Purpose: to examine the impact of coagulation abnormalities on survival and analyze the performance of routine coagulation tests to predict clinical outcomes in intention to improve the survival in intensive care unit (ICU) patients by the using of routine coagulation tests.
Methods: a prospective double cohort study during 10 months. All patients admitted were included. Those with congenital coagulation disorder and curative anticoagulation were excluded. Daily monitoring of blood count and standard coagulation tests were performed. Kaplan-Meier curves, Logistic regression and ROC curves were used for analysing the impact of coagulation disorders and assessment of usefulness of routine coagulation tests.
Results: Mortality was significantly higher in the coagulation disorders group (55% vs 14%, p <0.001) and disseminated intravascular coagulation (DIC) was an independent factor related to mortality (OR=3.37, CI95% [1.85-4.90], p=0.012). The leading cause of mortality was ischemic or bleeding complications (56%). We found that prolonged clotting time had significant sensitivity and negative predictive value (NPV) (94.7% and 98% respectively) to predict ischemia, whereas thrombocytopenia had a NPV at 91.4% to predict bleeding and hyper D dimer/hypofibrinogen have specificity at 91.6% to predict bleeding. An ISTH score > 3.5 was significantly related to death with a specificity of 98%, a PPV of 90% and a robust likelihood ratios: LR+=17 and LR-=0.66.
Conclusion: coagulation disorders were significantly associated to mortality. The conventional coagulation tests can be useful for the early detection and management of ischemic/bleeding events.
