Background: Serum concentration of high sensitive C-reactive protein (hsCRP) can predict the risk of chronic metabolic and cardiovascular diseases but it is unclear whether turbidimetric high sensitive assays of CRP are adequate.
Methods: Concentrations of serum CRP in 126 samples of serum were measured with high-sensitivity methods using nephelometry (BN II Nephelometer) and turbidimetry (Ortho Vitros FS 5.1).
Results: For CRP concentrations measured by nephelometry and turbidimetry intra-assay CVs were 3.2 and 0.9% at mean CRP concentrations of 1.4 and 2.1 mg/l, inter-assay CVs for commercial controls were 3.1% and 3.6% at mean concentrations of 1.3 and 1.7 mg/l, and mean biases were 7.62% and 2.26%, respectively. Measurements were strongly, linearly correlated (r = 0.99; CRP vitros = 0.03 +1.03 CRP (BN II)). When disease risk was assessed by nephelometry and turbidimetry, results were similar. If the risk of disease was classified as moderate (1.0 < CRP < or = 3.0 mg/l) or high (CRP > 3.0 mg/l), the frequency of misclassified cases was only 2.3 and 2.1%, respectively. The classification agreement weighted kappa coefficient was 0.94 (95% C.I.: 0.89-0.98).
Conclusions: turbidimetric high sensitive CRP assays can properly classify CRP-related prediction of chronic metabolic diseases with special consideration on cardiovascular risk.