External Validation of the Two Newly Proposed Criteria for Assessing Coagulopathy in Sepsis

Abstract

Background: Two different criteria for evaluating coagulopathy in sepsis were recently released: sepsis-induced coagulopathy (SIC) and sepsis-associated coagulopathy (SAC). Although both use universal haemostatic markers of platelet count and pro-thrombin time, significance and usefulness of these criteria remain unclear.

Objective: This article validates and evaluates the significance of SIC and SAC criteria compared with the International Society on Thrombosis and Haemostasis (ISTH) overt disseminated intravascular coagulation (DIC) and Japanese Association for Acute Medicine (JAAM) DIC criteria.

Methods: Clinical characteristics of patients from a nationwide Japanese cohort were classified by SIC, SAC or DIC status and relations between criteria were examined. We evaluated associations between in-hospital mortality and anticoagulant therapy according to the SIC, SAC or DIC status to clarify the significance of criteria for introducing anticoagulants. Intervention effects were analysed by Cox regression analysis adjusted by propensity scoring.

Results: Incidences of coagulopathy diagnosed by SIC and JAAM DIC were similar, whereas those of SAC and ISTH overt DIC were about half of the former two (61.4%, 60.8% vs. 45.3%, 29.3%). Severity and mortality of all criteria were almost comparable. For validating initiation of anticoagulation, favourable effects of anticoagulant therapy were observed only in sub-sets with, and not without, coagulopathy diagnosed by all four criteria. Slight non-significant differences between anticoagulant groupings were found in ISTH overt DIC- and SAC-negative populations, suggesting that some patients even 'without' these criteria may benefit from anticoagulant therapy.

Conclusion: Newly developed SIC diagnostic criteria for coagulopathy may be valuable in detecting appropriate candidates for anticoagulant therapy in sepsis and a useful alternative to conventional DIC scoring systems.