European Heart Journal Advance Access originally published online on April 16, 2008
European Heart Journal 2008 29(10):1335; doi:10.1093/eurheartj/ehn149
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
New cardiovascular risk determinants do exist and are clinically useful
Department of Pathology and
Laboratory Medicine
University of Louisville and Laboratory
Service
VAMC
800 Zorn Avenue
Louisville
KY 40206
USA
Department of Veterans Affairs
Medical Center
800 Zorn Avenue
Louisville
KY 40206
USA
Tel: +1 502 287 5565
Fax: +1 502 287 6265
Email: levinson{at}louisville.edu
Footnotes
Published by Oxford University Press on behalf of the European Society of Cardiology 2008
I read with interest the article by Smulders et al., in which the authors noted that the key to understanding a statistically significant relationship between independent risk predictions, but no improvement in risk classifications, lies with the statistical methodology used for the assessment.1 But, it seems to me that the final criteria for defining biomarkers for clinical use must be from information obtained from well-designed prospective outcome studies that show clinical management can be altered on the basis of test results leading to improved outcomes,2 in an economically reasonable way. The statistical tests used to evaluate studies preformed prior to outcome studies (preliminary studies) help in pointing towards which biomarkers might show sufficient diagnostic discrimination to be tested in what may be large expensive prospective studies that define clinical outcome.
In their article, Smulders et al.1 discussed a number of statistical techniques that may be used in preliminary studies to assess biomarkers for outcome studies. These include c-statistic, risk ratio/odds ratio (RR/OR), and reclassification. It has been demonstrated that RR/OR is more sensitive for identifying biomarkers than c-statistic.3 For inflammatory markers tested, so far there is a problem in that the RR/OR falls4–6 into a range of c-statistics between 0.6 and 0.5, denoting very poor diagnostic discrimination. With such weak associations, the between-study variance is large, giving rise to RR/ORs that vary between 1.0 and 4.4, so that the true RR/OR cannot be known.4–8 One way to determine whether such studies may be clinically useful is to calculate revised positive predictive values (those base on the actual incidence). Because the incidence of disease in the population is low and the relationship is weak, in the best case the predictive value is usually about 1%.9 That is to say for every 100 positive results (above a cut off) 99 are false positives. The same problem affects reclassification also. Thus, it has been shown in such cases that reclassification may cause many more persons without the condition to be reclassified into the high-risk group than persons with the condition, giving rise to worse performance.10 Moreover, because of the variability associated with weak relationships, reclassification of persons at intermediate risk for coronary disease to high risk varies substantially from study to study just as the RR/OR does. For example, in one study, reclassification of persons at intermediate risk for coronary disease to high risk on the basis of C-reactive protein was limited to 2.7%,6 whereas another study showed a reclassification of 12% for those at intermediate risk.3
As Wang discussed in the accompanying article,11 recent clinical studies underscore just how well traditional biomarkers (that are true risk factors whose modification directly reduces risk) perform with regard to the prediction of future cardiovascular risk. In my view, the problem with new biomarkers, tested so far, is that the weak relationships make the challenge great in terms of design and cost for prospective-outcome studies that are apt to fail.
Footnotes
Published by Oxford University Press on behalf of the European Society of Cardiology 2008
References
- Smulders YM, Thijs A, Twisk JW. New cardiovascular risk determinants do exist and are clinically useful. Eur Heart J (2008) 29:436–440.
[Abstract/Free Full Text] - Sackett DL, Haynes RB. Evidence base of clinical diagnosis. The architecture of diagnostic research. BMJ (2002) 324:539–541.
[Free Full Text] - Cook N. Use and misuse of the receiver operator characteristic curve in risk prediction. Circulation (2007) 115:928–935.
[Abstract/Free Full Text] - St-Pierre AC, Cantin B, Bergeron J, Pirro M, Dagenais GR, Després JP, Lamarche B. Inflammatory markers and long-term risk of ishemic heart disease in men a 13 year follow-up of the Quebec Cardiovascular Study. Atherosclerosis (2005) 182:315–321.[CrossRef][Web of Science][Medline]
- Folsom AR, Chambless LE, Ballantyne CM, Coresh J, Heiss G, Wu KK, Boerwinkle E, Mosley TH Jr, Sorlie P, Diao G, Sharrett AR. An assessment of incremental coronary risk prediction using C-reactive protein and other novel risk markers: The atherosclerosis risk in communities study. Arch Intern Med (2006) 166:1368.
[Abstract/Free Full Text] - Wang TJ, Gona P, Larson MG, Tofler GH, Levy D, Newton-Cheh C, Jacques PF, Rifai N, Selhub J, Robins SJ, Benjamin EJ, D'Agostino RB, Vasan RS. Multiple biomarkers for the prediction of first major cardiovascular events and death. N Engl J Med (2006) 355:2631–2639.
[Abstract/Free Full Text] - Ridker PM, Buring JE, Shih J, Matias M, Hennekens CH. Prospective study of C-reactive protein and risk of future cardiovascular events among apparently healthy women. Circulation (1998) 98:731–733.
[Abstract/Free Full Text] - 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.
[Abstract/Free Full Text] - Levinson SS, Elin RJ. What is the acute phase reactant c-reactive protein telling us about atherosclerosis? Arch Int Med (2002) 162:389–392.
[Free Full Text] - Pepe MS, Janes H, Gu JW. Use and misuse of receiver operator characteristic curve in risk prediction. (Letter). Circulation (2007) 116:e132.
[Free Full Text] - Wang TJ. New cardiovascular risk factors exist, but are they clinically useful? Eur Heart J (2008) 29:441–444.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||