Clinical Implications of Crosstalk Between Coagulation and Inflammation: The Role of Anticoagulation in Treating Sepsis

BACKGROUND:

Despite decades of active investigation, sepsis remains one of the leading causes of mortality worldwide. Multiple lines of evidence have illustrated that up-regulation of the activated Factor VII (FVIIa)/Tissue Factor (TF) complex, and its downstream extrinsic coagulation cascade, are major contributors to coagulopathies and inflammatory response during sepsis. For example, decreased mortality and inflammatory responses during sepsis were observed in mice with significantly reduced FVII expression. Another recent study demonstrated the association of increased mortality with higher levels of FVIIa in septic patients. Similar results have been demonstrated for Factor X (FX) and thrombin. In addition, several studies have been conducted to investigate the role of heparin in treating sepsis and have yielded promising results, however, the exact mechanisms remain elusive, and the clinical implications of crosstalk between coagulation pathways and sepsis are yet to be determined. Furthermore, the role of vitamin-K antagonist in sepsis has not been investigated.

OBJECTIVE:

To assess the effects of pre-existing anticoagulation with warfarin on the clinical course of septic patient.

METHODS:

This was a retrospective observational study undertaken in a community-based teaching hospital. Patients who were admitted with a primary diagnosis of sepsis from January 01 to June 30, 2012 were included in the study. The clinical characteristics between patient groups without and with prior anticoagulation were compared and analyzed. The primary outcomes include the severity of sepsis, length of hospitalization, and mortality rate during hospitalization.

RESULTS:

A total of 134 septic patients were included in the study. Among them, 105 patients were not anticoagulated, while 29 patients were anticoagulated, prior to admission (mean age: 76.0 + 1.2 vs. 77.5 + 2.6, p = 0.603). All of the patients with anticoagulation had been taking warfarin due to either pre-existing atrial fibrillation (79.3%) or deep vein thrombosis/pulmonary embolism (20.7%). There were significant differences in International Normalized Ratio (INR) of prothrombin time between groups without and with anticoagulation at the time of admission (1.28 + 0.04 vs. 4.59 + 0.83, p < 0.001). Septic patients who did not take warfarin prior to admission presented with higher Sepsis Indices (0.93 + 0.03 vs. 0.82 + 0.05, p < 0.05), resulting in longer hospitalizations (11.60 + 1.02 vs. 8.40 + 0.70, p < 0.001). The overall all-cause mortality rates during the hospitalization between patients without anticoagulation and those with anticoagulation were 23% vs. 14%, respectively.

CONCLUSION:

To our knowledge, this is the first study to demonstrate that septic patients with prior anticoagulation by a vitamin-K antagonist presented with less severity of sepsis, reduced length of hospital stay, and decreased all-cause mortality during hospitalization as compared with those without anticoagulation. In our study, prior administration of anticoagulation with warfarin may have had significant clinical implications in septic patients. This warrants further prospective studies.