Randomization was performed in computer-generated, permuted blocks of 6 to 10, stratified according to the participating ICU. Treatment assignments and a five-digit reference number were placed in sealed, opaque envelopes, which were opened by the person responsible for the preparation of the trial-drug solutions. The solutions of norepinephrine or dopamine were prepared in vials or syringes according to the preference of the local ICU. Each vial or syringe was then labeled with its randomly allocated number. The doctors and nurses administering the drugs, as well as the local investigators and research personnel who collected data, were unaware of the treatment assignments. The trial was approved by the ethics committee at each participating center. Written informed consent was obtained from all patients or next of kin.