Prospective observational study of hemostatic alterations during adult extracorporeal membrane oxygenation (ECMO) using point-of-care thromboelastometry and platelet aggregometry.

Objectives: To characterize the longitudinal hemostatic profile during adult ECMO using point-of-care tests (POCT) for coagulation and to compare these parameters to standard laboratory tests. In addition, the clinicians' responses during bleeding episodes using available information were compared to a POCT-based response.
Design: Prospective observational cohort study.
Setting: ECMO-referral center in a university teaching hospital.
Participants: Ten critically ill adult ECMO patients.
Interventions: Daily laboratory coagulation profile, transfusion history and near-daily thromboelastometry (ROTEM®) and platelet aggregometry (Multiplate®).
Main Results: Six male and four female patients, seven with VA- and three with VV-ECMO were studied over 110 days. Seventy-five thromboelastometry (TEM) and 36 platelet aggregometry (MEA) results were analyzed. A majority of TEM values were within the normal range, except for FIBTEM (majority high), which remained consistent over long (>5 days) ECMO runs. In MEA there were low values, particularly in the adenosine diphosphate- and ristocetin-induced assay, implying possibly a vWF-factor or GpIb-receptor defect. There was correlation between laboratory and POCT as well as good correlation between the clot firmness after 10 minutes (A10) and the maximum clot firmness in ROTEM, suggesting that reliable information can be obtained within 15 minutes. Twenty-two bleeding episodes were observed in five patients. When comparing the clinicians' response to a transfusion algorithm based on POCT, there was a concordance in less than 20% of episodes.
Conclusions: POCT for coagulation can provide specific, reliable, and timely information during bleeding episodes and the use of targeted therapy algorithms could improve outcomes and reduce costs.
(Crown Copyright © 2015. Published by Elsevier Inc. All rights reserved.)