Extracorporeal membrane oxygenation improves coagulopathy in an experimental traumatic hemorrhagic model

Abstract

Purpose: Hemorrhage is the most common cause of preventable death after trauma. Coagulopathy plays a central role in uncontrolled bleeding and is caused by multiple factors. Extracorporeal Membrane Oxygenation (ECMO) is an established treatment for patients with respiratory failure and has in recent years also been used in severely injured trauma patients with cardiopulmonary failure and coexisting bleeding shock. The aim of this study was to evaluate the effect of ECMO on hypothermia, acidosis, and coagulopathy in a traumatic hemorrhagic rabbit model.

Methods: After anesthesia and tracheostomy, ten New Zealand White rabbits sustained laparotomy, bilateral femur fractures and were hemorrhaged 45% of their estimated blood volume. After 90 min of hemorrhagic shock they were resuscitated with a standard transfusion protocol together with venoarterial ECMO (n = 5) or with a standard transfusion protocol only (n = 5) for 60 min. No systemic heparin was administered.

Results: ECMO during 60 min of resuscitation significantly increased heart rate (p = 0.01), mean arterial pressure (p = 0.01), body temperature (p = 0.01) and improved the metabolic acidosis, pH (p = 0.01), and lactate (p = 0.01). ECMO also improved the coagulation capacity measured in vitro by Rotational Thromboelastometry with a significant decrease in clot formation time (p < 0.01). This finding was confirmed in vivo with a significant reduction in the animals' ear bleeding time (p < 0.01) and cuticle bleeding time (p < 0.01); 5/5 animals survived in the ECMO group and 3/5 animals survived in the control group.

Conclusions: Heparin-free ECMO stabilizes circulation, improves coagulation, and may impact short-time survival, during the first 60 min, in an experimental traumatic model with severe hemorrhagic shock.

Keywords: Animal model; Coagulopathy; ECMO; Extracorporeal circulation; Hemorrhage; Resuscitation; Shock; Trauma.

Fig. 1

The Rabbit ECMO-Trauma model. The animal sustained laparotomy and bilateral femur fractures and were exsanguinated to class IV Shock. The venoarterial ECMO circuit: Venous draining cannulae in the right atrium (blue), roller pump, heparinized membrane oxygenator for saturation and carbondioxide removal, water heat exchanger (38.5 °C). Blood is reinfused in the descending aorta (red). Bilateral femurfractures are indicated

Fig. 2

Flowchart illustrating the study’s timeline. Induction of anesthesia was at −30 min. All rabbits (n10) sustained laparotomy and femurfractures at baseline, 0 min. During 30 min blood was drained from the IVC. The animals were kept in bleeding shock for another 60 min (goals were mean arterial pressure <20 mmHg, temperature <32 °C, and pH < 7.3). At 90 min resuscitation began with a standard protocol (n = 5) or with a standard protocol and venoarterial ECMO in addition (n = 5). Rotational thromboelastometry and standard coagulation tests were controlled at 0, 90 and 150 min. The animals were killed at 150 min

Fig. 3

The Hematocrit during the study. The exsanguination and transfusion in the two groups were equal. Control (solid line) ECMO (dashed line)

Fig. 4

The animals’ mean arterial pressure during the study. At 90 min after hemorrhagic shock there was no difference in mean arterial pressure between the two groups but after 60 min of resuscitation the mean arterial pressure in the ECMO group was significantly improved (p = 0.01 respectively). Control (solid line) ECMO (dashed line)

Fig. 5

a, b The animals acid–base balance during the study. 90 min after hemorrhagic shock there was no difference in Base Excess (a) or pH (b) between the two groups. After 60 min of resuscitation the Base Excess and the pH were significantly higher in the ECMO group (p = 0.01, respectively). Control (solid line) ECMO (dashed line). c The animals’ body temperature. Initiation of ECMO increased the animals body temperature efficiently and after 60 min the animals in the ECMO group were significantly warmer (>37 °C) than the controls (p = 0.001)