Objective To study the plasma degradation of platelet-activating factor in severely ill patients with clinical sepsis. Design A prospective, nonrandomized control study. Setting Intensive care unit in a university hospital. Patients Thirteen critically ill male patients with clinical sepsis, due to medical or surgical illness, and ten normal male volunteers were studied. Measurements were repeated in seven patients who survived. Measurements and Main Results The plasma activity of acetylhydrolase, the lipoprotein-associated enzyme that hydrolyses platelet-activating factor to its biologically inactive lyso-derivative was determined using an optimized enzyme assay. The plasma half-life of platelet-activating factor was also measured, along with phospholipase A2 activity, lyso-platelet-activating factor, and serum lipid concentrations. Patient results were compared with those results of normal controls and followed once in survivors. Acetylhydrolase activity in the patient group was significantly lower than in normal subjects (median 34, interquartile range 17 to 54 nmol/min/mL vs. median 60, interquartile range 56 to 80 nmol/min/mL; p < .002), while overall, the plasma half-life of platelet-activating factor did not differ significantly between the groups. However, the half-life of platelet-activating factor in six patients who died (median 3.3, range 3.3 to 4.3 mins) was significantly greater than in either survivors (median 2.1, range 1.4 to 2.9 mins; p < .001) or the normal group (median 2.5, range 2.2 to 2.8 mins; p < .001). Consistent with theoretical prediction, a significant linear relationship existed between platelet-activating factor half-life and the reciprocal of acetylhydrolase activity in the patient group (p < .05). Plasma phospholipase A2 activity was markedly increased in the patient group, while plasma lyso-platelet-activating factor and serum lipid concentrations were severely decreased. Conclusions Depression of acetylhydrolase activity was consistent with the concentration of lipids with which it is associated. Platelet-activating factor half-life was relatively well preserved because of the nature of its relationship with enzyme activity. The half-life was prolonged in those patients with the worst outcome and the breakdown in plasma degradation of platelet-activating factor could have contributed to pathophysiology in these subjects. (Crit Care Med 1994; 22:204–212)