Skip Navigation


American Journal of Epidemiology Advance Access originally published online on July 19, 2007
American Journal of Epidemiology 2007 166(7):760-764; doi:10.1093/aje/kwm151
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
166/7/760    most recent
kwm151v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kivimäki, M.
Right arrow Articles by Raitakari, O. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kivimäki, M.
Right arrow Articles by Raitakari, O. T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology © The Author 2007. Published by the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.

ORIGINAL CONTRIBUTIONS

Variants in the CRP Gene as a Measure of Lifelong Differences in Average C-Reactive Protein Levels

The Cardiovascular Risk in Young Finns Study, 1980–2001

Mika Kivimäki1,2, Debbie A. Lawlor3, George Davey Smith3, Carita Eklund4, Mikko Hurme4,5, Terho Lehtimäki6, Jorma S. A. Viikari7 and Olli T. Raitakari8

1 Department of Epidemiology and Public Health, Faculty of Biomedical Sciences, University College London, London, United Kingdom
2 Finnish Institute of Occupational Health, Helsinki, Finland
3 Department of Social Medicine, Faculty of Medicine and Dentistry, University of Bristol, Bristol, United Kingdom
4 Department of Microbiology and Immunology, Faculty of Medicine, University of Tampere, Tampere, Finland
5 Laboratory Centre, Tampere University Hospital, Tampere, Finland
6 Department of Clinical Chemistry, Tampere University Hospital and University of Tampere, Tampere, Finland
7 Department of Medicine, Faculty of Medicine, University of Turku, Turku, Finland
8 Department of Clinical Physiology, Faculty of Medicine, University of Turku, Turku, Finland

Correspondence to Prof. Mika Kivimäki, Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, London WC1E 6BT, United Kingdom (e-mail: m.kivimaki{at}ucl.ac.uk).

Received for publication February 7, 2007. Accepted for publication April 23, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Genetic association studies have used variants in the C-reactive protein (CRP) gene to estimate causal effects of lifelong circulating CRP levels on disease endpoints. However, the extent to which the genetic variants are actually associated with lifelong circulating CRP levels has not been demonstrated empirically. In a population-based prospective cohort study (1980–2001) of 1,609 young Finns (768 men and 841 women), the authors genotyped five single nucleotide polymorphisms in the CRP gene (–717A/G, –286C/T/A, +1059G/C, +1444T/C, and +1846G/A) and assessed circulating CRP levels at ages 3–18 years and 24–39 years. The haplotypes from the five single nucleotide polymorphisms were associated with circulating CRP levels in childhood and adulthood, with the strongest effect being found for average CRP level across these two measures taken at two time points in the life course. In combination, the haplotype pairs accounted for 3.9%, 3.3%, and 5.0% of the variation in circulating CRP levels in childhood, in adulthood, and for the mean of CRP levels at both time points, respectively. These findings support the assumption that the above genetic variants define groups with long-term differences in circulating CRP levels.

C-reactive protein; haplotypes; random allocation; variation (genetics)

Abbreviations: CRP, C-reactive protein; SNP, single nucleotide polymorphism


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
There is debate as to whether C-reactive protein (CRP), a nonspecific marker of acute-phase inflammatory response, is a cause of coronary heart disease or just a marker of other causal factors and preclinical disease. Several genetic association studies have suggested that CRP may be noncausal, because variants in the CRP gene have not consistently been related to coronary outcomes (1–5). In the epidemiologic literature, this approach of studying causality using genetic variants is often called "Mendelian randomization" (6). The underlying assumption is that, because alleles are randomly allocated from parents to offspring at birth, persons with genetic variants that are related to higher CRP levels will have been, in effect, randomly allocated to somewhat higher levels of CRP across their life course than persons with genotypes related to lower CRP levels (6, 7). In theory, therefore, these lifelong differences in CRP level should translate into differences in coronary heart disease risk, if CRP were causally related to coronary heart disease and if population stratification did not confound associations of genotype with CRP levels or coronary heart disease risk (6–9).

A major limitation of genetic association evidence on CRP is that no previous studies are available to demonstrate empirically that the genetic variants are indeed a valid measure of lifelong differences in CRP levels. Previous studies have demonstrated relations of genotype with circulating CRP levels measured at one point in adulthood (most commonly in middle or older age) (1–5). Demonstrating an association with mean CRP levels across the life course would support the inference that the association of CRP genotype with coronary heart disease reflects lifetime differences in CRP levels (10, 11).

In this study, we determined the magnitude of the association between variants in the CRP gene and circulating CRP levels from childhood to adulthood in a population-based cohort of Finns. Since various nonstable environmental factors affect circulating CRP levels, we hypothesized that the genetic variants would be more robustly related to average CRP level across the life course than to CRP measured at a single point in time only.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study population
The participants were from the Cardiovascular Risk in Young Finns Study, an ongoing multicenter follow-up study of cardiovascular disease risk factors in Finnish children and adolescents (12, 13). The original sample comprised 4,320 White children and adolescents aged 3, 6, 9, 12, 15, and 18 years in five areas of Finland who had been randomly chosen from a national register (the Finnish Population Information System). The baseline examination was conducted in 1980, with the participation rate being 83 percent (3,596 of those invited). At the latest follow-up in 2001, the participants had reached 24–39 years of age. A total of 1,609 persons (768 men and 841 women) with full information on five genetic variants of the CRP gene and circulating CRP levels in 1980 and 2001 formed the cohort for the present study (5). Their CRP values were less than 10 mg/liter in 1980 and 2001; they were free of a history of recent infection and chronic rheumatic disease; and they did not include women who were lactating or pregnant in 2001.

This study was conducted according to the guidelines of the Declaration of Helsinki, and the study protocol was approved by local ethics committees. All participants gave their informed consent.

CRP genotyping
We genotyped five single nucleotide polymorphisms (SNPs) in the CRP gene (CRP–717A/G (rs2794521), CRP–286C/T/A (rs3091244), CRP+1059G/C (rs1800947), CRP+1444T/C (rs1130864), and CRP+1846G/A (rs1205)) using the ABI Prism 7900HT Sequence Detection System (Applied Biosystems, Foster City, California) for both polymerase chain reaction and allelic discrimination. For the SNP CRP+1059G/C, a commercial kit from Applied Biosystems was used (Assays-on-Demand; C_177490_10 CRP). The other SNPs were genotyped using Assays-by-Design from Applied Biosystems under standard conditions; the only exception was the triallelic tagSNP, which was genotyped as described previously (14), except for the genotype calling, which was conducted manually from the polymerase chain reaction run component tab.

Haplotypes were estimated from the five SNPs with the PHASE program, version 2.0.2 (www.stat.washington.edu/stephens), which uses a Bayesian statistical method for reconstructing haplotypes from population genotype data and lists the most probable haplotype pair for each individual (15). We used 100 iterations, one thinning interval, and 100 burn-ins—that is, settings that are compatible to the relatively small region analyzed (all studied SNPs were within 2,900 base pairs of each other), one of the most conserved regions in the human genome (16). Repeated independent runs varying the value of the seed of the pseudo-random number generator produced identical haplotype frequencies, supporting the consistency of the results. The SNPs were strongly linked, with D' values of 0.98–0.99; such close physical proximity and strong linkage disequilibrium further reduce uncertainly in the determination of the most-like haplotype pairs (17). There were no genotype data missing in our study sample. From subsequent analyses of haplotype and CRP levels, we excluded 41 persons with rare haplotypes (frequency <1 percent) and eight persons with ambiguous haplotype estimation.

Measurement of CRP levels
We assessed serum CRP levels in 1980 (childhood) and 2001 (adulthood) using an automated analyzer (Olympus AU400; Olympus America Inc., Center Valley, Pennsylvania) and a highly sensitive turbidimetric immunoassay kit (CRP-UL assay; Wako Chemicals, Neuss, Germany) (18, 19). The detection limit of the assay was 0.06 mg/liter. The interassay coefficient of variation was 3.33 percent at the mean level of 1.52 mg/liter (n = 116) and 2.65 percent at the mean level of 2.51 mg/liter (n = 168). Childhood serum samples were taken in 1980, stored at –20°C, and analyzed in 2005 using the same method as that used in 2001. Samples were not thawed or refrozen during storage.

Statistical analysis
Using an exact test, we evaluated Hardy-Weinberg equilibrium at each SNP locus on a contingency table of observed-versus-predicted genotype frequencies. The distribution of circulating CRP levels was highly skewed, and the level of circulating CRP was dependent on age and sex (20, 21). Thus, the three variables for circulating CRP used in the analyses were age- and sex-standardized z scores for log CRP in childhood, in adulthood, and across the life course (i.e., the average of age- and sex-standardized z scores for log CRP in childhood and adulthood). In agreement with previous studies (1, 5), we chose to use a model for the haplotype-CRP association that assumed that each of a participant's two haplotypes contributed additively to his/her CRP value. We used age- and sex-adjusted least-squares regression analysis to assess the association of haplotypes with circulating CRP indicators. All haplotypes were simultaneously entered into the regression model to obtain the total variance in circulating CRP that was explained by the haplotypes. We repeated this regression analysis with a single independent variable of 14 haplotype pairs instead of five separate haplotypes. All analyses were performed with SAS statistical software, version 9.1 (SAS Institute Inc., Cary, North Carolina).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Of the five CRP gene SNPs selected for analysis, two are promoter-region polymorphisms (–717 and –286), one is exonic (+1059), and two are in the 3'-untranslated region (+1444 and +1846). The genotype frequencies for all five SNPs were in Hardy-Weinberg equilibrium. Median CRP levels were 0.21 mg/liter (interquartile range, 0.11–0.53) in childhood and 0.65 mg/liter (interquartile range, 0.30–1.52) in adulthood (5). Because of skewness, mean CRP levels were higher: 0.65 mg/liter (standard deviation, 1.29) in childhood and 1.25 mg/liter (standard deviation, 1.60) in adulthood.

After exclusion of rare haplotypes (frequency <1 percent), the remaining five haplotypes were associated with circulating CRP levels. In all cases, haplotype had a stronger effect on the mean level of CRP across the life course than on the level of CRP in childhood or adulthood alone (table 1). All five haplotypes together explained 3.8 percent of the variation in childhood CRP, 3.0 percent of the variation in adulthood CRP, and 4.9 percent of the variation in life-course CRP. The corresponding figures for 14 haplotype pairs were 3.9 percent, 3.3 percent, and 5.0 percent, respectively.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Median C-reactive protein (CRP) levels and mean age- and sex-standardized z scores for CRP level in childhood and adulthood and across the life course, according to five CRP gene haplotypes* (n = 1,560), Cardiovascular Risk in Young Finns Study, 1980–2001

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
In this study, we empirically demonstrated that haplotypes in the CRP gene were associated with serum CRP levels in childhood and adulthood, with the strongest effects being found for average serum CRP level across the life course. This supports the underlying assumption in previous genetic association studies that the genetic variants define groups with lifelong differences in circulating CRP levels (22, 23). Previously, genetic variation in the CRP gene has shown to be associated with circulating CRP levels in adulthood (1–5), but not from childhood to adulthood, as in the present study.

Variants in the CRP gene have not generally been associated with coronary heart disease (2, 24, 25) or blood pressure, components of the metabolic syndrome (1, 26), or carotid intima-media thickness (5), a measure of the extent of subclinical atherosclerotic disease, except in one study of older people, where an association was found with coronary heart disease but not with carotid intima-media thickness (4). In our data, as previously reported, CRP haplotypes were associated with carotid arterial compliance (also a marker of atherosclerosis), but only in men, and there were no associations between the haplotypes and carotid intima-media thickness in men or women (5, 27). To date, evidence from genetic epidemiologic studies provides no consistent support for the status of circulating CRP level as a causal risk factor for coronary heart disease. According to the present findings, the underlying assumption in these studies that the genetic variants define groups with long-term differences in circulating CRP level seems to be correct.

The variance in childhood and adulthood CRP levels accounted for by haplotype pairs ranged from 3.3 percent to 3.9 percent and was 5.0 percent for average CRP level across the life course. In the Cardiovascular Health Study, a US study that also assessed five CRP SNPs, the variance explained by genetic determinants ranged from 2.6 percent to 6.4 percent, depending on population (4), but in the Framingham Heart Study, available genetic determinants accounted for less than 2 percent of the variance in circulating CRP levels (28). These robust associations suggest that several haplotypes and SNPs could be used to clarify whether lifelong CRP levels are causally related to clinical disease, such as coronary heart disease. However, the magnitude of the associations between genetic variants and CRP is small, and so are the absolute differences in median CRP levels between genotypes (≤0.61 mg/liter in this study) (29). This suggests that much larger genetic studies, carried out perhaps by pooling data sets, are still needed in order to obtain precise estimates and provide definite evidence about causality (23, 26, 30). Yet this cost may be worth paying, because overcoming the fundamental problems of residual confounding and bias in conventional observational studies would mean that limited funds for drug development and trials for testing drugs' effectiveness would be directed towards what would probably be the most effective targets. Since the phenotype of cardiovascular disease risk is complex and is related to clustering of cardiovascular risk factors (31), future studies should also examine whether CRP exerts biologically independent effects on coronary heart disease risk only in the presence of other risk factors observable in subsets of individuals.


   ACKNOWLEDGMENTS
 
The Cardiovascular Risk in Young Finns Study has been supported by the Academy of Finland (grants 77841, 34316, and 210283), the Social Insurance Institution of Finland, the Finnish Work Environment Foundation, the Turku University Foundation, the Juho Vainio Foundation, the Finnish Foundation of Cardiovascular Research, and the Finnish Cultural Foundation. M. K. was supported by the Academy of Finland (grants 105195 and 117604), D. A. L. by a United Kingdom Department of Health Career Scientist Award, M. H. by the Research Foundation of Tampere University Hospital, T. L. by the Emil Aaltonen Foundation and the Medical Research Fund of Tampere University Hospital, and J. S. A. V. and O. T. R. by research grants from Turku University Central Hospital.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Timpson NJ, Lawlor DA, Harbord RM, et al. C-reactive protein and its role in metabolic syndrome: mendelian randomisation study. Lancet (2005) 366:1954–9.[CrossRef][Web of Science][Medline]
  2. Casas JP, Shah T, Cooper J, et al. Insight into the nature of the CRP–coronary event association using Mendelian randomization. Int J Epidemiol (2006) 35:922–31.[Abstract/Free Full Text]
  3. 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) 350:1387–97.[Abstract/Free Full Text]
  4. Lange LA, Carlson CS, Hindorff LA, et al. Association of polymorphisms in CRP gene with circulating C-reactive protein levels and cardiovascular events. JAMA (2006) 296:2703–11.[Abstract/Free Full Text]
  5. Kivimäki M, Lawlor DA, Eklund C, et al. Mendelian randomization suggests no causal association between C-reactive protein and carotid intima-media thickness in the Young Finns Study. Arterioscler Thromb Vasc Biol (2007) 27:978–9.[Free Full Text]
  6. Davey Smith G, Ebrahim S. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol (2003) 32:1–22.[Abstract/Free Full Text]
  7. Davey Smith G, Ebrahim S. What can mendelian randomisation tell us about modifiable behavioural and environmental exposures? BMJ (2005) 330:1076–9.[Free Full Text]
  8. Knowler WC, Williams RC, Pettitt DJ, et al. Gm3;5,13,14 and type 2 diabetes mellitus: an association in American Indians with genetic admixture. Am J Hum Genet (1988) 43:520–6.[Web of Science][Medline]
  9. Cardon LR, Palmer LJ. Population stratification and spurious allelic association. Lancet (2003) 361:598–604.[CrossRef][Web of Science][Medline]
  10. Stock JH, Wright JH, Yogo M. A survey of weak instruments and weak identification in generalized method of moments. J Bus Econ Stat (2002) 20:518–29.[CrossRef]
  11. Staiger D, Stock JH. Instrumental variables regression with weak instruments. Econometrica (1997) 65:557–86.[CrossRef][Web of Science]
  12. Akerblom HK, Uhari M, Pesonen E, et al. Cardiovascular risk in young Finns. Ann Med (1991) 23:35–9.[Web of Science][Medline]
  13. Raitakari OT, Juonala M, Kahonen M, et al. Cardiovascular risk factors in childhood and carotid artery intima-media thickness in adulthood: The Cardiovascular Risk in Young Finns Study. JAMA (2003) 290:2277–83.[Abstract/Free Full Text]
  14. Carlson CS, Aldred SF, Lee PK, et al. Polymorphisms within the C-reactive protein (CRP) promoter region are associated with plasma CRP levels. Am J Hum Genet (2005) 77:64–77.[CrossRef][Web of Science][Medline]
  15. Stephens M, Donnelly P. A comparison of Bayesian methods for haplotype reconstruction from population genotype data. Am J Hum Genet (2003) 73:1162–9.[CrossRef][Web of Science][Medline]
  16. Kilpatrick JM, Volanakis JE. Molecular genetics, structure, and function of C-reactive protein. Immunol Res (1991) 10:43–53.[Web of Science][Medline]
  17. Mensah FK, Gilthorpe MS, Davies CF, et al. Haplotype uncertainty in association studies. Genet Epidemiol (2007) 31:348–57.[CrossRef][Web of Science][Medline]
  18. Kivimäki M, Lawlor DA, Juonala M, et al. Lifecourse socioeconomic position, C-reactive protein, and carotid intima-media thickness in young adults: The Cardiovascular Risk in Young Finns Study. Arterioscler Thromb Vasc Biol (2005) 25:2197–202.[Abstract/Free Full Text]
  19. Juonala M, Viikari JS, Ronnemaa T, et al. Childhood C-reactive protein in predicting CRP and carotid intima-media thickness in adulthood: The Cardiovascular Risk in Young Finns Study. Arterioscler Thromb Vasc Biol (2006) 26:1883–8.[Abstract/Free Full Text]
  20. Raitakari M, Mansikkaniemi K, Marniemi J, et al. Distribution and determinants of serum high-sensitive C-reactive protein in a population of young adults: The Cardiovascular Risk in Young Finns Study. J Intern Med (2005) 258:428–34.[CrossRef][Web of Science][Medline]
  21. Bansal N, Cruickshank JK, McElduff P, et al. Cord blood lipoproteins and prenatal influences. Curr Opin Lipidol (2005) 16:400–8.[Web of Science][Medline]
  22. Davey Smith G. Randomised by (your) God: robust inference from an observational study design. J Epidemiol Community Health (2006) 60:382–8.[Free Full Text]
  23. Lawlor DA, Harbord RM, Sterne JAC, et al. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. In: Stat Med. (in press).
  24. Zee RYL, Ridker PM. Polymorphism in the human C-reactive protein (CRP) gene, plasma concentrations of CRP, and risk of future arterial thrombosis. Atherosclerosis (2002) 162:217–19.[CrossRef][Web of Science][Medline]
  25. Kardys I, de Maat MP, Uitterlinden AG, et al. C-reactive protein gene haplotypes and risk of coronary heart disease: The Rotterdam Study. Eur Heart J (2006) 27:1331–7.[Abstract/Free Full Text]
  26. Davey Smith G, Harbord R, Ebrahim S. Fibrinogen, C-reactive protein and coronary heart disease: does Mendelian randomization suggest the associations are non-causal? QJM (2004) 97:163–6.[Free Full Text]
  27. Eklund C, Kivimäki M, Shaheenul Islam M, et al. C-reactive protein genetics is associated with carotid artery compliance in men in the Cardiovascular Risk in Young Finns Study. Atherosclerosis (2007) Mar 8 [Epub ahead of print].
  28. Kathiresan S, Larson MG, Vasan RS, et al. Contribution of clinical correlates and 13 C-reactive protein gene polymorphisms to interindividual variability in serum C-reactive protein level. Circulation (2006) 113:1415–23.[Abstract/Free Full Text]
  29. Sabatine MS, Morrow DA, Jablonski KA, et al. Prognostic significance of the Centers for Disease Control/American Heart Association high-sensitivity C-reactive protein cut points for cardiovascular and other outcomes in patients with stable coronary artery disease. Circulation (2007) 115:1528–36.[Abstract/Free Full Text]
  30. Davey Smith G, Ebrahim S. Mendelian randomization: genetic variants as instruments for strengthening causal influence in observational studies. (2007) Washington, DC: National Academy of Sciences.
  31. Lloyd-Jones DM, Liu K, Tian L, et al. Narrative review: assessment of C-reactive protein in risk prediction for cardiovascular disease. Ann Intern Med (2006) 145:35–42.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Eur Heart JHome page
M. Kivimaki, G. D. Smith, N. J. Timpson, D. A. Lawlor, G. D. Batty, M. Kahonen, M. Juonala, T. Ronnemaa, J. S.A. Viikari, T. Lehtimaki, et al.
Lifetime body mass index and later atherosclerosis risk in young adults: examining causal links using Mendelian randomization in the Cardiovascular Risk in Young Finns study
Eur. Heart J., October 2, 2008; 29(20): 2552 - 2560.
[Abstract] [Full Text] [PDF]

Home page
 Int J EpidemiolHome page
D Gimeno, J E Ferrie, M Elovainio, L Pulkki-Raback, L Keltikangas-Jarvinen, C Eklund, M Hurme, T Lehtimaki, J Marniemi, J S A Viikari, et al.
When do social inequalities in C-reactive protein start? A life course perspective from conception to adulthood in the Cardiovascular Risk in Young Finns Study
Int. J. Epidemiol., April 1, 2008; 37(2): 290 - 298.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
166/7/760    most recent
kwm151v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kivimäki, M.
Right arrow Articles by Raitakari, O. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kivimäki, M.
Right arrow Articles by Raitakari, O. T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?