Comparison between regional citrate anticoagulation and systemic heparin anticoagulation in patients undergoing continuous renal replacement therapy and venoarterial extracorporeal membrane oxygenation: A retrospective multicentric study

Introduction: Regional citrate anticoagulation (RCA) is the recommended first-line strategy for continuous renal replacement therapy (CRRT) circuits. Acute kidney injury is common in patients undergoing venoarterial extracorporeal membrane oxygenation (VA-ECMO), and CRRT is frequently required. However, the use of RCA in this setting remains debated due to concerns about metabolic complications and limited evidence. This study aimed to compare the efficacy and safety of RCA versus unfractionated heparin (UFH) anticoagulation in this high-risk population. Materials and methods: In this retrospective multicentric study, we included adult patients receiving simultaneous VA-ECMO and CRRT between January 2019 and December 2021 in four French intensive care units. The primary outcome was CRRT filter thrombosis, defined as unplanned circuit cessation due to clotting. Filter lifespan, defined as the time from CRRT initiation to cessation for any reason, was recorded as a secondary outcome. Additional secondary outcomes included bleeding events, transfusion requirements, coagulation parameters, and metabolic complications. Association of CRRT filter thrombosis with anticoagulation modality was assessed using a Fine-Gray regression model accounting for competing events, and adjusted for coagulation parameters and CRRT site. Missing data were handled using multiple imputation. Results: Among the 253 patients included (77.9% male, median age 61 years), 599 CRRT sessions were analyzed (RCA: 106; UFH: 493). Median filter lifespan did not differ significantly between groups (RCA: 47 [IQR 21-90] vs UFH: 44 [IQR 20-72] hours, p = 0.144). The estimated cumulative incidence of filter thrombosis at 72 h was 17.0% (95% CI, 10.5-24.8%) in the RCA group versus 26.6% (95% CI, 22.8-30.5%) in the UFH group; p = 0.033. RCA was associated with a significantly lower risk of filter thrombosis compared to UFH (aHR = 0.35 (95% CI, 0.21-0.59; p < 0.001)), after adjustment for coagulation parameters and CRRT site. RCA was not associated with decreased bleeding events (33.3% vs 37.6%, p = 0.417) and did not increase the incidence of severe metabolic complications. Conclusion: In patients receiving VA-ECMO and CRRT, RCA was safe and effective. RCA was associated with a significantly lower risk of filter thrombosis, despite similar median filter lifespan, without increasing bleeding or severe metabolic complications. These findings support the use of RCA as a promising anticoagulation strategy in this high-risk population. Specialty: Critical Care.