Background: C-reactive protein (CRP) is a nonspecific marker of inflammation that can be used as a marker for atherosclerotic risk. This application requires increased precision at low CRP concentrations compared to traditional assays. Methods: The Dimension RXL is an automated chemistry analyzer for central laboratory use. The limit of detection, limit of quantification, linearity and imprecision of a high-sensitivity CRP assay developed for it were assessed. Method comparison studies were performed using samples both inside and outside the reference interval. The presence of a prozone effect was also evaluated. Results: The limit of detection was 0.7 mg/l. The method was linear from 2 to 60 mg/l and from 1 to 60 mg/l using systematic error limits of 10% and 20%, respectively. The total imprecision was <10% for CRP concentrations above 1.5 mg/l. No prozone effect was seen at a CRP concentration of 450 mg/l, the highest concentration tested. Using samples from 212 apparently healthy adults, the Dimension RXL method demonstrated good concordance with the BN II high-sensitivity CRP method for samples in the highest quartile. It also compared well using samples with elevated CRP concentrations. Conclusions: The Dimension RXL high-sensitivity CRP method may be adequate for atherosclerotic risk prediction in clinical practice if accurate and precise measurement is only required for the highest quartile. However, the total error of this method for CRP concentrations <3 mg/l appears too large for accurate assignment to lower risk groups.