Feasibility of Perflutren Microsphere Contrast Transthoracic Echocardiography in the Visualization of Ventricular Endocardium during Venovenous Extracorporeal Membrane Oxygenation in a Validated Ovine Model

Background Transthoracic echocardiography ( TTE ) during extra corporeal membrane oxygenation ( ECMO ) is important but can be technically challenging. Contrast‐specific TTE can improve imaging in suboptimal studies. These contrast microspheres are hydrodynamically labile structures. This study assessed the feasibility of contrast echocardiography ( CE ) during venovenous ( VV ) ECMO in a validated ovine model. Method Twenty‐four sheep were commenced on VV ECMO . Parasternal long‐axis (Plax) and short‐axis (Psax) views were obtained pre‐ and postcontrast while on VV ECMO . Endocardial definition scores ( EDS ) per segment were graded: 1 = good, 2 = suboptimal 3 = not seen. Endocardial border definition score index ( EBDSI ) was calculated for each view. Endocardial length ( EL ) in the Plax view for the left ventricle ( LV ) and right ventricle ( RV ) was measured. Results Summation EDS data for the LV and RV for unenhanced TTE ( UE ) versus CE TTE imaging: EDS 1 = 289 versus 346, EDS 2 = 38 versus 10, EDS 3 = 33 versus 4, respectively. Wilcoxon matched‐pairs rank‐sign tests showed a significant ranking difference (improvement) pre‐ and postcontrast for the LV (P < 0.0001), RV (P < 0.0001) and combined ventricular data (P < 0.0001). EBDSI for CE TTE was significantly lower than UE TTE for the LV (1.05 ± 0.17 vs. 1.22 ± 0.38, P = 0.0004) and RV (1.06 ± 0.22 vs. 1.42 ± 0.47, P = 0.0.0006) respectively. Visualized EL was significantly longer in CE versus UE for both the LV (58.6 ± 11.0 mm vs. 47.4 ± 11.7 mm, P < 0.0001) and the RV (52.3 ± 8.6 mm vs. 36.0 ± 13.1 mm, P < 0.0001), respectively. Conclusions Despite exposure to destructive hydrodynamic forces, CE is a feasible technique in an ovine ECMO model. CE results in significantly improved EDS and increased EL .