Preliminary findings on how ECMOdisrupts haemostasis: an ovine model

Background: Extracorporeal membrane oxygenation (ECMO)is a vital treatment modality for patients with respiratory failurerefractory to standard ventilation. However, the extended con-tact of patients’ blood with an artificial circuit, routinely leads tobleeding and clotting complications. Defining what factors disrupthaemostasis is difficult in ECMO patients because of their com-plex underlying pathologies. We sought to overcome this withan in vivo ovine ECMO model.Methods: Ten healthy female merino sheep were placed on veno-venous ECMO. Seven were supported for 2hr, and 3 for 24hr. Sam-ples were collected for FBC, routine coagulation (PT, APTT, ClaussFibrinogen) and rotational thromboelestometry (ROTEM), pre-ECMO (0 hr) and post-ECMO (0.25, 1, 2, 6, 7, 12, 18 and 24 hr).Results: ECMO induced a quick and significant decrease inEXTEM maximum clot firmness (MCF) (Fig A) which enduredfor 2 hours. Correspondingly PT increased initially but was tem-poral (Fig B). Fibrinogen levels decreased initially but not signi-ficantly (Fig D), this was reflected in FIBTEM MCF (Fig C). Nosignificant changes to INTEM or platelet number and functionwere detected (data not shown).Discussion and conclusions: This preliminary data confirmsthat even in a healthy host, ECMO disrupts haemostasis imme-diately. The EXTEM data correlated with the PT, but may bemore sensitive as it detected a decreased MCF over a longer period. The decrease in EXTEM MCF suggests that the qualityof the clot is compromised.As the platelet count and function remained unchanged this mayreflect the fall in fibrinogen. These haemostatic imbalances war-rant further investigation