The aim of the present study was to assess the longterm stability of endogenous Btype natriuretic peptide (BNP) and amino terminal proBNP (NTproBNP) in plasma samples stored at -20degreesC without addition of protease inhibitors (e.g., aprotinin). Stability of BNP and NTproBNP was tested in 60 EDTA plasma samples with BNP values between 30 and 420 pg/ml. Initial BNP and NTproBNP plasma concentrations were determined within four hours after blood collection using the AxSYM BNP and the Elecsys NTproBNP assays. Subsequently, all samples were stored at -20degreesC and were thawed for the second BNP and NTproBNP determination on the two instruments after one day, 30 days, 60 days, 90 days and 120 days, respectively. Mean recovery (i.e., residual immunoreactivity) of BNP and NTproBNP expressed in percent of the initial value for the given time interval of storage was calculated. Mean recovery of BNP was less than 70% after one day of storage at -20degreesC and decreased to less than 50% after two to four months of storage (e.g., recovery of endogenous BNP after three months of storage at -20degreesC ranging from 0% to 71%). In contrast, mean recovery of NTproBNP was generally greater than 90%, irrespective of the duration of storage at -20degreesC (e.g., recovery of endogenous NTproBNP after three months of storage at -20degreesC ranging from 91% to 112%). In conclusion, the determination of endogenous BNP with the AxSYM assay using frozen plasma samples may not be valid under the conditions tested. In contrast, NTproBNP as measured by the Elecsys assay may be stored at -20degreesC for at least four months without a relevant loss of the immunoreactive analyte.