Cardiac Output From Right Ventricular Pressure: Not the Solution for Venovenous Extracorporeal Membrane Oxygenation

Cardiac output (CO) monitoring is challenging during venovenous extracorporeal membrane oxygenation (VV ECMO), as thermodilution techniques are unreliable. Cardiac output from right ventricular pressure (RVP) waveforms has recently been considered. We evaluated whether RVP-derived CO (CO RV ) accurately reflects cardiac output measured by echocardiography (CO US ) during VV ECMO. Ten anesthetized pigs underwent femoro-jugular VV ECMO implantation. A Swan-Ganz IQ catheter provided continuous CO RV monitoring throughout the experiment. Cardiac output was modulated using either dobutamine (5–20 µg/kg/min) or landiolol (10–40 µg/kg/min) infusions. Agreement between CO RV and CO US was assessed using Pearson correlation and Bland–Altman analysis. Overall, 596 paired CO measurements were performed in seven animals. Cardiac output ranged from 1.05 to 9.46 L/min. Correlation between CO RV and CO US was poor ( R = 0.24; R 2 = 0.057; p < 0.05). Bland–Altman analysis showed a mean bias of +1.94 L/min with wide limits of agreement (−1.44 to +5.32 L/min). Correlations between individual right ventricular pressure waveform components and CO US were heterogeneous and weak to moderate (diastolic RVP, R = 0.36; right ventricle +d P /d t [delta pressure/delta time] R = 0.40; systolic RVP R = 0.60). CO RV is not reliable during VV ECMO in our experiment. Further studies are needed to assess CO RV performances.