study examination. ST-segment depression 50 V (n 127) and CRP 7.0 mg/l (defining the upper quartile of CRP levels, n 540) were considered abnormal. After years follow-up there were 95 CVD deaths and 310 ACD. In univariate Cox analyses, the combination of CRP and ECG STD improved risk stratification compared to either alone, with the presence of both CRP 7.0 and ECG STD associated with a 7.7-fold increased risk of CVD death (95% confidence interval [CI] 3.3 to 18.2) and a 6.5-fold increased risk of ACD (95% CI 4.1 to 10.3). After adjustment for age, gender, and relevant risk factors, the combination of high CRP and STD remained predictive of CVD death and ACD, with the presence of both abnormal CRP and STD associated with the highest risks of CVD death (hazard ratio [HR] 3.2, 95% CI 1.1 to 10.5) and ACD (HR 3.9, 95% CI 2.1 to 7.2) and the presence of either high CRP or abnormal STD associated with intermediate risks of CVD death (HR 2.2, 95% CI 1.4 to 3.4) and ACD (HR 1.5, 95% CI 1.2 to 2.0). The combination of ECG STD and CRP increases the risk of mortality, demonstrating the additive impacts of active inflammation and preclinical CVD on prognosis. (J Am Coll Cardiol 2005;45: ) 2005 by the American College of Cardiology Foundation The resting electrocardiogram (ECG) is an inexpensive and widely used noninvasive tool for assessing risk in population-based studies and clinical practice. ST-segment depression (STD) on the ECG, a sign of ventricular repolarization abnormality, is a well-established marker of risk in the general population (1 10) that is also strongly associated with underlying cardiovascular disease (CVD), including left ventricular hypertrophy (11,12). Indeed, even minimal degrees of computer-measured STD are associated with higher left ventricular mass and greater prevalence of anatomic ventricular hypertrophy (11) and provide additional prognostic information beyond that afforded by echocardiographic hypertrophy and traditional risk factors (10). C-reactive protein (CRP), a marker of systemic inflammation, has been demonstrated to predict incident CVD, hypertension, systemic atherosclerosis, sudden death, and From the *Division of Cardiology, Department of Medicine, Weill Medical College of Cornell University, New York, New York; Missouri Breaks Industries Research Inc., Timber Lake, South Dakota; College of Public Health, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma; University of Arizona, Tucson, Arizona; and MedStar Research Institute, Washington, DC. This work was supported by grants HL-41642, HL-41652, HL-41654, and HL from the National Heart, Lung, and Blood Institute, Bethesda, Maryland; and by grant M10RR (GCRC) from the National Institutes of Health, Bethesda, Maryland. The views expressed in this paper are those of the authors and do not necessarily reflect those of the Indian Health Service. Manuscript received December 8, 2004; revised manuscript received February 15, 2005, accepted February 22, CVD and all-cause mortality (13 26). The additive prognostic value of combining CRP with other markers of risk, such as low-density lipoprotein (LDL) or total cholesterol (19), glycated hemoglobin (20), metabolic syndrome (21,22), systemic hypertension (23), and Framingham risk scores (24,25) to improve risk stratification has been well documented. However, there are few data regarding the relationship between STD and CRP (27,28), and whether these two readily available noninvasive risk factors provide additive prognostic value has not been examined. Thus, the present study examined the relationship between highsensitivity CRP and ECG STD and whether CRP and STD have additive utility for predicting CVD and all-cause mortality, controlling for clinical and demographic variables that could confound these relationships. METHODS Study population. The Strong Heart Study is a population-based study of CVD and its risk factors in American Indians from 13 communities in Arizona, Oklahoma, and North and South Dakota. Information about the population, methods, and enrollment procedures for the study has previously been reported in detail (8,9,11,29,30). The current study examined a subset of 2,155 of the 3,638 original participants in the second Strong Heart Study exam (64% women, mean age 59 8 years) who had digital ECG

2 1788 Okin et al. JACC Vol. 45, No. 11, 2005 C-Reactive Protein and ST-Segment Depression June 7, 2005: Abbreviations and Acronyms ACD all-cause death CI confidence interval CRP C-reactive protein CVD cardiovascular disease ECG electrocardiogram/electrocardiographic HDL high-density lipoprotein HR hazard ratio LDL low-density lipoprotein STD ST-segment depression records showing sinus rhythm with no bundle branch block and with CRP levels. Electrocardiography. Standard 12-lead ECGs were performed with MAC-PC or MAC-12 digital ECG systems (GE Medical Systems, Milwaukee, Wisconsin) as previously described (8,9,11). Absolute ST-segment deviation was measured by computer with 5- V precision at the midpoint of the ST-segment. ST-segment depression 50 V in any lead (excluding avr) was considered abnormal, a value that corresponded to the 95th percentile of ECG STD and that stratified mortality risk in participants in the first Strong Heart Study examination (8). CRP determination. C-reactive protein was measured using an in-house developed enzyme-linked immunosorbent assay using purified CRP and anti-crp antibodies from CalBioChem (La Jolla, California) (31). This assay has been used extensively in epidemiologic studies and in the validation of the commercially available assay for high-sensitivity CRP (32). The coefficient of variation is approximately 8%. C-reactive protein was considered abnormal if 7.0 mg/l, defining the upper quartile of CRP levels in the study population. Determination of end points. Deaths were identified and verified and were classified as due to CVD if caused by myocardial infarction, stroke, sudden death from coronary heart disease, or congestive heart failure as determined by standardized review by the mortality committee (8,29). Data and statistical analyses. Data were analyzed with SPSS, release 12.0 (SPSS Inc., Chicago, Illinois). Data are presented as mean SD for continuous variables and as proportions for categorical variables. Mean values were compared between groups using two-way analysis of variance to adjust for possible gender differences. Proportions were compared by chi-square tests. Mortality rates were calculated and plotted by the Kaplan-Meier product-limit method; death rates were compared between groups with the log-rank test. Mortality analyses were performed by fitting Cox proportional hazards models to the data with stratification by center. The estimated hazard ratio of death associated with positive test outcomes was computed as the antilog of the estimated coefficient for dichotomous variables. The 95% CI of each relative risk was calculated from the estimated coefficients and their standard errors, and Wald chi-square statistics and probability values were calculated. To test the independence of STD and CRP as predictors of mortality, both variables were entered together into multivariate Cox models with a forward selection procedure. The models also included covariates that were significant predictors of CVD or all-cause mortality found in the univariate Cox models. To test the complementary information provided by CRP and STD criteria, a combined test criterion was derived that incorporated both measures into three categories: both STD and CRP negative, either STD or CRP positive, and both STD and CRP positive. For all tests, a two-tailed p value 0.05 was considered significant. RESULTS Patient characteristics. After a mean follow-up of years there were 310 deaths from all causes, including 95 CVD deaths. Clinical characteristics of participants grouped Table 1. Clinical Characteristics According to Level of C-Reactive Protein and Degree of ST-Segment Depression at Baseline Variable CRP <7.0 mg/l (n 1,615) CRP >7.0 mg/l (n 540) p Value STD <50 V (n 2,028) STD >50 V (n 127) p Value Age (yrs) Gender (% female) Body mass index (kg/m 2 ) Systolic BP (mm Hg) Diastolic BP (mm Hg) HDL cholesterol (mg/dl) LDL cholesterol (mg/dl) Triglycerides (mg/dl) Albuminuria (log mg/g) Fasting glucose (mg/dl) Prevalent CHD (%) Smoking (% current) Diabetes (%) Fibrinogen (mg/dl) C-reactive protein (mg/l) STD ( V) BP blood pressure; CHD coronary heart disease; CRP C-reactive protein; HDL high-density lipoprotein; LDL low-density lipoprotein; STD ST-segment depression.

3 JACC Vol. 45, No. 11, 2005 June 7, 2005: Okin et al. C-Reactive Protein and ST-Segment Depression 1789 Table 2. Cox Proportional Hazards Models of High-Sensitivity C-Reactive Protein and ST-Segment Depression Criteria for Prediction of All-Cause and Cardiovascular Mortality All-Cause Mortality Cardiovascular Mortality Variable Hazard Ratio 95% CI Chi-Square p Value Hazard Ratio 95% CI Chi-Square p Value Univariate STD 50 V CRP 7.0 mg/l Multivariate* STD 50 V CRP 7.0 mg/l *Including both ST-segment depression 50 V and hscrp 7.0 mg/l in the model. Age, body mass index, LDL cholesterol, fasting glucose, history of smoking, and albuminuria also remain in the multivariate Cox model. Age, gender, and albuminuria also remained in the multivariate Cox model. CI confidence interval; other abbreviations as in Table 1. according to the level of CRP and degree of STD are examined in Table 1. Compared to those with lower CRP levels, the 540 participants with CRP 7.0 mg/l were slightly younger, more likely to be women, had higher body mass indexes, lower LDL cholesterol, higher fasting glucose and fibrinogen levels, greater albuminuria, higher prevalences of coronary heart disease and diabetes, and were slightly less likely to be current smokers, but did not differ with respect to systolic or diastolic pressure, high-density lipoprotein (HDL) cholesterol or triglyceride levels, or degree of STD. The 127 participants with STD 50 V had higher systolic pressures, higher triglyceride, fasting glucose, fibrinogen, and CRP levels; had greater albuminuria and prevalences of coronary heart disease and diabetes; and were less likely to be current smokers, but did not differ with respect to age, gender, body mass index, diastolic pressure, and HDL or LDL cholesterol levels compared to participants with less STD. C-reactive protein and STD were only minimally correlated with each other (r 0.06, p 0.006). Prediction of cardiovascular and all-cause mortality. In Cox analyses stratified for study center, abnormal STD and CRP were each significant predictors of CVD and all-cause mortality (Table 2, Figs. 1 and 2). ST-segment depression 50 V was associated with a 4-fold increased risk of CVD death and with a 2.5-fold increased risk of all-cause mortality, with CVD mortality of 15.0% (19 of 127) and all-cause mortality of 32.3% (41 of 127), compared with 3.7% (76 of 2,028) and 13.3% (269 of 2,028), respectively, in those with STD 50 V. C-reactive protein 7.0 mg/l was associated with a two-fold increased risk of CVD death and with a nearly two-fold higher risk of death from any cause, with CVD mortality of 6.7% (35 of 540) and total mortality of 20.0% (128 of 540), compared with 3.7% (59 of 1,615) and 12.5% (202 of 1,615), respectively, in those with lower CRP levels. After adjustment for the possible predictive values of age, gender, body mass index, diastolic and systolic blood pressures, HDL and LDL cholesterol, triglyceride, fasting glucose and fibrinogen levels, albuminuria, alcohol use, diabetes, and history of smoking or prevalent coronary heart disease, both STD 50 V and an elevated CRP remained significant predictors of CVD and all-cause mortality (Table 2). Predictive values of STD and CRP for CVD and all-cause mortality were not dependent on the partition values selected for test positivity: considering STD and CRP as continuous variables preserved their independent predictive value for both CVD (HR 1.25, 95% CI 1.03 to 1.52 per 25 V of STD and HR 1.015, 95% CI Figure 1. Kaplan-Meier plots of cumulative cardiovascular mortality (A) and all-cause mortality (B) according to the magnitude of ST-segment depression (STD) partitioned at 50 V.

4 1790 Okin et al. JACC Vol. 45, No. 11, 2005 C-Reactive Protein and ST-Segment Depression June 7, 2005: Combined high-sensitivity CRP and ECG STD criteria. Because CRP and STD criteria provided independent prognostic information, the ability of the combination of these variables to improve prediction of mortality was assessed (Table 3, Fig. 3). In Cox analyses stratified by study center, the combined CRP and ECG STD variable improved risk stratification compared to either CRP or STD alone for both CVD and all-cause mortality, with the presence of both ECG STD and an elevated CRP associated with the greatest risks. Cardiovascular disease mortality was 16.7% (6 of 36) in participants with both STD 50 V and CRP 7.0 mg/l, 7.2% (43 of 595) in those with either ECG STD or an abnormal CRP, and only 3.0% (46 of 1,524) in those in whom both variables were negative (p ). All-cause mortality was 55.6% (20 of 36) in the presence of both ECG STD and an elevated CRP, 18.3% (109 of 595) when either variable was abnormal, and 11.9% (181 of 1,524) when both tests were negative (p ). Multivariate Cox analyses (Table 3) demonstrated that after adjustment for other potential predictors of mortality, the combination of ECG STD and an elevated CRP level remained a significant predictor of CVD and all-cause mortality, with the presence of both ECG STD and increased CRP associated with a 3.2-fold increased risk of CVD death and a 3.9-fold increased risk of all-cause mortality after adjusting for covariates. Predictive value of the combined criterion was not dependent on use of the upper quartile of CRP to define abnormality: the combined predictive value of CRP and STD remained significant if CRP was partitioned using the population median (3.8 mg/l) or 90th percentile (15.0 mg/l) values. Figure 2. Kaplan-Meier plots of cumulative cardiovascular mortality (A) and all-cause mortality (B) according to high-sensitivity C-reactive protein (CRP) partitioned at 7.0 mg/l. to per 1 mg/l of CRP) and all-cause mortality (HR 1.13, 95% CI 1.03 to 1.25 and HR 1.014, 95% CI to 1.023) in multivariate analyses. DISCUSSION This study demonstrates that the combination of STD on the ECG and elevated high-sensitivity CRP is associated with an increased risk of CVD and all-cause mortality. Electrocardiographic STD and increased CRP provide ad- Table 3. Cox Proportional Hazards Models of Combined High-Sensitivity C-Reactive Protein and ST-Segment Depression Criteria for Prediction of All-Cause and Cardiovascular Mortality All-Cause Mortality Cardiovascular Mortality Hazard Hazard Variables Ratio 95% CI Chi-Square p Value Ratio 95% CI Chi-Square p Value Univariate Combined CRP STD CRP /STD (n 1,524) 1 1 CRP /STD or CRP /STD (n 595) CRP /STD (n 36) Multivariate* Combined CRP STD CRP /STD (n 1,524) 1 1 CRP /STD or CRP /STD (n 595) CRP /STD (n 36) *Same variables remain in the multivariate models as listed in Table 2. Abbreviations as in Table 2.

5 JACC Vol. 45, No. 11, 2005 June 7, 2005: Okin et al. C-Reactive Protein and ST-Segment Depression 1791 Figure 3. Kaplan-Meier plots of cumulative cardiovascular mortality (A) and all-cause mortality (B) according to combined ST-segment depression (STD) and high-sensitivity C-reactive protein (CRP) criteria. (STD / CRP represents both negative; STD or CRP represents either positive; STD /CRP represents both positive). ditive prognostic information, independent of each other and of other risk factors known to predict mortality. The additive prognostic value of increased CRP and STD persists across the full range of CRP and STD values and is independent of factors previously demonstrated to provide additional prognostic information in the presence of an elevated CRP (19 26). The absence of both an elevated CRP and ECG STD identifies a large group at relatively low five-year risk of death, whereas the presence of both abnormalities identifies a subgroup with a markedly increased five-year mortality. These findings support the ability of combining simple computerized ECG STD and CRP criteria to improve risk stratification. Relationship of CRP and STD to outcome. The separate predictive values of CRP and ECG STD have been well documented in a range of populations, with convincing evidence for continuous relationships between increasing values of both STD and CRP and event rates (8,9,11,13 26). Quantitative measures of STD using computerized ECG have been associated with an increased risk of anatomic hypertrophy (11) and with both CVD and all-cause mortality (8,9), with the combination of ECG STD and echocardiographic left ventricular hypertrophy providing complementary prognostic information for these outcomes (9). Similarly, the prognostic value of CRP has been extensively documented in a variety of prospective epidemiologic studies, with evolving evidence that CRP provides additive prognostic information beyond that afforded by assessment of cholesterol levels (19) and the Framingham risk score (24,25), further supporting the additive value of CRP as a risk factor for vascular disease and outcomes. However, only limited data exist on the relationship between CRP and ECG findings (27,28), and the value of combining CRP and ECG STD for risk prediction has previously not been examined. In a population-based study of 8,076 subjects, Asselbergs et al. (27) demonstrated that although ST-segment and T-wave abnormalities by Minnesota code were modest univariate correlates of an increased CRP, only Q-wave myocardial infarction by Minnesota code remained associated with increased CRP levels after adjusting for standard cardiovascular risk factors. The current study demonstrates only a weak correlation between CRP and ECG STD as continuous variables and that each provides significant independent prognostic information for both CVD and ACD (Table 2), providing the impetus for combining the variables to enhance risk stratification. Importantly, the present study further demonstrates that the additive predictive value of STD and CRP is independent of the prognostic value of cholesterol and other risk factors that constitute the Framingham risk score, and of serum fibrinogen levels and albuminuria, risk factors previously demonstrated to predict outcome in this population (9,30). In light of the recent report demonstrating the additive value of ECG STD and echocardiographic hypertrophy for predicting mortality in this population (9), it is important to note that inclusion of echocardiographic hypertrophy in alternative multivariate Cox models (data not shown) did not affect the predictive value of the combined ECG STD and CRP variable. These findings suggest that active inflammation (reflected by increased CRP) and preclinical CVD (as denoted by even minor degrees of STD on the ECG) provide complementary and additive stimuli in the development of mortality due to atherosclerosis. Accumulating evidence suggests that CRP may play a direct role in atherogenesis at the level of the endothelial cell and vascular smooth muscle (33 36), whereas STD on the ECG has been directly linked with CVD, including left ventricular hypertrophy (1 12). The interaction of inflammation with ventricular and vascular hypertrophy to produce atherosclerosis provides an attractive hypothesis for the additive impact of these two risk markers in predicting outcome. Several aspects of the study population need to be considered with respect to these findings. First, it is unclear to what degree these findings in American Indians can be extrapolated to other ethnic populations. However, the

6 1792 Okin et al. JACC Vol. 45, No. 11, 2005 C-Reactive Protein and ST-Segment Depression June 7, 2005: demonstrated predictive value of CRP and minor degrees of STD in other populations when examined separately from each other (1 7,13 26) suggests that the combination of CRP and ECG STD will stratify risk in other populations as well. Second, the current population has a high prevalence of diabetes and metabolic syndrome and is predominantly women, with the attendant issues of gender differences in CRP levels and possible effects of estrogen on CRP. However, CRP performed similarly in men and women and there were no significant interactions between gender and CRP in Cox analyses of the entire population. Additionally, estrogen use was limited in the population (n 137, 6.4%), and neither including estrogen use as an additional variable nor excluding women using estrogen altered the results. A number of other potential limitations should also be taken into account. First, the absence of serial CRP and ECG STD determinations precludes analysis of the impact of changes in CRP and/or STD on risk. Second, the values of CRP in this population of American Indians are significantly higher than those found in most prior populationbased studies of the prognostic value of CRP (13 26). However, the predictive value of both STD and CRP in the current population persisted when CRP was examined using both lower (3.8 mg/l) and higher (15.0 mg/l) partition values and when CRP was considered as a continuous variable. Previous analyses have also documented the predictive value of very high ( 10 mg/l) levels of CRP (25). In addition, exclusion of participants with CRP levels in the highest decile ( 15 mg/l) did not substantively alter the results of the current analyses. Lastly, it should be noted that use of computer-measured STD 50 V for prediction of risk has been validated in American Indian participants only in the Strong Heart Study (8,9,11). However, utility of this threshold of STD, using both manual and computerized ST-segment measurements, has been extensively examined and confirmed in other populations using Minnesota codes 4-2 and 4-3 (37,38), which are defined by the presence of STD 50 V and 100 V. The major implication of this study is that both highsensitivity CRP and ECG STD aid in routine clinical identification of patients at high risk. The recent interest in development of novel biomarkers that provide additional information beyond that available from standard risk factors and are inexpensive and readily available to practitioners (39) provides further impetus for the use of CRP levels and determination of the magnitude of STD from the widely available and inexpensive digital ECG. Although data are emerging that demonstrate clear associations between improved clinical outcomes and treatment-related reductions in CRP (40,41) and ECG markers of risk (42), whether therapies targeted specifically at reducing these biomarkers will improve prognosis requires further evaluation and will prove crucial in further delineating the role of these biomarkers in the serial assessment of risk. Acknowledgments The authors wish to thank the Strong Heart Study participants, staff, and coordinators. Reprint requests and correspondence: Dr. Peter M. Okin, Weill Medical College of Cornell University, 525 East 68th Street, New York, New York pokin@med.cornell.edu. REFERENCES 1. Liao Y, Liu K, Dyer A, et al. Sex differential in the relationship of electrocardiographic ST-T abnormalities to risk of coronary death: 11.5 year follow-up findings of the Chicago Heart Association Detection Project in Industry. Circulation 1987;75: Liao Y, Liu K, Dyer A, et al. Major and minor electrocardiographic abnormalities and risk of death from coronary heart disease, cardiovascular diseases and all causes in men and women. J Am Coll Cardiol 1988;12: De Bacquer D, Martins Pereira LS, De Backer G, et al. The predictive value of electrocardiographic abnormalities for total and cardiovascular disease mortality in men and women. Eur Heart J 1994;15: Dekker JM, Schouten EG, Klootwijk P, et al. ST-segment and T-wave characteristics as indicators of coronary heart disease risk: the Zutphen Study. J Am Coll Cardiol 1995;25: Sigurdsson E, Sigfusson N, Sigvaldason H, et al. Silent ST-T changes in an epidemiologic cohort study-a marker of hypertension or coronary heart disease or both: the Reykjavik Study. J Am Coll Cardiol 1996;27: Daviglus ML, Liao Y, Greenland P, et al. Association of nonspecific minor ST-T abnormalities with cardiovascular mortality: the Chicago Western Electric Study. JAMA 1999;281: Okin PM, Devereux RB, Lee ET, Galloway JM, Howard BV. Ventricular repolarization complexity and abnormality predict allcause and cardiovascular mortality in diabetes: the Strong Heart Study. Diabetes 2004;53: Okin PM, Devereux RB, Kors JA, et al. Computerized ST depression analysis improves prediction of all-cause and cardiovascular mortality: the Strong Heart Study. Ann Noninvas Electrocardiol 2001;6: Okin PM, Roman MJ, Lee ET, Galloway JM, Howard BV, Devereux RB. Combined echocardiographic left ventricular hypertrophy and electrocardiographic ST depression improve prediction of mortality in American Indians: the Strong Heart Study. Hypertension 2004; 43: Okin PM, Devereux RB, Nieminen MS, et al. Electrocardiographic strain pattern and prediction of cardiovascular morbidity and mortality in hypertensive patients. Hypertension 2004;44: Okin PM, Devereux RB, Fabsitz RR, et al. Quantitative assessment of electrocardiographic strain predicts increased left ventricular mass: the Strong Heart Study. J Am Coll Cardiol 2002;40: Okin PM, Devereux RB, Nieminen MS, et al. Relationship of the electrocardiographic strain pattern to left ventricular structure and function in hypertensive patients: the LIFE Study. J Am Coll Cardiol 2001;38: Ridker PM, Cushman M, Stampfer MJ, et al. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med 1997;336: Danesh J, Whincup P, Walker M, et al. Low grade inflammation and coronary heart disease: prospective study and updated meta-analyses. BMJ 2000;321: 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: Mendall MA, Strachan DP, Butland BK, et al. C-reactive protein: relation to total mortality, cardiovascular mortality and cardiovascular risk factors in men. Eur Heart J 2000;21: Sesso HD, Buring JE, Rifai N, et al. C-reactive protein and the risk of developing hypertension. JAMA 2003;290: Danesh J, Wheeler JG, Hirschfield GM, et al. C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med 2004;350: Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the

7 JACC Vol. 45, No. 11, 2005 June 7, 2005: Okin et al. C-Reactive Protein and ST-Segment Depression 1793 prediction of first cardiovascular events. N Engl J Med 2002;347: Schillinger M, Exner M, Amighi J, et al. Joint effects of C-reactive protein and glycated hemoglobin in predicting future cardiovascular events of patients with advanced atherosclerosis. Circulation 2003;108: Ridker PM, Buring JE, Cook NR, Rifai N. C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: an 8-year follow-up of initially healthy American women. Circulation 2003;107: Sattar N, Gaw A, Scherbakova O, et al. Metabolic syndrome with and without C-reactive protein as a predictor of coronary heart disease and diabetes in the West of Scotland Coronary Prevention Study. Circulation 2003;108: Blake GJ, Rifai N, Buring JE, Ridker PM. Blood pressure, C-reactive protein, and risk of future cardiovascular events. Circulation 2003;108: Koenig W, Löwel H, Baumert J, Meisinger C. C-reactive protein modulates risk prediction based on the Framingham score. Circulation 2004;109: 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: Pearson TA, Mensah GA, Alexander RW, 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: Asselbergs FW, van Boven AJ, Stuveling EM, et al. Relation of electrocardiographic abnormalities to levels of serum C-reactive protein. Am J Cardiol 2003;91: Cosín-Sales J, Pizzi C, Brown S, Kaski JC. C-reactive protein, clinical presentation, and ischemic activity in patients with chest pain and normal coronary angiograms. J Am Coll Cardiol 2003;41: Lee ET, Welty TK, Fabsitz R, et al. The Strong Heart Study: a study of cardiovascular disease in American Indians: design and methods. Am J Epidemiol 1990;132: Palmieri V, Celentano A, Roman MJ, et al. Relation of fibrinogen to cardiovascular events is independent of preclinical cardiovascular disease: the Strong Heart Study. Am Heart J 2003;145: Macy E, Hayes T, Tracy R. Variability in the measurement of C-reactive protein in healthy subjects: implications for reference interval and epidemiological applications. Clin Chem 1997;43: Rifai N, Tracy RP, Ridker PM. Clinical efficacy of an automated high-sensitivity C-reactive protein assay. Clin Chem 1999;45: Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation 2002;105: Szmitko PE, Wang CH, Weisel RD, et al. New markers of inflammation and endothelial cell activation: part 1. Circulation 2003;108: Wang CH, Li SH, Weisel RD, et al. C-reactive protein upregulates angiotensin type 1 receptors in vascular smooth muscle. Circulation 2003;107: Verma S, Kuliszewski MA, Li S-H, et al. C-reactive protein attenuates endothelial progenitor cell survival, differentiation and function: further evidence of a mechanistic link between C-reactive protein and cardiovascular disease. Circulation 2004;109: Diercks GFH, Hillege HL, van Boven AJ, et al. Microalbuminuria modifies the mortality risk associated with electrocardiographic ST-T segment changes. J Am Coll Cardiol 2002;40: Kors JA, Crow RS, Hannan PJ, Rautaharju PM, Folsom AR. Comparison of computer-assigned Minnesota codes with the visual standard method for new coronary heart disease events. Am J Epidemiol 2000;151: Ridker PM, Wilson PWF, Grundy SM. Should C-reactive protein be added to metabolic syndrome and to assessment of global cardiovascular risk? Circulation 2004;109: Ridker PM, Rifai N, Clearfield M, et al., for the Air Force/Texas Coronary Atherosclerosis Prevention Study Investigators. Measurement of C-reactive protein for targeting of statin therapy in the primary prevention of acute coronary events. N Engl J Med 2001;344: Ridker PM, Cannon CP, Morrow D, et al. C-reactive protein levels and outcomes after statin therapy. N Engl J Med 2005;352: Okin PM, Devereux RB, Jern S, et al. Regression of electrocardiographic left ventricular hypertrophy during antihypertensive treatment and prediction of major cardiovascular events: the LIFE Study. JAMA 2004;292: