Chemokines and Incident Coronary Heart Disease

【摘要】  Objectives- The chemokines monocyte chemoattractant protein-1 (MCP-1/CCL2), interleukin-8 (IL-8/CXCL8), and interferon- -inducible protein-10 (IP-10/CXCL10) have been reported to be involved in the development of atherosclerosis and type 2 diabetes. The aim of this study was to assess whether elevated systemic levels of these chemokines precede coronary events.

Methods and Results- We investigated MCP-1, IL-8, and IP-10 serum levels in a case-cohort design based on data from 381 individuals (294 men, 87 women) with and 1977 individuals (1006 men, 971 women) without incident coronary heart disease (CHD) from the prospective, population-based MONICA/KORA Augsburg study (1984 to 2002). The mean follow-up time was 11.0 years. Baseline concentrations were significantly higher in cases compared with noncases ( P 0.001 for all chemokines). MCP-1 and IL-8 remained associated with CHD risk after adjustment for age, sex, and survey with hazard ratios (95% confidence intervals) comparing extreme tertiles of 1.39 (1.05 to 1.84) for MCP-1 and 1.48 (1.10 to 1.99) for IL-8. However, adjustment for further cardiovascular and immunologic risk factors attenuated the observed associations, and they became nonsignificant.

Conclusions- Elevated systemic levels of the chemokines MCP-1, IL-8, and IP-10 precede CHD but do not represent independent risk factors. Thus, the associations are less pronounced than previously shown for type 2 diabetes.

The study investigates whether elevated serum concentrations of the chemokines MCP-1, IL-8, and IP-10 precede coronary events in a prospective case-cohort design. Baseline concentrations of all 3 chemokines were significantly higher in cases compared with noncases, but the associations with CHD risk were no longer statistically significant in multivariable analysis.

【关键词】  chemokines inflammation coronary heart disease casecohort study prognosis

Introduction

Inflammatory mechanisms play an important role in the initiation and progression of cardiovascular diseases. 1 Key mediators in these processes are chemokines, a family of low-molecular-weight proteins regulating cell migration to sites of inflammation and triggering activation of many cell types. 2 They have chemotactic properties, and the repertoire of chemokine receptors on the cell surface determines the migration behavior of leukocytes and other cells. Chemokines like monocyte chemoattractant protein-1 (MCP-1/CCL2), interleukin-8 (IL-8/CXCL8), and INF- -inducible protein-10 (IP-10/CXCL10) are expressed in atherosclerotic lesions. 3-5 Their release by endothelial cells, smooth muscle cells, macrophages, and perivascular adipose tissue has been shown to mediate leukocyte adhesion to endothelial cells as well as infiltration into the arterial wall 6 and presumably also into epicardial fat, with implications for the development of coronary heart disease (CHD). 7,8

There is further evidence from preclinical and clinical studies for a contribution of chemokines to CHD. On the genetic level, deletion of MCP-1/CCL2 or its receptor CCR2 strongly attenuate macrophage recruitment and lesion formation in atherosclerosis-prone mice. 9,10 In humans, gene variants of MCP-1 and CCR2 have also been reported to increase the susceptibility to CHD in some, 11,12 but not all, studies. 13 Cross-sectional studies investigating circulating chemokine levels suggest that these processes might also be reflected on the systemic level. MCP-1 and IL-8 concentrations have been reported to be associated with traditional cardiovascular risk factors, atherosclerosis, and prevalent CHD. 14-16 In a previous study on stable coronary artery disease, we found significantly higher levels of IL-8 and IP-10 and lower levels of RANTES ( R egulated on A ctivation, N ormal T -cell E xpressed and S ecreted)/CCL5 in patients compared with controls, whereas no clear disease association could be found for MCP-1, macrophage inflammatory protein-1 (MIP-1 )/CCL3 and eotaxin/CCL11, indicating a differential upregulation of chemokines in cardiovascular disease. 16 Two studies reported that high MCP-1 levels may be associated with incident CHD events, 17,18 and IL-8 was found to predict cardiovascular events in a small study in hemodialysis patients. 19

Because there are only a few prospective studies that have investigated whether elevated chemokine levels precede coronary events, the question of whether chemokines may represent novel independent CHD risk factors is still open. Therefore, we measured circulating levels of MCP-1, IP-10, and IL-8 in sera from a large, prospective, population-based case-cohort study to address the following questions: (1) Do elevated levels of these 3 chemokines precede CHD end points? (2) Are associations with CHD risk independent of traditional cardiovascular risk factors? (3) If not, which of the established risk factors may mediate an increase in circulating chemokine levels? (4) Does the combination of all 3 chemokines in 1 model help to identify individuals at increased CHD risk?

Methods

Study Design

The design of the prospective case-cohort study 20 within the population-based MONICA/KORA Augsburg cohort study, 1984 to 2002, has been described previously. 21,22 The outcome CHD variable was a combined end point of incident fatal and nonfatal myocardial infarction (MI) and sudden cardiac death (SCD) before the age of 75 years. Cases were identified through the MONICA/KORA Augsburg coronary event registry and through follow-up questionnaires for persons who had moved out of the study area. Until December 2000, the diagnosis of a major nonfatal MI event was based on the MONICA algorithm taking into account symptoms, cardiac enzymes, and electrocardiogram changes. Since January 1, 2001, all patients with MI diagnosed according to European Society of Cardiology and American College of Cardiology criteria were included. 23,24 Deaths from MI were validated by autopsy reports, death certificates, chart review, and information from the last treating physician. The present analysis comprised a total of 2358 participants (294 men and 87 women with incident CHD and 1006 men and 971 women without incident CHD; Figure I in the online data supplement, available at http://atvb.ahajournals.org). The mean follow-up time (±SD) for the study population was 11.0 (±5.0) years. For further information on the study design, please see the online data supplement.

Data Collection, Laboratory Measurements, and Statistical Analyses

Most aspects of data collection and laboratory measurements have been described before. 22,25,26 For additional details and information on statistical analyses, please see the online data supplement.

Results

Study Population

Baseline characteristics of the study participants (n=2358) are given in supplemental Table I. Briefly, individuals who had a coronary event during the follow-up period (cases) differed significantly from subjects without coronary event (noncases) by higher age, body mass index (BMI), waist-to-hip ratio (WHR), prevalence of hypertension or diabetes mellitus, levels of total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C), and TC/high-density lipoprotein (HDL)-C ratio, as well as lower levels of HDL-C, more frequent parental history of MI, higher frequency of smokers, and less physical activity. In cases, baseline systemic concentrations of the chemokines MCP-1, IL-8, and IP-10 and of C-reactive protein (CRP) and IL-6 were significantly elevated compared with noncases ( P =0.001 for IP-10 and P <0.001 for the other mediators).

In the randomly drawn subcohort, we observed significantly higher concentrations of MCP-1 and IL-8, but not of IP-10, in men compared with women, with geometric means of 190.4 and 170.7 pg/mL, respectively, for MCP-1 ( P =0.004); 7.5 and 6.9 pg/mL, respectively, for IL-8 ( P =0.003); and 213.9 and 222.6 pg/mL, respectively, for IP-10 ( P =0.259). In general, we observed modest positive, but significant, correlations between chemokine levels and age, BMI, WHR, and blood pressure, whereas only IP-10 concentrations were correlated with HDL-C or the ratio TC/HDL-C ( Table 1 ). Associations were stronger between the immunologic parameters: MCP-1, IL-8, and IP-10 were highly correlated with each other and with IL-18 ( r 0.2, P <0.001 for each correlation). Moderate correlations were observed between chemokines and CRP and IL-6. The strength of these associations were similar in men and in women.

TABLE 1. Correlation of Chemokines With Age, Markers of Inflammation, and Other Continuous Risk Factors for CHD in the Randomly Sampled Subcohort: MONICA/KORA Case-Cohort Study, 1984 to 2002

Systemic Levels of MCP-1, IL-8, and IP-10 and Incident CHD

Using Cox proportional hazards analysis, we found that elevated serum concentrations of MCP-1, IL-8, and IP-10 were significantly associated with increased CHD risk during follow-up ( Table 2, model 1). Adjustment for age, sex, and survey (model 2) attenuated the associations between all 3 chemokines and incident CHD, resulting in lower hazard ratios (HRs) for all mediators and in loss of significance for IP-10. This attenuation was also seen in analyses stratified for sex (data not shown). Adjustment for clinical, metabolic and lifestyle risk factors of CHD (model 3) and additionally for CRP and IL-6 (model 4) had a similar effect on the associations of MCP-1 and IL-8 and incident CHD, leading to additional slight reductions of the observed HRs. Comparisons of extreme tertiles resulted in HRs of 1.30 to 1.38 and borderline nonsignificance ( P between 0.053 and 0.099) for MCP-1 and IL-8 in models 3 and 4.

TABLE 2. HRs (95% CI) for the Risk of Incident CHD According to Baseline Levels of MCP-1, IL-8, and IP-10

To assess which factors had the strongest impact on the association between elevated chemokine levels and CHD risk, separate models were performed adjusting for all single covariables. We found that age was the strongest single confounder for all 3 chemokines because adjustment for age alone led to more pronounced reduction of CHD risk than adjustment for any other single factor (age-adjusted HRs for the comparison of extreme tertiles: 1.39 [1.05 to 1.82] for MCP-1; 1.44 [1.09 to 1.90] for IL-8; 1.15 [0.88 to 1.50] for IP-10).

To test for a joint effect of increased levels of MCP-1, IL-8, and IP-10 on CHD risk, we combined these 3 chemokines in 1 model and investigated whether HRs increased with increasing number of elevated chemokine levels in the study subjects (elevated levels as defined by chemokine concentrations above median level measured in the cohort random sample). As shown in the Figure, elevated chemokine levels were associated with CHD risk in the unadjusted model ( P for trend <0.001) and in the model adjusted for age, sex, survey ( P for trend=0.007), whereas additional adjustment for further metabolic and immunologic risk factors led to borderline nonsignificant results. This analysis thus confirmed that chemokines are associated with CHD risk and that established cardiovascular risk factors may contribute to these associations.

Combined effect of elevated serum concentrations of MCP-1, IL-8, and IP-10 on CHD risk. Variables were stratified based on weighted median cut-off points for each mediator using the distribution in the randomly drawn subcohort. The numbers along the x-axis give the number of immune mediators in 1 individual whose concentrations are above the median cut-off point in the subcohort for the respective protein. Unadjusted and adjusted HRs (95% CI) were obtained as described for models 1 to 4 in Table 2. * P <0.05, *** P <0.001 compared with the reference group (all immune mediator concentrations below or equal to median). Probability values given in the figure are for trend.

Stratification for sex showed stronger associations between baseline chemokine levels and incident CHD in women than in men in unadjusted and adjusted analyses, but there was no significant interaction (probability values for interaction were 0.120, 0.580, and 0.155 for MCP-1, IL-8, and IP-10, respectively). To exclude potentially confounding effects by diabetes mellitus, all analyses were repeated without subjects with prevalent diabetes at baseline, but this had virtually no effect on HRs. Because the definition of MI was changed during follow-up, additional analyses were carried out for all cases diagnosed until December 2000 according to the MONICA algorithm based on typical symptoms, elevated cardiac enzymes or electrocardiogram changes (330 cases, 2028 noncases), whereas cases in 2001 and 2002 also included less severe MI diagnosed as resulting from elevated troponin levels. Comparisons of extreme tertiles in model 3 resulted in HRs of 1.24 (0.90 to 1.69) for MCP-1, 1.58 (1.12 to 2.24) for IL-8, and 1.29 (0.94 to 1.76) for IP-10 and thus indicated a slightly attenuated association for MCP-1 and a strengthened association for IL-8, whereas the association with IP-10 was not affected.

Discussion

The major findings of this study were as follows: (1) Elevated serum levels of the chemokines MCP-1, IL-8, and IP-10 precede CHD, which we defined as a composite end point of fatal and nonfatal MI and SCD. (2) However, these associations were not independent of established cardiovascular risk factors, so that MCP-1, IL-8, and IP-10 do not appear to be novel CHD risk factors. (3) We found significant positive correlations between chemokine concentrations and age, and adjustment for age only attenuated HRs more than adjustment for any other covariable. (4) In a combined model of all 3 chemokines, there was a trend toward higher CHD risk in individuals with elevations of all 3 markers, but again no significant association after adjustment for known cardiovascular risk factors.

These findings extend results from previous studies that mainly focused on MCP-1 and CHD, 17,18 because, to the best of our knowledge, prospective data on IL-8, IP-10, and CHD risk from population-based studies have not yet been reported. In addition, this study comprises the analysis of 3 functionally different chemokines, which also enabled us to investigate whether a combination of elevated chemokine levels may be more strongly associated with incident CHD than isolated immune markers.

The fact that individuals who subsequently developed CHD exhibited significantly higher serum levels of MCP-1, IL-8, and IP-10 than noncases is in line with clinical and preclinical studies that have suggested an implication of these mediators in atheroma formation and MI risk, 3-5,9,10,14-16 although our study cannot prove a causal relationship between elevated chemokine levels and CHD. Regarding the question of whether higher MCP-1 may precede coronary events independent of other risk factors, our data appear to contradict 2 previous reports 17,18 for coronary events. In 1 of these studies, discrepancies may be attributed to the selection of patients with acute coronary symptoms as study participants. 17 Such patients may have a general disease-related immune activation, which might not be comparable to unstimulated immune marker concentrations in a prospective, population-based study such as ours. In addition, it should be noted that the HR (95% confidence interval ) in the study by de Lemos et al was 1.42 (1.02 to 1.98) comparing the fourth quartile with the first 3 quartiles (adjustment for age; history of diabetes, hypertension, hypercholesterolemia, or prior CHD; creatinine clearance; ST deviation; cardiac troponin I; n=2127 patients), 17 whereas we found an HR (95% CI) comparing extreme MCP-1 tertiles of 1.32 (0.98 to 1.78) after adjustment for age, sex, survey, BMI, systolic blood pressure, ratio of TC/HDL-C, physical activity, alcohol intake, smoking status, parental history of MI, and history of diabetes. Hence, the point estimate was only marginally lower, and it cannot be excluded that differences in baseline populations and the availability of potentially confounding risk factors for the statistical adjustment might have contributed to this difference.

The second report describes the association of MCP-1 and incident CHD in the Atherosclerosis Risk in Communities (ARIC) study in 412 cases and 709 noncases. 18 Data appear ambiguous because incident CHD risk increased significantly per 1 SD difference in MCP-1 concentrations in multivariable-adjusted models, but not after adjustment for only age, sex, and race. In addition, comparing extreme tertiles of MCP-1 concentrations yielded a nonsignificant HR (95% CI) of 1.25 (0.79 to 1.97) after adjustment for age, sex, race, BMI, hypertension, diabetes, TC, HDL-C, and cigarette years for smokers. 18 Again, the point estimate is rather similar to our findings in a comparable multivariable model.

Taken together, these data show that the postulated pathophysiological role of the investigated chemokines might be reflected on the systemic level but that traditional cardiovascular risk factors may mediate the increase in systemic concentrations. In the MONICA/KORA sample, we could demonstrate that age was the dominant covariable that mediated the association. This finding is in line with other observations that suggest an intricate association among ageing, proinflammatory state, and cardiovascular risk factors. 27-29 Whereas we found that age was a confounder in the association between chemokine levels and CHD risk, cardiovascular risk factors and morbidity have been reported to represent confounders in the positive association between age and circulating immune mediators. 29 The mechanism of this age-related immune activation is still poorly understood, but it can be assumed that impaired immune regulation, presence of metabolic risk factors, and lifestyle factors like lack of physical activity have their share in this association. Therefore, although elevations of serum concentrations of MCP-1, IL-8, and IP-10 precede CHD events, they might not represent novel independent cardiovascular disease risk factors and are thus unlikely to significantly improve disease prediction by established cardiovascular risk factors. Even if the temporal relationship were highly significant after adjustment, this would not suffice to establish a causal relationship. However, our findings would have to be assessed on the background of other studies demonstrating expression of MCP-1, IL-8, and IP-10 in atherosclerotic lesions, 3-5 so that the lack of a robust statistical association would not preclude that local chemokine expression in coronary arteries and atheromas may be important in disease initiation and progression. In this localized process, chemokines may very well be "independent" pathogenetic factors, but further studies are warranted to investigate whether or which chemokines represent potential therapeutic targets for the treatment of atherosclerosis.

Given that a common proinflammatory background has been postulated for both cardiovascular disease and type 2 diabetes (T2D), 30 it is interesting that the association of MCP-1 and of the combination of elevated levels of MCP-1, IL-8, and IP-10 remained significantly associated with incident T2D in our MONICA/KORA case-cohort sample, even after comparable multivariable adjustment. 22 Immune activation leading to systemically elevated chemokine concentrations may thus be more relevant in the development of T2D, but further research is necessary to define the proinflammatory profile, which would enable to discriminate between CHD or T2D risk.

Our study has several strengths and limitations that need to be considered. The strengths include the population-based, prospective design with a large number of initially healthy incident cases and appropriate noncases. The definition of a composite end point was meant to reduce problems attributable to survival bias. Finally, the extensive phenotype data on the study participants allowed us to consider a range of potential confounding variables and to adjust for them in stepwise regression analyses. Regarding limitations, it should be noted that the MONICA/KORA sample includes only men and women of German nationality with moderate absolute CHD risk, so that we could not investigate the importance of chemokines in populations with either very high or very low CHD risk or with other genetic/ethnic background. In addition, the assessment of systemic chemokine levels is still not standardized, which precludes their use in clinical routine. In our study, interassay coefficients of variation for IP-10 were relatively high, 22 which might have attenuated the association between IP-10 and incident CHD. The definition of MI was adapted to new guidelines in 2001 and thus did not remain constant during the follow-up period. Our sensitivity analyses indicated that definition and severity of MI might affect the association between systemic chemokine levels and CHD risk, but larger studies would be required to investigate this aspect in detail.

In summary, we demonstrated that elevated levels of the chemokines MCP-1, IL-8, and IP-10 precede coronary events. Comparing extreme tertiles of chemokine concentrations and adjustment for age, sex, BMI, systolic blood pressure, ratio of TC/HDL-C, physical activity, alcohol intake, smoking status, parental history of MI, history of diabetes, and survey resulted in HRs of 1.26 to 1.38 ( P between 0.044 and 0.105) for the risk of incident CHD with age being the major confounder in the association between chemokines and CHD risk. Thus, the contribution of chemokines to CHD risk appears to be only moderate and lower than to T2D risk, but the associations need to be studied in larger samples with different risk profiles and genetic background to give a more definite answer as to whether elevated chemokine levels contribute independently to the development of CHD.

Acknowledgments

We thank all members of the GSF Institute of Epidemiology who were involved in the planning and conduct of the MONICA/KORA Augsburg studies. We thank the MONICA/KORA myocardial infarction registry team in Augsburg and the MONICA Augsburg survey team. Furthermore, we are grateful to Prof L. Chambless (School of Public Health, University of North Carolina at Chapel Hill) for statistical advice concerning the analysis of the case-cohort dataset; to G. Gornitzka, U. Poschen, K. Röhrig (all from the German Diabetes Center, Düsseldorf, Germany), and G. Trischler (University of Ulm, Germany) for excellent technical assistance; and to W. de Jager (University Medical Center Utrecht/Immune Tolerance Network Luminex Core Facility, the Netherlands) for support with establishing the Luminex assay. Finally, we are indebted to all study participants.

Sources of Funding

The MONICA/KORA Augsburg cohort study was financed by the GSF-National Research Center for Environment and Health and supported by grants from the Federal Ministry of Education and Research, Berlin, Germany. The present study was primarily funded by the Deutsche Forschungsgemeinschaft (German Research Foundation; TH784/2-1) and the European Foundation for the Study of Diabetes. Additional support was obtained from the German Federal Ministry of Health and Social Security; the German Federal Ministry of Education, Science, Research and Technology/NGFN-2; the Ministry of Science and Research of the State North Rhine-Westphalia; and the Department of Internal Medicine II-Cardiology at the University of Ulm.

Disclosure(s)

None.

【参考文献】
  Libby P. Inflammation in atherosclerosis. Nature. 2002; 420: 868-874.

Gerard C, Rollins BJ. Chemokines and disease. Nat Immunol. 2001; 2: 108-115.

Nelken NA, Coughlin SR, Gordon D, Wilcox JN. Monocyte chemoattractant protein-1 in human atheromatous plaques. J Clin Invest. 1991; 88: 1121-1127.

Wang N, Tabas I, Winchester R, Ravalli S, Rabbani LE, Tall A. Interleukin 8 is induced by cholesterol loading of macrophages and expressed by macrophage foam cells in human atheroma. J Biol Chem. 1996; 271: 8837-8842.

Mach F, Sauty A, Iarossi AS, Sukhova GK, Neote K, Libby P, Luster AD. Differential expression of the T lymphocyte-activating CXC chemokines by human atheroma-associated cells. J Clin Invest. 1999; 104: 1041-1050.

Charo IF, Taubman MB. Chemokines in the pathogenesis of vascular disease. Circ Res. 2004; 95: 858-866.

Mazurek T, Zhang L, Zalewski A, Mannion JD, Diehl JT, Arafat H, Sarov-Blat L, O?Brien S, Keiper EA, Johnson A, Martin J, Goldstein BJ, Shi Y. Human epicardial adipose tissue is a source of inflammatory mediators. Circulation. 2003; 108: 2460-2466.

Henrichot E, Juge-Aubry CE, Pernin A, Pache JC, Velebit V, Dayer JM, Meda P, Chizzolini C, Meier CA. Production of chemokines by perivascular adipose tissue: a role in the pathogenesis of atherosclerosis? Arterioscler Thromb Vasc Biol. 2005; 25: 2594-2599.

Boring L, Gosling J, Cleary M, Charo IF. Decreased lesion formation in CCR2-/- mice reveals a role for chemokines in the initiation of atherosclerosis. Nature. 1998; 394: 894-897.

Gosling J, Slaymaker S, Gu L, Tseng S, Zlot CH, Young SG, Rollins BJ, Charo IF. MCP-1 deficiency reduces susceptibility to atherosclerosis in mice that overexpress human apolipoprotein B. J Clin Invest. 1999; 103: 773-778.

Ortlepp JR, Vesper K, Mevissen V, Schmitz F, Janssens U, Franke A, Hanrath P, Weber C, Zerres K, Hoffmann R. Chemokine receptor (CCR2) genotype is associated with myocardial infarction and heart failure in patients under 65 years of age. J Mol Med. 2003; 81: 363-367.

McDermott DH, Yang Q, Kathiresan S, Cupples A, Massaro JM, Keaney JF, Larson MG, Vasan RS, Hirschhorn JN, O?Donnell CJ, Murphy PM, Benjamin EJ. CCL2 polymorphisms are associated with serum monocyte chemoattractant protein-1 levels and myocardial infarction in the Framingham Heart Study. Circulation. 2005; 112: 1113-1120.

Bjarnadottir K, Eiriksdottir G, Aspelund T, Gudnason V. Examination of genetic effects of polymorphisms in the MCP-1 and CCR2 genes on MI in the Icelandic population. Atherosclerosis. In press.

Deo R, Khera A, McGuire DK, Murphy SA, Meo Neto J de P, Morrow DA, de Lemos JA. Association among plasma levels of monocyte chemoattractant protein-1, traditional cardiovascular risk factors, and subclinical atherosclerosis. J Am Coll Cardiol. 2004; 44: 1812-1818.

Larsson PT, Hallerstam S, Rosfors S, Wallen NH. Circulating markers of inflammation are related to carotid artery atherosclerosis. Int Angiol. 2005; 24: 43-51.

Rothenbacher D, Müller-Scholze S, Herder C, Koenig W, Kolb H. Differential expression of chemokines, risk of stable coronary heart disease and correlation with established cardiovascular risk markers. Arterioscler Thromb Vasc Biol. 2006; 26: 194-199.

de Lemos JA, Morrow DA, Sabatine MS, Murphy SA, Gibson CM, Antman EM, McCabe CH, Cannon CP, Braunwald E. Association between plasma levels of monocyte chemoattractant protein-1 and long-term clinical outcomes in patients with coronary artery syndromes. Circulation. 2003; 107: 690-695.

Hoogeveen RC, Morrison A, Boerwinkle E, Miles JS, Rhodes CE, Sharrett AR, Balantyne CM. Plasma MP-1 level and risk for peripheral arterial disease and incident coronary heart disease: Atherosclerosis Risk in Communities Study. Atherosclerosis. 2005; 183: 301-307.

Panichi V, Taccola D, Rizza GM, Consani C, Ghiadoni L, Filippi C, Cristofani R, Panicucci E, Migliori M, Sidoti A, Biagioli M, Boracelli D, Barsotti G, Tetta C. Interleukin-8 is a powerful prognostic predictor of all-cause and cardiovascular mortality in dialytic patients. Nephron Clin Pract. 2006; 102: c51-c58.

Barlow WE. Robust variance estimation for the case-cohort design. Biometrics. 1994; 50: 1064-1072.

Thorand B, Kolb H, Baumert J, Koenig W, Chambless L, Meisinger C, Illig T, Martin S, Herder C. Elevated levels of interleukin-18 predict the development of type 2 diabetes: results from the MONICA/KORA Augsburg Study, 1984-2002. Diabetes. 2005; 54: 2932-2938.

Herder C, Baumert J, Thorand B, Koenig W, de Jager W, Meisinger C, Illig T, Martin S, Kolb H. Chemokines as risk factors for type 2 diabetes-results from the MONICA/KORA Augsburg Study, 1984-2002. Diabetologia. 2006; 49: 921-929.

Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarction redefined-a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol. 2000; 36: 959-969.

Luepker RV, Apple FS, Christenson RH, Crow RS, Fortmann SP, Goff D, Goldberg RJ, Hand MM, Jaffe AS, Julian DG, Levy D, Manolio T, Mendis S, Mensah G, Pajak A, Prineas RJ, Reddy KS, Roger VL, Rosamond WD, Shahar E, Sharrett AR, Sorlie P, Tunstall-Pedoe H. AHA Council on Epidemiology and Prevention; AHA Statistics Committee; World Heart Federation Council on Epidemiology and Prevention; European Society of Cardiology Working Group on Epidemiology and Prevention; Centers for Disease Control and Prevention; National Heart, Lung, and Blood Institute. Case definitions for acute coronary heart disease in epidemiology and clinical research studies: a statement from the AHA Council on Epidemiology and Prevention; AHA Statistics Committee; World Heart Federation Council on Epidemiology and Prevention; the European Society of Cardiology Working Group on Epidemiology and Prevention; Centers for Disease Control and Prevention; and the National Heart, Lung, and Blood Institute. Circulation. 2003; 108: 2543-2549.

Meisinger C, Thorand B, Schneider A, Stieber J, Döring A, Löwel H. Sex differences in risk factors for incident type 2 diabetes mellitus-The MONICA Augsburg cohort study. Arch Intern Med. 2002; 162: 82-89.

Thorand B, Löwel H, Schneider A, Kolb H, Meisinger C, Fröhlich M, Koenig W. C-Reactive protein as a predictor for incident diabetes mellitus among middle-aged men: results from the MONICA Augsburg cohort study, 1984-1998. Arch Intern Med. 2003; 163: 93-99.

Inadera H, Egashira K, Takemoto M, Ouchi Y, Matsushima K. Increase in circulating levels of monocyte chemoattractant protein-1 with aging. J Interferon Cytokine Res. 1999; 19: 1179-1182.

Krabbe KS, Pedersen M, Bruunsgaard H. Inflammatory mediators in the elderly. Exp Gerontol. 2004; 39: 687-699.

Ferrucci L, Corsi A, Lauretani F, Bandinelli S, Bartali B, Taub DD, Guralnik JM, Longo DL. The origins of age-related proinflammatory state. Blood. 2005; 105: 2294-2299.

Pickup JC, Crook MA. Is type II diabetes a disease of the innate immune system? Diabetologia. 1998; 41: 1241-1248.

作者单位：Christian Herder; Jens Baumert; Barbara Thorand; Stephan Martin; Hannelore Löwel; Hubert Kolb; Wolfgang KoenigFrom the German Diabetes Clinic (C.H., S.M., H.K.), German Diabetes Center, Leibniz Center at Heinrich Heine University, Düsseldorf; GSF-National Research Center for Environment an