Venoarterial Extracorporeal Membrane Oxygenation in High-Risk Pulmonary Embolism: A Case Series and Literature Review.

ABSTRACT

Background: High-risk Pulmonary Embolism (PE) has an ominous prognosis and requires emergent reperfusion therapy, primarily systemic thrombolysis (ST). In deteriorating patients or with contraindications to ST, Veno-Arterial Extracorporeal Membrane Oxygenation (VA-ECMO) may be life-saving, as supported by several retrospective studies. However, due to the heterogeneous clinical presentation (refractory shock, resuscitated cardiac arrest (CA) or refractory CA), the real impact of VA-ECMO in high-risk PE remains to be fully determined. In this study, we present our centre experience with VA-ECMO for high-risk PE. Method: From 2008 to 2020, we analyzed all consecutive patients treated with VA-ECMO for high-risk PE in our tertiary 35-bed intensive care unit (ICU). Demographic variables, types of reperfusion therapies, indications for VA-ECMO (refractory shock or refractory CA requiring extra-corporeal cardiopulmonary resuscitation, ECPR), hemodynamic variables, initial arterial blood lactate and ICU complications were recorded. The primary outcome was ICU survival, and secondary outcome was hospital survival. Results: Our cohort included 18 patients (9F/9M, median age 57 years old). VA-ECMO was indicated for refractory shock in 7 patients (2 primary and 5 following resuscitated CA) and for refractory CA in 11 patients. Eight patients received anticoagulation only, 9 received ST, and 4 underwent surgical embolectomy. ICU survival was 1/11 (9%) for ECPR vs 3/7 (42%) in patients with refractory shock (p = 0.03, log-rank test). Hospital survival was 0/11 (0%) for ECPR vs 3/7 for refractory shock (p = 0.01, log-rank test). Survivors and Non-survivors had comparable demographic and hemodynamic variables, pulmonary obstruction index, and amounts of administered vasoactive drugs. Pre-ECMO lactate was significantly higher in non-survivors. Massive bleeding was the most frequent complication in survivors and non-survivors, and was the direct cause of death in 3 patients, all treated with ST. Conclusions: VA-ECMO for high-risk PE has very different outcomes depending on the clinical context. Furthermore, VA-ECMO was associated with significant bleeding complications, with more severe consequences following systemic thrombolysis. Future studies on VA-ECMO for high-risk PE should therefore take into account the distinct clinical presentations and should determine the best strategy for reperfusion in such circumstances.