Abstract
Background: We organized and conducted a proficiency testing study (CardioOrmocheck) to evaluate the differences in analytical performance of brain natriuretic peptide (BNP) and N-terminal proBNP (NT-proBNP) immunoassays.
Methods: Approximately 90 Italian laboratories were involved in the 2005–2007 proficiency testing cycles, while 112 laboratories took part in the 2008 cycle (from January to May 2008). A total of 28 study samples were measured by participating laboratories for a total of 2354 determinations.
Results: The mean total variability for BNP (50.6 %CV) was significantly higher than that for NT-proBNP (8.4 %CV). In addition, the mean variability due to differences between-methods (46.4 %CV) comprised the majority of the total variability for BNP. Between-method variability for BNP comprised, on average, 84% of total variability, while the within-method variability comprised an average of 20.2 %CV. On the contrary, for NT-proBNP the within-method variability (7.3 %CV) represented the majority of total variability (average 75%), while between-method variability was smaller (4.1 %CV). Imprecision around the cut-off value showed marked differences among methods, especially for BNP immunoassay methods. In addition, BNP methods were affected by large systematic differences, for example an average 2.7-fold difference between Access and ADVIA Centaur methods, while agreement between NT-proBNP methods was better (an average 1.2-fold difference between Dimension and ECLIA on the Elecsys methods).
Conclusions: This multicenter collaborative study demonstrates that there are significant differences in analytical characteristics and measured values among the most popular commercial methods for BNP and NT-proBNP. Clinicians should be very careful when comparing results obtained by laboratories that use different methods.
Clin Chem Lab Med 2009;47:762–8.
©2009 by Walter de Gruyter Berlin New York
