Continuous renal replacement therapy (CRRT) is a dialysis modality used to treat patients with severe acute kidney injury. Nevertheless, there is limited information on the predictors of weaning from CRRT. The present study examined whether the N-terminal prohormone of brain natriuretic peptide (NT-proBNP) can predict weaning from CRRT, based on the fact that this cardiac neurohormone is known to predict kidney dysfunction. Plasma NT-proBNP and several other baseline parameters at the time of starting CRRT were retrieved from 160 patients. The odds ratio (OR) for weaning from the CRRT within two weeks was calculated using a multivariate stepwise logistic model. We calculated the cut off value predicting weaning outcome by using the receiver operating characteristic curve and corresponding Youden index, and then divided patients into high (n = 74) and low (n = 86) NT-proBNP groups. The high NT-proBNP group had a lower weaning rate than the low NT-proBNP group [adjusted OR,; P = 0.007]. We additionally found other predictors of weaning, such as sex, serum creatinine, urine output, and the score from the Acute Physiology and Chronic Health Evaluation, but all of these were not better than NT-proBNP in the predictability of weaning outcome. Neutrophil gelatinase-associated lipocalin, a wellknown biomarker of acute kidney injury and originating from kidney, was not related with the CRRT weaning, which indicated the usefulness of NT-proBNP in the cases of CRRT despite originating from heart. The present study addresses the potential of NT-proBNP as an independent predictor of weaning from CRRT.