Serum B-type natriuretic peptide (BNP) is increased after heart transplantation (HT), but it has not been well established whether BNP could be used to detect acute rejection in asymptomatic patients after HT. A total of 259 routine endomyocardial biopsy specimens from 50 consecutive patients after HT (83% men; age 50 +/- 15 years) were studied. Serial BNP measurements were performed at the time of each biopsy. BNP was evaluated as an absolute level (picograms per milliliter) and percentage of change from the previous biopsy (BNP - BNP at previous biopsy)/BNP at previous biopsy] x 100). Rejection was defined as grade > or =2R International Society of Heart and Lung Transplantation grading system. BNP correlated independently with time after HT (p <0.001), pulmonary artery systolic pressure (p <0.001), creatinine (p = 0.001), and age (p = 0.0012). Asymptomatic rejection was found in 15 biopsy specimens (6%), for which absolute BNP (106 pg/ml; interquartile range [IQR] 67 to 495) did not differ from nonrejection biopsy specimens (92 pg/ml; IQR 49 to 230; p = 0.286). BNP percentage of change showed a median of +60% (IQR -29 to +154%) in rejection versus -17% (IQR -47 to +19%) in nonrejection biopsy specimens (p = 0.009). After multivariable adjustment, BNP percentage of change was a consistent predictor of rejection (+10%; odds ratio 1.05, 95% confidence interval 1.01 to 1.09, p = 0.021). Receiver-operator characteristic analysis showed an area under the curve of 0.71 (95% confidence interval 0.643 to 0.768) and identified percentage of change <+38% as an optimal cut-off point, with a negative predictive value of 97%. In conclusion, serial monitoring of BNP, evaluated as a percentage of change, may be a useful noninvasive tool in the clinical management of rejection.