Right Ventricle Dysfunction in Patients Undergoing Lung Transplant

Overview

Patients enlisted for bilateral lung transplantation (LUTX) have subclinical right ventricle (RV) dysfunction1, which is usually clinically silent until LUTX. During LUTX, several reasons (i.e., sequential pulmonary arteries cross-clamp, hypoxia, hypercapnia) lead to de-compensation of RV function, cardiac failure and shock2. In this clinical scenario, extracorporeal life support (ECLS) with cardiopulmonary bypass (CBP) or extracorporeal membrane oxygenation (ECMO) is emergently implemented. ECLS is associated with prolonged mechanical ventilation, primary graft dysfunction (PGD), bleeding, and graft rejection3. This may be due to: 1) the activation of pro-inflammatory cascade due to blood-circuit contact; 2) the increased need for allogenic blood components, which per se has been associated to an increased risk of PGD4. Avoiding intraoperative ECLS may thus have significant positive clinical outcomes. In the general cohort of patients undergoing LUTX, pulmonary hypertension, and right ventricular dysfunction have been identified as risk factors for intraoperative ECLS5. At enlistment for LUTX, patients undergo a comprehensive evaluation of right cardiac function comprising: transthoracic echocardiography, pulmonary artery catheterization, and calculation of RV ejection fraction (RVEF) by multiple gated radionuclide ventriculography. Echocardiography is non-invasive, can be performed repeatedly and at the bedside. The free-wall RV longitudinal strain (RVLS) is a novel echocardiographic method for quantification of myocardial deformation6 with high diagnostic accuracy to predict depressed RV ejection fraction. RVLS may be used for non-invasive, repeated and bedside assessment of RV function before LUTX. We envision the employment of RVLS to document subclinical RV dysfunction before LUTX.

This study is a prospective observational cohort analysis of echocardiographic studies and medical records of consecutive patients who underwent LUTX at our Institution from January 2021 to March 2023. All patients enlisted for LUTX during the study period were considered for inclusion. Exclusion criteria were: 1) single LUTX; 2) re-transplantation; 3) patients bridged to LUTX with veno-venous ECLS; 4) missing medical records. At our Institution, at enlistment for LUTX, patients undergo a comprehensive cardiac evaluation, comprising: 1) invasive right heart catheterization; 2) multi-gated radionuclide ventriculography; 3) trans-thoracic echocardiography performed by a specialized cardiologist. For further details on the management of LUTX at our Institution, see Online Supplement, Additional Methods. To conduct this study, following enlistment, the research team contacted the patients, and a specialized sonographer (SS) and a specialized cardiologist (PM) blinded to the results of the enlistment echocardiography, carried out a further echocardiographic examination for the measurement of RV strain. Specifically, a GE Vivid IQ machine (GE Healthcare, Milwaukee, WI) was used. Images were acquired during breath holds with stable electrocardiographic recordings and digitally stored for subsequent offline analysis using EchoPAC Clinical Workstation Software (GE Healthcare, Milwaukee, WI). RV global longitudinal strain (RVGLS) and RV free wall strain (RVFWS) were calculated ex-post using conventional the two-dimensional echocardiographic apical 4-chamber (17,18) or - when inaccessible - subcostal view. Thus, according to the most recent data available, patients were classified as having and abnormal (\>-16.9%), borderline (between -16.9% and -19.2%), and normal (\< -19.2%) RVFWLS. We obtained the following measurement following international guidelines: right atrium (RA) area, RV end-diastolic area (RV EDA), RV free wall thickness, fractional area change (FAC), M-mode measured tricuspid annular plane excursion (TAPSE), pulsed-wave tissue Doppler imaging (TDI) tricuspid peak annulus systolic velocity (S'), and pulmonary artery systolic pressure (PAPs). The following data at the time of enlisting for LUTX were prospectively collected: demographics, weight, height, the indication to LUTX (further aggregated in pulmonary fibrosis vs. not pulmonary fibrosis), comorbidities, lung allocation score (LAS), oxygen requirement at rest, spirometry, arterial blood gas analyses, diffusing capacity of carbon monoxide (DLCO), six-minute walking test (6MWT), pulmonary arterial pressures and cardiac output (by invasive cardiac catheterization); pulmonary scintigraphy; right ventricle ejection fraction (RVEF) measured by multi-gated radionuclide ventriculography. Statistical analysis Data were reported as the median \[first-third quartile\] and number of events (percentage of the subgroup) for continuous and categorical variables, respectively. Patients without an available echocardiographic window for RV evaluation were not considered for the echocardiographic analysis. The Z-test was utilized to compare the patients' population with standard normality values(12,21,22). The correlation between continuous variables was tested with the R2 linear regression. Sensitivity, specificity, positive predictive value (PPV) and negative predictive values (NPV) and associated confidence intervals (CI) of RVFWS (vs. TAPSE, FAC, S', multi-gated radionuclide ventriculography) were computed. Comparison between patients' cohorts (i.e., normal RVLS vs. compromised RVLS) was performed with Chi2 or Fisher Exact Test, and logistic regressions, as appropriate. The odds ratios (OR) and associated 95% likelihood ratio-based confidence intervals were calculated. Statistical significance was accepted at P \< 0.05. The JMP® pro 16.0 (SAS, Cary, NC) was utilized.