Extremely low doses of tissue factor pathway inhibitor decrease mortality in a rabbit model of septic shock

Byline: R. Matyal (1), Y. Vin (1), R. Delude (1), C. Lee (1), A. Creasey (2), M. Fink (1) Keywords: Peritonitis Coagulation Escherichia coli Disseminated intravascular coagulation Tissue factor pathway inhibitor Abstract: Objective: We sought to determine the lowest dose of recombinant human tissue factor pathway inhibitor (TFPI) that can provide protection from lethality in a rabbit model of septic shock. Methods: Sepsis was induced in New Zealand white rabbits by intraperitoneal implantation of 7.0 ml of a solution containing hemoglobin (4.8 g/dl), porcine mucin (6 g/dl), and 0.8--1.4x10.sup.4 viable Escherichia coli (strain O:18 K+). Gentamicin (5 mg/kg) was administered 4 h following surgery, and this dose was repeated every 12 h for 3 days. Beginning 4 h following the induction of sepsis, animals were treated with a bolus (1 ml) plus a continuous infusion (100 ml over 24) of either TFPI (various doses) or its vehicle. Four different doses of TFPI were studied, and each experiment included a contemporaneous control group. The primary outcome parameter was survival time. Results were analyzed using the Wilcoxen log rank test. Results: The average survival time for rabbits treated with the highest dose of TFPI tested (50 Aug/kg bolus and 0.5 Aug/kg per minute infusion) was 118 h, as compared to 81 h in vehicle-treated controls). The average survival time for septic rabbits treated with a much lower dose of TFPI (100 ng/kg bolus and 1.0 ng/kg per minute infusion) was 119 h as compared to 57 h in surviving vehicle-treated controls. Treatment with an even lower dose of TFPI (10 ng/kg bolus and 0.1 ng/kg per minute infusion) still produced a marginally significant prolongation of average survival time (80 h) relative to contemporaneously studied controls (47 h). When the dose of TFPI was decreased still further (1.0 ng/kg bolus and 0.01 ng/kg per minute infusion), average survival times were not significantly different between TFPI-treated and vehicle-treated rabbits (77 and 51 h, respectively). Conclusions: Delayed infusion with remarkably low doses of recombinant human TFPI prolongs survival in a rabbit model of antibiotic-treated Gram-negative bacterial sepsis. In planning human trials of TFPI as an adjuvant treatment for sepsis it may be reasonable to use much lower doses of the agent than were heretofore contemplated. Author Affiliation: (1) Department of Surgery, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA 02215, USA (2) Chiron Corporation, Emeryville, CA 94608, USA (3) Mailing address: Division of Crit Care Med, 616 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261, USA Article History: Received Date: 28/06/2000 Accepted Date: 30/04/2001 Article note: Final revision received: 4 April 2001 Electronic Publication