The goals of this multicenter controlled trial were: (1) to study the short-term effect of plasma exchange in the Guillain-Barré syndrome when applied alone within 17 days of onset of the disease; and (2) to compare two replacement fluids, diluted albumin and fresh frozen plasma (FFP) with regard to efficacy and morbidity. Two hundred twenty patients were included, 111 in the control group, 109 in the plasma exchange group, 57 of whom were assigned to receive albumin and 52 to receive fresh frozen plasma. Treatment consisted of four plasma exchanges of two plasma volumes each, initiated on the day of randomization and repeated on alternate days. Significant short-term benefits appeared in the group that received plasma exchange as demonstrated by the reduction in the proportion of patients who required assisted ventilation after randomization, and the decrease in time before beginning weaning from ventilator, time to onset of motor recovery, and time to walk with and without assistance. No statistically significant difference was found between the group that received albumin and the group that received fresh frozen plasma. Incidents during exchanges and complications related to the plasma exchange were more frequent in patients who received fresh frozen plasma. Plasma exchange is beneficial in Guillain-Barré syndrome when it is carried out early in the course of the disease. Given its risks and the lack of its clear superiority over albumin, however, we recommend that fresh frozen plasma be abandoned in the treatment of Guillain-Barré syndrome.