ABSTRACT
BACKGROUND: In the second generation of the point-of-care (POC) assay Roche CARDIAC proBNP, the upper limit of the measuring range was extended from 3000 to 9000 ng/L. METHODS: A thirteen-site multicentre evaluation was carried out to assess the analytical performance of the POC NT-proBNP assay and to compare it with a laboratory N-terminal pro-brain natriuretic peptide (NT-proBNP) assay. RESULTS: In method comparisons of six lots of POC NT-proBNP with the lab reference method (Elecsys proBNP) mean bias ranged from -10 to +17%. In lot-to-lot comparisons all six investigated lots of POC NT-proBNP showed excellent agreement, with mean bias between -7% and +2%. The majority of all coefficients of variation obtained from ten-fold measurements using 56 native blood samples were below 8%. No interference was observed with hemolytic blood (hemoglobin concentrations up to 0.12 mmol/L), lipemic blood (triglyceride concentrations up to 14.0 mmol/L) nor icteric blood (bilirubin concentrations up to 63 micromol/L). Hematocrit values between 24% and 51% had no influence on the assay result. High NT-proBNP concentrations above the measuring range of POC NT-proBNP did not lead to false low results due to potential high-dose hook effect. Results with POC NT-proBNP were not influenced by different ambient temperatures (18 degrees C to 32 degrees C), the sample material used, nor by over- or underdosing by 15 microL compared to the regular sample volume of 150 microL. CONCLUSIONS: The POC NT-proBNP assay showed an excellent analytical performance including a good agreement with the laboratory method. The assay is therefore suitable for its intended use in point-of-care settings.