1.BNP and ANP are important research indices of severity of heart failure. However, uncertainty regarding the stability of these peptides at room temperature has limited their use to assess cardiac function in routine clinical practice. 2. We assessed the stability of BNP and ANP in blood samples left for 2 h or 2 days at room temperature compared with levels in blood processed immediately (initial). These times were chosen to reflect possible times for samples to be processed in a hospital outpatient clinic (2 h) or a blood sample posted to a laboratory from general practice (2 days). Samples were obtained from eight heart transplant recipients. Blood was separated and plasma stored immediately after collection (initial) and after 2 h or 2 days at room temperature respectively. 3. Initial plasma BNP and ANP values measured by radioimmunoassay after Sep-Pak extraction were 38.9+/-11.1(S.E.M.) pg/ml and 113.6+/-28.1 pg/ml, respectively. After 2 h at room temperature there was no significant fall in either peptide level (35.5+/-9.9 pg/ml, BNP; 104. 9+/-30.6 pg/ml, ANP). However, after 2 days at room temperature there was a significant fall in ANP to 38.1+/-12.6 pg/ml (P<0.005 versus initial level). In contrast, there was no significant fall in BNP after 2 days (32.0+/-8.4 pg/ml). After 2 days at room temperature only 30.4+/-4.3% of the ANP remained, but 86.0+/-5.0% of BNP compared with the initial ANP and BNP measurements. 4. Our study clearly showed that ANP is stable for 2 h and thus could be useful as a screening test for heart disease in hospital. In contrast, BNP remained stable for 2 days. Measuring BNP may thus be practical as a test of heart function both for routine use in hospital and by general practitioners in the community.