Novel mesh-derived right ventricular global longitudinal strain predicts ICU outcomes in patients undergoing coronary artery bypass grafting

Introduction Right ventricular (RV) strain analysis proved to be prognostically relevant in various cardiac diseases. However, data regarding outcome prediction by RV strain analysis during surgical coronary revascularization are limited. Three-dimensional (3D) transesophageal echocardiography-based (TEE) methods dedicated to the RV might be advantageous over established measures of cardiac function. Methods 87 patients undergoing coronary artery bypass grafting with (CABG, n=49) or without cardiopulmonary bypass (off-pump coronary artery bypass grafting, OPCAB, n=38) were enrolled retrospectively if suitable 3D-TEE was performed after anesthesia induction. Post-hoc offline analysis using 3D-STE software (4D RV-Function 2.0, Tomtec) was used to generate RV meshes. Regional and RV global longitudinal and circumferential strain analysis on the surface of the meshes was performed using a custom-made software algorithm based on the Visualization Toolkit (www.vtk.org). Global RV and left ventricular (LV) ejection fraction (EF) as well as LV global longitudinal strain (LV-GLS) were assessed using commercially available 3D-STE software packages (4D RV-Function 2.0 and 4D LV-Analysis, Tomtec). Prolonged (>48 hours) post-operative mechanical ventilation was used as the primary end-point. Results RV strain analysis was feasible in all patients (100%). Mean RV global longitudinal strain (RV-GLS) and mean RV global circumferential strain (RV-GCS) was -14.1±4.6% and -15.7±4.7% respectively. Between CABG and OPCAB patients, neither RV-GLS (-14.3±4.6% vs. 13.8±4.7%, p=0.62) nor RV-GCS (-15.9±4.4% vs. -15.4±5.1%, p=0.60) was significantly different. In univariate regression models, septal longitudinal RV strains (p Discussion In patients undergoing coronary artery bypass surgery, right ventricular global longitudinal strain analysis on meshes derived from transesophageal 3D-STE is superior in predicting short-term ICU outcomes compared to conventional parameters of right or left ventricular function and might be helpful to identify patients at higher risks for post-operative complications.