Stacey Chen, MD
Surgical Resident
NYU Langone Health, United States
Disclosure information not submitted.
Brendan Wu, DMD MMSc
Medical Student
NYU Langone Health, United States
Disclosure information not submitted.
Zachary Kon, MD
Associate Professor
Northwell Health, United States
Disclosure information not submitted.
Philip Sommer, MD
Clinical Assistant Professor of Anesthesiology and Critical Care
NYU Langone Medical Center
New York, New York
Disclosure information not submitted.
Greta Piper, MD
Associate Professor of Surgery
NYU Langone Health, United States
Disclosure information not submitted.
Deane Smith, MD
Director, Extracorporeal Membrane Oxygenation Program
NYU Langone Health, United States
Disclosure information not submitted.
Title: Potential Predictors of Circuit Exchange in COVID-19 Patients Receiving ECMO Support
Introduction: An extracorporeal membrane oxygenation (ECMO) circuit exchange is a commonly encountered procedure with potential for significant hemodynamic risks in patients with already compromised physiology. We aim to examine potential predictors and safety of circuit exchanges (CE) in COVID-19 patients receiving veno-venous (VV) ECMO support.
Methods: This is a single-center, retrospective study of 29 patients with COVID-19 infection requiring VV ECMO support between March 2020 and April 2020. Pre-ECMO characteristics, complications, and blood product requirements were compared between patients with and without CE using independent two-sample Student’s t-tests for continuous variables and chi-squared test for categorical variables.
Results: Overall survival to discharge was 97% (28/29). 13 patients did not require any CE and 16 patients required a total of 42 CEs. There was no difference in baseline characteristics, pre-ECMO ventilator settings, SOFA, or VIS scores between patients who required a circuit exchange versus those who did not. However, patients who required circuit exchanges had a higher lactate at time of cannulation (1.83 vs 1.32, p=0.011). The CE group had a longer duration of ECMO support (53.31 vs 13.31 days, p=0.004), higher pulmonary complications defined as hemothorax or pneumothorax (8/16 vs 0/13, p=0.003), and higher blood product transfusions including packed red blood cell (5,525 mL vs 1,261 mL, p=0.0001), cryoprecipitate (1,735 mL vs 501 mL, p=0.004), and platelet (486 mL vs 111 mL, p=0.019). In the 42 CEs, no major complications, including hemodynamic instability or cardiac arrest, occurred during the exchange.
Conclusions: Severity of underlying lung injury, ECMO duration, and pre-cannulation lactate may be markers for potential CE in COVID-19 patients requiring ECMO. They may be associated with higher blood product transfusion requirements and while CEs are procedures performed in hemodynamically tenuous patients, they are able to be performed safely.