Abstract: Objective To explore the effect of ultrasound pretreatment for the detection of plasma endotoxin. Methods We used the platelet rich plasma and platelet poor plasma as the research objects. These plasma samples were pretreated with ultrasound or other methods and divided into 6 groups with 5 samples in every group: group A (platelet rich plasma pretreated with ultrasound), group B (platelet rich plasma pretreated with vortex), group C (platelet rich plasma pretreated with endotoxin reagent), group D (platelet poor plasma pretreated with ultrasound), group E (platelet poor plasma pretreated with vortex), and group F (platelet
poor plasma pretreated with endotoxin reagent). Trace endotoxin in plasma was assayed by limulus lysate reagent using end point azo chromogenic method. Results ①The recovery rate for endotoxin was significantly different, but the optimal recovery rate had no significant difference between group A and group B (t=
9.424, P＜0.001 for recovery rate; x²=1.111, P=0.500 for optimal recovery rate), and between group D and group E (t=30.395, P＜0.001 for recovery rate; x²=2.500, P=0.222 for optimal recovery rate). The recovery rate for endotoxin had no significant difference, so did the optimal recovery rate between group A and group C (t= - 0.815, P=0.439 for recovery rate; x²=0.000, P=0.778 for optimal recovery rate), between group D and group F (t= - 0.375, P=0.717 for recovery rate; x²=0.000, P=0.527 for optimal recovery rate), and between group A and group D (t=0.355, P=0.732 for recovery rate; x²=0.476, P=0.500 for optimal recovery rate). ② The variation coefficient of the results from the groups pretreated with ultrasound was narrower (2.7% and 3.5%), and the recovery rate was 90-100%, more accurate than the national standard of 50～200%. Conclusions The pretreatment with ultrasound for the detection of endotoxin in plasma has the advantages of lower cost, higher sensitivity, reliable result and higher efficiency.

Key words: endotoxin, ultrasonicmethod, plasma, detection