Reliability of two different bedside assays for C-reactive protein in newborn infants
Background: Bedside tests for C-reactive protein (CRP) have been studied in pediatric patients, but not in neonates. Methods: This study compared the results of two rapid bedside tests for CRP (Quick-Read CRP, Orion Diagnostic, Espoo, Finland and NycoCard CRP-Single Test, Axis-Shield, Oslo, Norway) with those of our central laboratory method (CRP-Lab) in newborn infants. CRP concentrations were determined using 72 samples obtained from 43 infants with suspected sepsis occurring between 1 and 28 days of life. Results: Considering positive CRP concentrations to be ≥10 mg/L, both bedside tests had good specificity (Quick-Read 80.5%, NycoCard 83.3%) and sensitivity (Quick-Read 97.2%, NycoCard 94.4%) when compared with our CRP-Lab. The agreement of measurement with central laboratory values was high for both the bedside tests, without statistically significant differences between the methods. The Quick-Read and NycoCard methods did not show any statistically significant systematic proportional bias when compared with the central laboratory values. The accuracy of the results of both bedside tests is somewhat decreased when CRP concentrations are >100 mg/L. Conclusions: This study shows that both the Quick-Read and the NycoCard test can be used for serial determinations of CRP concentrations in newborn infants. They require small volumes of blood and provide reliable results in <5 min. Clin Chem Lab Med 2009;47:1081–4.