Predictors of Recovery in Patients Supported With Venoarterial Extracorporeal Membrane Oxygenation for Acute Massive Pulmonary Embolism

Abstract

Venoarterial extracorporeal membrane oxygenation (VA-ECMO) has emerged as a promising initial support strategy for acute massive pulmonary embolism. However, it remains unclear which patients will ultimately require surgical pulmonary embolectomy (SPE) vs anticoagulation alone. All consecutive patients (2015-2018) with confirmed massive PE, placed on VA-ECMO utilizing a protocolized approach, were reviewed. Per protocol, patients were supported for 3 to 5 days before reevaluation of right ventricular (RV) function via echocardiography. If RV function recovered, VA-ECMO was discontinued with no further intervention (no-SPE group). In patients with persistent RV dysfunction, SPE was performed. Forty-five patients were identified, and 41 patients were treated per protocol. Seventy-three percent responded to anticoagulation alone, and 27% required SPE. Factors associated with SPE rather than decannulation with anticoagulation alone included prolonged shortness of breath, elevated N-terminal prohormone of brain natriuretic peptide, enlarged pulmonary artery diameter, and history of venous thromboembolism. A predictive algorithm was developed with a negative predictive value of 97% and a specificity of 97% for a low-risk score, and a positive predictive value of 100% and sensitivity of 67% for a high-risk score. Overall, 90-day survival was 97% in the no-SPE group and 100% in the SPE group. In this cohort, greater than 70% of patients who presented with massive PE and supported with VA-ECMO ultimately recovered with anticoagulation alone. Specific risk factors, likely related to thrombus chronicity, may be associated with lack of RV recovery, and can be utilized for consideration of early surgical intervention to minimize VA-ECMO duration.