Fresh frozen plasma (FFP) has been proposed as a specific therapy for acute pancreatitis. Reduced mortality encountered in an uncontrolled clinical study and a controlled experimental study may be attributable to replenishment by FFP of the naturally occurring antiprotease system. To investigate this potential therapy further, 202 patients presenting with acute pancreatitis were randomized to receive FFP (2 units daily for 3 days) or a similar volume of colloid control as part of their intravenous fluid therapy. Clinical progress was monitored and the major serum antiproteases (alpha 1-antiprotease and alpha 2-macroglobulin) were measured on days 1, 3 and 7. There was no significant difference between the two groups in terms of clinical outcome. alpha 1-Antiprotease levels rose significantly from day 1 to day 3 in both groups (P less than 0.0001) and remained elevated at day 7. alpha 1-Antiprotease is an acute phase protein in man and raised serum levels would be anticipated. FFP appears to have no effect on the magnitude of this rise. Serum alpha 2-macroglobulin levels were reduced in both groups on day 1 and continued to fall significantly from day 1 to day 3 in the colloid control group (P less than 0.005) whilst remaining substantially unaltered in patients receiving FFP (P = 0.6527). alpha 2-Macroglobulin plays a central role in the elimination of proteases during acute pancreatitis and the ability of relatively low volumes of FFP to reduce the fall in serum alpha 2-macroglobulin levels seen during the early stages of this disease may have therapeutic implications.