RESUMO
(1) Background: Extracorporeal membrane oxygenation (ECMO) is a complex procedure affecting both the risk of thrombosis and bleeding. High-quality data to personalize anticoagulation management in ECMO are lacking, resulting in a high variability in practice among centers. For this reason, we review coagulation methods and monitoring and share a pragmatic proposal of coagulation management, as performed in our high-volume ECMO Referral Centre; (2) Methods: We revised the anticoagulation options and monitoring methods available for coagulation management in ECMO through PubMed search based on words including "anticoagulation," "coagulation assays," "ECMO," "ELSO," and "ISTH"; (3) Results: Actual revision of the literature was described as our routine practice regarding ECMO anticoagulation and monitoring; (4) Conclusions: No coagulation test is exclusively predictive of bleeding or thrombotic risk in patients undergoing ECMO support. An approach that allows for a tailored regimen of anticoagulation (regardless of agent used) and monitoring is mandatory. To accomplish this, we propose that the titration of anticoagulation therapies should include multiple laboratory tests, including anti-Xa, aPTT, ACT, viscoelastic tests, AT levels, platelet count, fibrinogen, and FXIII levels. Anticoagulation regimens should be tailored to a specific patient and personalized based on this complex array of essays.
RESUMO
OBJECTIVES: Activated clotting time (ACT) is commonly used to monitor anticoagulation during cardiac surgeries. Final ACT values may be essential to predict postoperative bleeding and transfusions, although ideal values remain unknown. Our aim was to evaluate the utility of ACT as a predictor of postoperative bleeding and transfusion use. METHODS: Retrospective study (722 patients) submitted to surgery between July 2018-October 2021. We compared patients with final ACT < basal ACT and final ACT ≥ basal ACT and final ACT < 140 s with ≥140 s. Continuous variables were analysed with the Wilcoxon rank-sum test; categorical variables using Chi-square or Fisher's exact test. A linear mixed regression model was used to analyse bleeding in patients with final ACT < 140 and ≥140. Independent variables were analysed with binary logistic regression models to investigate their association with bleeding and transfusion. RESULTS: Patients with final ACT ≥ 140 s presented higher postoperative bleeding than final ACT < 140 s at 12 h (P = 0.006) and 24 h (**P = 0.004). Cardiopulmonary bypass (CPB) time [odds ratio (OR) 1.009, 1.002-1.015, 95% confidence interval (CI)] and masculine sex (OR 2.842,1.721-4.821, 95% CI) were significant predictors of bleeding. Patients with final ACT ≥ 140 s had higher risk of UT (OR 1.81, 1.13-2.89, 95% CI; P = 0.0104), compared to final ACT < 140 s. CPB time (OR 1.019,1.012-1.026, 95% CI) and final ACT (OR 1.021,1.010-1.032, 95% CI) were significant predictors of transfusion. Female sex was a predictor of use of transfusion, with a probability for use of 27.23% (21.84-33.39%, 95% CI) in elective surgeries, and 60.38% (37.65-79.36%, 95% CI) in urgent surgeries, higher than in males. CONCLUSIONS: Final ACT has a good predictive value for the use of transfusion. Final ACT ≥ 140 s correlates with higher risk of transfusion and increased bleeding. The risk of bleeding and transfusion is higher with longer periods of CPB. Males have a higher risk of bleeding, but females have a higher risk of transfusion.