C-reactive protein (CRP), the classic acute phase reactant, is strongly associated with increased risk of cardiovascular events. The demand for measuring serum CRP levels has been predicted to increase. We developed an ultra-sensitivity in-house immunometric assay on polystyrene beads for measuring CRP and studied its analytical and clinical performance. The assay used a pair of monoclonal anti-CRP antibodies and detected CRP in a 1-step immunometric assay with a chemiluminescence signal. The calibration was traceable to the World Health Organization reference material. The assay covered a linear range of 0.01 to 50.00 mg/L. The analytical detection limit calculated from the mean level plus 3 SD of the zero calibrator was 0.004 mg/L. The within-run imprecision was 7.0%, 5.2%, and 4.1% for mean CRP levels of 0.02 mg/L, 1.44 mg/L, and 11.04 mg/L, respectively. The between-run imprecision was 9.2%, 7.0%, and 6.0% for mean CRP levels of 0.02 mg/L, 1.49 mg/L, and 10.90 mg/L, respectively. The average recovery was 102.0% (n=6). The assay correlated well with a high-sensitivity latex-enhanced nephelometric assay (regression line y=0.865 x +1.333, r=0.974, S(y/x)=3.415, n=47 for 0-50.00 mg/L and y=1.076 x-0.080, r=0.985, S(y/x)=0.989, n=29 for 0-20.00 mg/L). The central 95 percentile reference interval for Han Chinese residing in Taiwan was 0.02-4.33 mg/L (n=469). There was no significant difference in serum CRP levels between healthy male and female subjects (median, 0.34 and 0.31 mg/L, respectively); however, CRP levels increased moderately with age (r=0.276, P<.05). The reference values for the Chinese population were about 5-fold lower than those for the United States population. This ultra-sensitivity immunochemiluminometric assay for CRP is rapid and accurate and can be used to assess cardiovascular risk.