Effect of hypofibrinolysis on clinical outcomes of patients with septic disseminated intravascular coagulation

Abstract

Background: This study investigated the utility of thromboelastometry (ROTEM) in assessing hypofibrinolysis among septic patients, specifically the association of hypofibrinolysis, as determined by ROTEM, with septic disseminated intravascular coagulation (DIC), organ dysfunction, and clinical outcomes.

Methods: This single-center, retrospective analysis included adult septic patients admitted to Saga University Hospital from 2013 to 2017, with available ROTEM data. Hypofibrinolysis was assessed using the lysis index at 60 min (LI60) in extrinsic thromboelastometry (EXTEM). Based on their LI60 values, patients were classified into three groups: Hyper (LI60 ≤ 85), Normal (LI60 86-96), and Hypo (LI60 ≥ 97).

Results: Among the 63 cases analyzed, the Hypo group showed significantly higher APACHEII and SOFA scores than the Normal group, indicating greater disease severity. Similarly, DIC and sepsis-induced coagulopathy (SIC) scores were notably higher in the Hypo group. The diagnostic performance of LI60 for ISTH-overt DIC showed an area under the curve (AUC) of 0.954, with an optimal cutoff value of 97 %, achieving 100 % sensitivity and 83.3 % specificity. The odds ratio for ISTH-overt DIC was 2.894, indicating a strong association between elevated LI60 and occurrence of DIC. Hypofibrinolysis predicted 28-day mortality and high SOFA scores (≥ 10) with high specificity and negative predictive value (NPV). A Kaplan-Meier curve revealed that the Hypo Group showed significantly worse clinical outcomes than the Normal and Hyper groups.

Conclusion: For septic patients, fibrinolysis suppression presenting as "hypofibrinolysis" (elevated LI60) is associated with poor prognosis and risk of higher organ dysfunction. Moreover, it is a significant predictor of adverse clinical outcomes in sepsis.

Keywords: Disseminated intravascular coagulation; Hypofibrinolysis; Organ dysfunction; Sepsis; Thromboelastometry.