Performance of a Prospective Anticoagulation Stratification Algorithm After Liver Transplantation

Abstract

Background. Venous thromboembolism (VTE) occurs in 0.4% to 15.5% and bleeding occurs in 20% to 35% of patients after liver transplantation (LT). Balancing the risk of bleeding from therapeutic anticoagulation and risk of thrombosis in the postoperative period is challenging. Little evidence exists regarding the best treatment strategy for these patients. We hypothesized that a subset of LT patients who develop postoperative deep vein thromboses (DVTs) could be managed without therapeutic anticoagulation. We implemented a quality improvement (QI) initiative using a standardized Doppler ultrasound-based VTE risk stratification algorithm to guide parsimonious implementation of therapeutic anticoagulation with heparin drip. Methods. In a prospective management QI initiative for DVT management, we compared 87 LT historical patients (control group; January 2016–December 2017) to 182 LT patients (study group; January 2018–March 2021). We analyzed the rates of immediate therapeutic anticoagulation after DVT diagnosis within 14 d of LT, clinically significant bleeding, return to the operating room, readmission, pulmonary embolism, and death within 30 d of LT before and after the QI initiative. Results. Ten patients (11.5%) in the control group and 23 patients (12.6%; P = 0.9) in the study group developed DVTs after LT. Immediate therapeutic anticoagulation was used in 7 of 10 and 5 of 23 patients in the control and study groups, respectively (P = 0.024). The study group had lower odds of receiving immediate therapeutic anticoagulation after VTE (21.7% versus 70%; odds ratio = 0.12; 95% confidence interval, 0.019-0.587; P = 0.013) and a lower rate of postoperative bleeding (8.7% versus 40%; odds ratio = 0.14, 95% confidence interval, 0.02-0.91; P = 0.048). All other outcomes were similar. Conclusions. Implementing a risk-stratified VTE treatment algorithm for immediate post-LT patients appears to be safe and feasible. We observed a decrease in the use of therapeutic anticoagulation and a lower rate of postoperative bleeding without adverse impacts on early outcomes.