Bacterial sepsis is frequently accompanied by intravascular activation of coagulation and fibrinolysis, resulting in a coagulopathy with thrombotic, hemorrhagic, and inflammatory components (disseminated intravascular coagulation - DIC) that contributes significantly to morbidity and mortality. Recently, using vascular occlusion models, the plasma coagulation protease factor XI (FXI) has been shown to contribute to pathologic thrombus formation in primates, mice, and rabbits. These results are supported by epidemiologic studies in human populations that show an association between elevated plasma FXI levels and an increased risk for venous and arterial thrombosis. FXI has also been shown to contribute to the lethal consumptive coagulopathy in protein C deficient mice. To investigate the contribution of FXI to consumptive coagulopathy and mortality in sepsis, we compared wild type (WT) and FXI deficient (FXI−/−) mice in models of polymicrobial sepsis and endotoxemia. Polymicrobial sepsis was induced in WT and FXI−/− C57Bl/6 mice (n = 24 for each genotype) using an established model of cecal ligation and puncture (CLP). Overall survival was 21% for WT and 42% for FXI−/− mice (p < 0.01), with WT deaths occurring between days 2 and 4, and FXI−/− deaths between days 3 and 6. Average survival for mice succumbing to sepsis was 55 ± 10 hr for WT, and 88 ± 22 hr in FXI−/− mice (p < 0.01). 24 hrs after CLP, platelet count in WT and FXI−/− animals decreased 27 ± 13% and 13 ± 12% (p = 0.05), and leukocyte counts decreased by 64 ± 13% and 40 ± 23% (p < 0.01), respectively. Fibrinogen levels 24 hours post-CLP decreased slightly in WT mice (37 ± 34% from baseline), while increasing 212 % ± 74% in FXI−/− mice (p < 0.01). Mice were also challenged with intraperitoneal Escherichia coli endotoxin. Mortality 24 hrs after endotoxin administration was identical for WT and FXI−/− mice (n =15 for each dose) at doses of 50 mg/kg (mortality 80%) and 25 mg/kg (mortality 70%). Twelve hours after administering 25 mg/kg endotoxin, the mean platelet counts in WT animals decreased by an average of 60% (from 440 ± 159/mm3 at baseline to 175 ± 77/mm3, n = 10), and by 36% in FXI−/− animals (from 474 ± 218/mm3 to 303 ± 100/mm3). In summary, prognosis was better for FXI−/− mice in CLP-induced sepsis compared to WT mice, and the CLP and endotoxin data indicate that consumptive coagulopathy is less severe in the absence of FXI. The results suggest that therapeutic inhibition of FXIa activity could be beneficial in treating sepsis-related DIC. Humans with severe congenital FXI deficiency typically have, at most, a mild trauma-induced bleeding disorder. It is possible, therefore, that inhibition of FXIa in DIC may be associated with a lower risk of exacerbating bleeding when compared to anticoagulant therapies such as heparin or activated protein C.

Disclosures: I receive consultation fees from Bristol-Myers Squibb for work related to anticoagulation therapies.

Author notes

*

Corresponding author

Sign in via your Institution