Chhagan Bihari1, Anupama Patil1, Saggere Muralikrishna Shasthry2, Sukriti Baweja3, Guresh Kumar4, Shiv Kumar Sarin5,6. 1. Department of Pathology and Hematology, Institute of Liver and Biliary Sciences, D1, Vasant Kunj, New Delhi, India. 2. Department of Hepatology, Institute of Liver and Biliary Sciences, D1, Vasant Kunj, New Delhi, 11007, India. 3. Department of Molecular and Cellular Medicine, Institute of Liver and Biliary Sciences, D1, Vasant Kunj, New Delhi, India. 4. Department of Clinical Research, Institute of Liver and Biliary Sciences, D1, Vasant Kunj, New Delhi, India. 5. Department of Hepatology, Institute of Liver and Biliary Sciences, D1, Vasant Kunj, New Delhi, 11007, India. shivsarin@gmail.com. 6. Department of Molecular and Cellular Medicine, Institute of Liver and Biliary Sciences, D1, Vasant Kunj, New Delhi, India. shivsarin@gmail.com.
Abstract
BACKGROUND AND AIMS: Coagulopathic bleeding risk prediction is challenging in decompensated cirrhosis (DC) by conventional assays. Viscoelastic tests (VETs) are likely to be more useful for assessing coagulation status in these patients. We investigated whether the VET (Sonoclot) parameters with fibrinogen could predict coagulopathic bleeding in these patients. PATIENTS AND METHODS: Coagulation parameters studied in 874 patients [124 compensated cirrhosis (CC), 521 DC, and 229 acute-on-chronic liver failure (ACLF)] and 190 controls. DC patients were enrolled in derivation (n = 266) and validation (n = 255) cohorts. Sonoclot variables [activated clotting time (ACT), clot rate (CR), platelet function (PF), time to peak (TP) and peak amplitude (PA)] and fibrinogen levels were measured. Coagulopathic bleeding was recorded along with 1-year survival. RESULTS: DC patients had prolonged ACT (p < 0.001), depressed CR (p = 0.059), reduced PF (p = 0.09), longer TP (p < 0.001) and smaller PA (p < 0.001), compared to CC and controls (p < 0.001 each). In derivation cohort, 32.3% patients had coagulopathic bleeding. Cox regression analysis of derivation cohort revealed; ACT > 190 s, PF < 1.25 and fibrinogen < 1.2 g/l could predict coagulopathic bleeding and were used to develop a bleeding risk score. In validation cohort; this score was comparable, correlated to real events, and had similar bleed free events with time. The score was also useful in predicting bleed in ACLF patients. In DC patients, 1-year mortality was higher those who bled and received transfusions. CONCLUSION: Viscoelasticity-based bleeding risk score using ACT, PF and fibrinogen, predicts coagulopathic bleeding in DC patients and should be useful in rationalizing transfusion of blood products. We designed a viscoelastic test-based bleeding risk score which is useful in advanced liver disease to predict the coagulation-related bleeding. This figure shows the lower bleeding-free events in advanced cirrhosis with each incremental bleeding risk score.
BACKGROUND AND AIMS: Coagulopathic bleeding risk prediction is challenging in decompensated cirrhosis (DC) by conventional assays. Viscoelastic tests (VETs) are likely to be more useful for assessing coagulation status in these patients. We investigated whether the VET (Sonoclot) parameters with fibrinogen could predict coagulopathic bleeding in these patients. PATIENTS AND METHODS: Coagulation parameters studied in 874 patients [124 compensated cirrhosis (CC), 521 DC, and 229 acute-on-chronic liver failure (ACLF)] and 190 controls. DC patients were enrolled in derivation (n = 266) and validation (n = 255) cohorts. Sonoclot variables [activated clotting time (ACT), clot rate (CR), platelet function (PF), time to peak (TP) and peak amplitude (PA)] and fibrinogen levels were measured. Coagulopathic bleeding was recorded along with 1-year survival. RESULTS: DC patients had prolonged ACT (p < 0.001), depressed CR (p = 0.059), reduced PF (p = 0.09), longer TP (p < 0.001) and smaller PA (p < 0.001), compared to CC and controls (p < 0.001 each). In derivation cohort, 32.3% patients had coagulopathic bleeding. Cox regression analysis of derivation cohort revealed; ACT > 190 s, PF < 1.25 and fibrinogen < 1.2 g/l could predict coagulopathic bleeding and were used to develop a bleeding risk score. In validation cohort; this score was comparable, correlated to real events, and had similar bleed free events with time. The score was also useful in predicting bleed in ACLF patients. In DC patients, 1-year mortality was higher those who bled and received transfusions. CONCLUSION: Viscoelasticity-based bleeding risk score using ACT, PF and fibrinogen, predicts coagulopathic bleeding in DC patients and should be useful in rationalizing transfusion of blood products. We designed a viscoelastic test-based bleeding risk score which is useful in advanced liver disease to predict the coagulation-related bleeding. This figure shows the lower bleeding-free events in advanced cirrhosis with each incremental bleeding risk score.