Objective Recent clinical studies have shown that anticoagulant therapy might be effective only in specific at-risk subgroups of patients with sepsis and coagulation dysfunction. The definition of sepsis was recently modified, and as such, old scoring systems may no longer be appropriate for the diagnosis of sepsis-associated coagulopathy. The aim of this study was to evaluate prognostic factors in patients diagnosed with sepsis and coagulopathy according to the new sepsis definition and assess their accuracy in comparison with existing models.

Design Retrospective analysis of the nationwide survey for recombinant human soluble thrombomodulin.

Setting General emergency and critical care centres in secondary and tertiary care hospitals.

Participants We evaluated the prognostic value of the newly proposed diagnostic criteria for sepsis-induced coagulopathy (SIC). A total of 1498 Japanese patients with sepsis and coagulopathy complications who were treated with recombinant thrombomodulin were analysed in this study.

Main outcome measures The platelet count, prothrombin time (PT) ratio, fibrinogen/fibrin degradation products, systemic inflammatory response syndrome score and Sequential Organ Failure Assessment (SOFA) score obtained just before the start of treatment were examined in relation to the 28-day mortality rate.

Results The platelet count, PT ratio and total SOFA were independent predictors of a fatal outcome in a logistic regression model. A SIC score was defined using the three above-mentioned variables with a positivity threshold of 4 points or more. The SIC score predicted higher 28-day mortality rate compared with the current Japanese Association for Acute Medicine-disseminated intravascular coagulation score (38.4%vs34.7%).

Conclusion The SIC score is based on readily available parameters, is easy to calculate and has a high predictive value for 28-day mortality. Future studies are warranted to evaluate whether the SIC score may guide the decision to initiate anticoagulant therapy.