We thank Ridker and colleagues for their interest in our recent publication.1 C-reactive protein level clearly predicts risk for cardiovascular disease events, but the extent to which CRP level improves such estimates is controversial.
A critical issue is the identification of individuals with higher levels of CRP with the assay we used compared with the assay mentioned by Ridker et al. In 2002, we performed a crossover analysis to compare the Hemagen assay (Hemagen Diagnostics Inc, Columbia, Md) we used and the Dade Behring nephelometric assay (Dade Behring Inc, Deerfield, Ill). We found a very high correlation of 0.98 and high degree of agreement for CRP level categorized by quartiles. Similar results have been obtained by others in comparisons of medium and high-sensitivity CRP assays.2 We share Ridker and colleagues' interest in the development of a CRP standardization program, but at present there is no formal program in place.
Wilson PWF, Pencina M, D’Agostino RB, O’Donnell CJ. C-Reactive Protein and Cardiovascular Risk in the Framingham Study—Reply. Arch Intern Med. 2006;166(12):1327-1328. doi:10.1001/archinte.166.12.1328