Passive Leg Raising for Predicting Fluid Responsiveness in VA-ECMO Assisted Patients

Overview

Peripheral Veno-Arterial Extra Corporeal Membrane Oxygenation (VA ECMO) is a temporary assistance that provides a mechanical circulatory support in patients victim of cardiogenic shock (CS) or refractory cardiac arrest. During VA-ECMO support, hypotension may frequently occur due to deteriorated cardiac function, vasoplegia, or hypovolemia. Volume expansion is a common means to correct hypotension and improve systemic perfusion, but inappropriate fluid therapy is associated with adverse outcomes. As other intensive care unit (ICU) patients, VA-ECMO assisted patients have been shown to have higher mortality in case of large early fluid administration. Prediction of fluid responsiveness could achieve a lower fluid balance and improve outcomes of patients treated with VA-ECMO. Several dynamic hemodynamic parameters based on cardio-pulmonary interactions (stroke volume, pulse pressure or inferior vena cava variations induced by invasive ventilation cycles) have been described and validated for predicting fluid responsiveness in critically ill patients. Unfortunately, the VA-ECMO conditions (native cardiac circulation by-pass, low pulsatility, presence of drainage canulation in the inferior vena cava, the use of low tidal volume) make this parameters less reliable. Simulation of a fluid loading by shifting blood from the lower limbs and splanchnic compartment thanks to a revisable maneuver is another feasible approach to assess fluid responsiveness. Whereas the use of different maneuvers have been validated in the classical ICU population, very few data exist in the ECMO population and their application is questioning because blood transfer may be modified by the preload dependence of the ECMO. Recently, Luo et al showed that the variation of aortic Velocity Time Integral (VTI) measured using echocardiography induced by a Trendelenburg maneuver was predictive of fluid responsiveness during VA-ECMO support. However, their study excluded patients with low cardiac ejection (pulse pressure \< 15 mmHg) so that their data may not be extrapolated to the acute phase of heart failure requiring full mechanical support. Moreover, aortic VTI measurement suffers from low reproducibility in case of low native cardiac output (NCO) and arrythmia; and can be time-consuming. The investigators previously demonstrated in an observational prospective study that End-tidal CO2 (EtCO2) and Pulse Pressure (PP) were strongly correlated to NCO during VA-ECMO when NCO \< 2l/min. The investigators aim to study the variations of aortic VTI, EtCO2 and PP induced by Passive Leg Rising (PLR) and their ability to predict fluid responsiveness in patients under VA-ECMO.

Settings : This prospective non interventional study is conducted since february 2020 in the investigators' ICU and has been approved by their hospital's institutional review board. Informed consent is obtained from all patients or their surrogates. Patients : All patients who receive VA-ECMO support and mechanical ventilation for refractory cardiogenic shock or cardiac arrest of any etiology (acute myocardial infarction, end-stage dilated cardiomyopathy, heart surgery, fulminant myocarditis…) are screened. Patients conditioning : Patients are sedated by propofol/dexmedetomidine and sufentanyl and under invasive mechanical ventilation. Continuous blood pressure is monitored via a radial arterial catheter. PP is defined as systolic arterial pressure-diastolic arterial pressure. Central venous catheter is placed at the upper body for central venous oxygen saturation (ScVO2) monitoring. Lung ventilation is managed with low levels of respiratory rate (10-14 breaths/min) and tidal volume (4-6mL/ kg), and with a modest level of positive end- expiratory pressure (8-10 cmH2O) to ensure protective ventilation. EtCO2 is measured noninvasively from exhaled breath on a ventilator circuit and monitored using a ventilator CO2 analyzer (Maquet servo U, Drager Evita Infinity V500). ECMO circuit settings and patients management under ECMO : VA-ECMO consist of polyvinyl chloride tubing with a membrane oxygenator (PH.I.S.I.O and EOS; Sorin Group, Clamart, France), a centrifugal pump (Stockert; Sorin Group), and percutaneous or surgically inserted arterial and venous femoral cannulae (Fem-Flex and Fem-Track, Edwards Life- sciences, Guyancourt, France) with or without an additional 7 F cannula inserted distally into the femoral artery to prevent lower limb ischemia. An oxygen-air blender (Sechrist Industries, Anaheim, CA) ventilate the membrane oxygenator. Unfractionated heparin is administrated to maintain an anti-factor-Xa activity of between 0.2 and 0.3 IU/mL. In the initial phase of the circulatory assistance, VA-ECMO flow is set to provide adequate tissue perfusion (ScVO2 ≥ 65%) and to obtain correction of metabolic acidosis (serum lactate clearance). Thereafter, the VA-ECMO flow is set at the lowest rate necessary to ensure adequate tissue perfusion, while the highest NCO is wanted. Respiratory minute ventilation and ECMO sweep gas flow are adjusted to maintain baseline PaCO2 in a normal range, of around 40 mmHg. Protocol : Following data are recorded before inclusion : * Age, gender, weight. * VA-ECMO indication and reason to perform vascular expansion * Vasopressive support : noradrenaline mg/h, dobutamine mg/h; inhaled nitric oxyde. * Richmond agitation-Sedation scale (RASS) and Behavioral Pain Scale (BPS) levels. * Ventilatory settings : tidal volume, positive end-expiratory pressure. Protocol steps : The protocol include four sequential steps : 1. Baseline position: semirecumbent position (45°) 2. Supine position with a 0° angulation 3. Passive leg raising (after securing of VA ECMO tubing) : using an automatic bed elevation technique, the lower limbs are raised to a 45° angle while the patient's trunk is lowered in supine position 4. Return to baseline and realization of a fluid challenge (administration of 500 mL of isotonic saline serum over 15 min). Data collection : All following data are recorded after 1 minute of stabilization at step 1, 3 and at the end of step 4 : * Clinical data : heart rate, systolic, diastolic, mean and pulse arterial pressures, EtCO2. * Doppler echocardiography (transthoracic or trans-esophageal) data : E and A waves measured with mitral inflow Doppler (ratio E/A), Velocity Time Integral (VTI) at the level of the left ventricular outflow, using the 5-chamber apical view. Three consecutive measurements are recorded to calculate a mean VTI value. * ECMO data : pump outflow (PO, in mL/min), pump rotation speed (round/min). Left ventricular outflow tract (LVOT) diameter is measured by echocardiography at step 1 Central venous pressure is measured at step 2 and 3 Tele-expiratory inferior vena cava diameter is measured at step 1 and 3 Statistical analysis : Categorical variables (expressed as absolute value and percentage) will be compared using the chi-squared test. Continuous variables (expressed as median \[25th-75th percentile\]) will be compared with Student's t test or the Mann-Whitney U test, as appropriate according to the normality distribution assessed graphically. Fluid responsiveness will be defined by a 15% increase of aortic VTI (ΔVTI\>15%) between step 1 (half sitting position) and the end of step 4 (after fluid expansion). Linear regression analysis will be used to demonstrate relationships between percent change of VTI (ΔVTI), PP (ΔPP), EtCO2 (ΔEtCO2) induced by PLR maneuver and fluid challenge. Receiver Operating Characteristic (ROC) curves will be generated to evaluate percent changes in VTI, PP and EtCO2 induced by the PLR maneuver to predicts fluid responsiveness. The area under ROC curve will be compared using the DeLong test. Sensitivity, specificity, positive and negative predictive values (PPV and NPV), and associated 95% confidence intervals (CI) will be calculated based on the cutoff value as determined by the Youden Index (specificity+sensitivity - 1). Statistical significance will be defined as p \<0.05.