LAUSR

BACKGROUND AND OBJECTIVES Due to rebalanced haemostasis in cirrhosis, viscoelastometric testing (VET) is more accurate than standard coagulation tests (SCTs) in preprocedural haemostatic evaluation, resulting in decreased unnecessary transfusion. We aimed to determine the impact of VET-guided strategy on postprocedural bleeding, periprocedural transfusion rates and quantities, transfusion-related adverse events (TRAEs), lengths of stay (LOS) and mortality from randomized controlled trials (RCTs) of cirrhotic patients. METHODS PubMed and EMBASE were searched for RCTs comparing VET-guided with SCT-guided transfusion in cirrhotic adults undergoing esophagogastroduodenoscopy, liver transplantation or other invasive interventions. Using random-effects models, the pooled risk ratios (RRs) and/or mean differences (MDs) of postprocedural bleeding-free events and the other outcomes were estimated alongside 95% confidence intervals (CIs). RESULTS Of seven included RCTs (n = 421; 72.2% men; mean age 49.1 years), VET-guided transfusion did not change postprocedural bleeding-free statuses (RR 1.05; 95% CI 0.94-1.17). However, VET-based algorithms decreased the rates of fresh frozen plasma (FFP; RR 0.52; 95% CI 0.35-0.77) and platelet transfusions (RR 0.34; 95% CI 0.16-0.73), the quantities of transfused FFP (MD -1.39 units; 95% CI -2.18 to -0.60), platelets (MD -1.06 units; 95% CI -2.01 to -0.12) and cryoprecipitate (MD -7.13 units; 95% CI -14.20 to -0.07) and the risk of TRAEs (RR 0.42; 95% CI 0.27-0.65). The overall mortality rates and LOS were not significantly different between two groups. CONCLUSION Compared with conventional SCT-guided, VET-guided strategy decreases periprocedural plasma and platelet transfusions and TRAEs, without increasing haemorrhagic complications, LOS or mortality in cirrhosis.