Endovascular Resuscitation

Abstract

Haemorrhage is the leading cause of preventable death in trauma. The key in management of patients with haemorrhagic shock is timely haemorrhage control before the patient dies from exsanguination. However, this can be challenging, particularly in non-compressible torse haemorrhage leading to delays and eventual exsanguination and death. Despite significant advances in trauma care and resuscitation strategies over the past few decades, mortality related to traumatic haemorrhage remains very high. Endovascular resuscitation is an emerging concept of catheter-based techniques designed to control haemorrhage, support central blood pressure, and tissue perfusion. While Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA) and related endovascular techniques present optimistic treatment solutions, early clinical data do not show a clear benefit and there are significant risks related to these techniques such as arterial access complications and ischaemia reperfusion injury. It is unknown who would most benefit from these techniques and there is limited knowledge of precise physiology of exsanguination to allow optimal management strategies. The aims of this thesis are to analyse the effectiveness and safety of Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA) and explore the physiology of exsanguination. We used data from a local trauma registry to study outcomes in trauma patients treated with REBOA at a high-volume centre. We applied propensity score matching to compare patients treated with REBOA to those who were not. Our study demonstrated improved in- hospital, and 30- day survival in REBOA patients. Using a local REBOA registry data from a large volume centre we also compared trauma patients who were treated with partial and complete occlusion REBOA as well as total occlusion time less than and more than 30 min. This study showed that prolonged REBOA was associated with increased mortality. Surviving patients treated with partial occlusion required less need for vasopressors. iii With safety concerns of blind REBOA inflation, ex-vivo porcine aortic tissues were used to compare inflation parameters of compliant and semi-compliant balloons. We found that the latter allows for concomitant use with a safety valve which would prevent any overinflation. Finally, a swine model of controlled haemorrhage was used to define two distinct physiological phases of exsanguination cardiac arrest which have implications for treatment: preload support or coronary perfusion. This model was later used to study the relationship between the duration of cardiac arrest and the likelihood of successful resuscitation with Selective Aortic Arch Perfusion (SAAP). This showed that SAAP can accomplish return of spontaneous circulation even after 10 minutes of unsupported cardiac arrest due to haemorrhage.