Critical Care Medicine
Management of ECMO
- 1. Description of the problem
2. Emergency Management
Special considerations for nursing and allied health professionals.
What's the evidence?
Management of extracorporeal membrane oxygenation (ECMO)
Acute life threatening pulmonary failure.
1. Description of the problem
What is ECMO?
ECMO is a temporary mechanical support system for the management of reversible or potentially reversible life-threatening cardiac and/or respiratory failure unresponsive to conventional supportive measures.
ECMO principle: blood is removed from the venous system through an inflow cannula, oxygen is added and carbon dioxide is removed through a membrane oxygenator, and blood is returned by pump through an outflow cannula to an artery, V-A ECMO, or to a vein, V-V ECMO.
ECMO provides time for heart and lung to recover from initial insult, acting as a bridge to recovery. Although ECMO is considered a short term mechanical temporizing solution, the potential duration of ECMO support with newer technology has dramatically increased from days to weeks.
Venovenous ECMO (V-V ECMO) supports oxygenation and CO2 removal and requires an adequate cardiac function.
Veno-arterial ECMO (V-A ECMO) provides oxygenation, CO2 removal and also adequate perfusion.
E-CPR is an emerging concept of low flow V-A ECMO for emergent resuscitations, using small, easy to insert percutaneous cannulas.
A-V ECMO (AVCO2R) is an arterio-venous pumpless circuit for CO2 removal.
V-V ECMO should be considered in patients with acute lung injury and preserved cardiac function:
Major airway trauma.
Acute lung transplant failure, primary graft failure.
V-A ECMO should be considered in patients cardiogenic shock with or without acute lung injury:
Acute heart transplant failure.
Post cardiotomy cardiogenic shock.
Post acute myocardial infarction shock.
Severe cardiac depression due to sepsis.
Severe cardiac depression due to drug overdose.
Cardiac or major vessel trauma.
High risk PCI.
Acute on chronic cardiomyopathy: as a bridge to bridge.
Irreversible neurological impairment.
Active malignancy with limited survival.
Advance liver cirrhosis, MELD score 20-30.
Contraindication to systemic anticoagulation.
Age over 65.
Multiple trauma with severe bleeding.
Difficult vascular access (IVC filter, severe PVD).
High pressure, high FIO2 ventilation for more than 7 days.
Contraindications specific to V-V ECMO
Severe pulmonary hypertension (mPAP >45-50mmHg, or >75% of SBP).
Severe RV or LV failure.
Contraindications specific to V-A ECMO
Severe aortic regurgitation.
Indications for ECMO referral
Acute, potentially reversible, respiratory failure.
Bridge to transplant.
No absolute contraindications.
Advanced modes of ventilation, high PEEP.
SPO2 less than 85% for more than 1 hour.
Oxygenation index (OI) greater than 25 for more than 6 hours. OI = (FIO2 x mean Paw x 100)/PaO2.
PaO2/FiO2 less than 100mmHg with PEEP greater than 10 cm H2O for more than 24-48 hours.
PaO2/FiO2 less than 60 mmHg for more than 1 hour.
Respiratory acidosis pH less than 7.20.
PaO2/FiO2 less than 100mmHg and PCO2 greater than 100mmHg for more than 1 hour.
Potentially reversible cardiogenic shock.
Bridge to destination support/transplant.
No absolute contraindication.
Maximized medical therapy with or without IABP.
SVO2 less than 55-60% with HCT over 30.
CI less than 1.6-1.8 L/min/m2.
CVP more than 15-20 mmHg.
The ultimate desicion is made by the intensivist/surgeon experienced in ECMO management.
Basic circuit: blood pump, oxygenator (membrane lung), blood tubing and venous and arterial cannulas. Auxiliary components include heat exchanger, monitors and alarms.
Centrifugal or roller pump are the most common. Axial or peristaltic pump can also be considered.
Centrifugal pump: Disadvantages: hemolysis (much improved with new pumps), resistance dependent flow; Advantages: reduction of platelet deposition, increased run-time.
Disadvantages: rupture and embolism of tube particle, air embolism; Advantages: constant flow.
Silicone membrane or hollow fiber.
Silicone membrane oxygenator: Highly efficient.
Hollow fiber oxygenator: Fast circuit priming, reduced risk of thrombosis, less platelet and inflammation activation, less hemolysis secondary to lower pressure gradient across the membrane. In the past, plasma leakage through the microporous hollow fiber membrane caused reductions in gas transfer. The membrane currently used, solid hollow fiber, is a thin plasma resistant layer of polymethylpentene, which prevents plasma leak.
Membrane surface area and blood flow determines maximal oxygen delivery. The capacity of oxygen exchange for specific membranes is described as a rated flow. Rated flow is the volume of desaturated blood (SO2 75%) that the membrane can return to an SO2 of 95% per minute (L/min).
CO2 removal is more efficient than oxygenation because of solubility and diffusion properties. Because of this, CO2 removal can be done with low flows. Hence membranes with a low rated flow are sufficient for this purpose.
The new generation of solid hollow fiber membrane oxygenators and centrifugal pumps are capable of several weeks of continuous function.
Sweep gas is either 100% oxygen or carbogen (95% O2 and 5% CO2). Sweep gas flow is a gas flow in liters per minute through the membrane oxygenator. Sweep gas flow rates are equal to blood flow. An increase of sweep gas flow increases CO2 elimination. When interested only in CO2 removal with low rated flow membranes, the sweep gas (100% O2) flow rate is usually ten times the blood flow.
If the sweep gas pressure is higher than blood pressure, oxygen emboli can occur. Specifically, hollow fiber membranes are at risk. Keeping the membrane below patient level, increasing blood pressure, and controlling sweep gas pressure, servo regulation control or pop off valves all minimize the risk of this phenomenon.
Tubing should have minimal resistance to venous drainage and should avoid high resistance on outflow side. Length and internal diameter of the tubing determines the resistance. The blood flow resistance is directly proportional to length and inversely proportional to radius to the fourth power. Blood flow will depend on the pressure gradient and resistance of the tubing/cannula ECMO system. 100 mmHg pressure gradient through one meter of half-inch blood tubing delivers 10 L/min flow.
Used to compensate for heat loss secondary to extracorporeal flow. Water temperature is kept between 37-40°C. Higher temperatures can lead to hemolysis and bubble formation.
In line monitoring components of ECMO
Blood flow, gas flow, oxygen fraction, Hb, Sv/aO2, pH, pCO2, blood pressure before and after the blood pump and oxygenator, bubble detector. Pressure before and after oxygenator is used to monitor a clot formation in the oxygenator. An increase in pressure differential without a concomitant increase in ECMO flow is a sign of clot formation.
Line after pump pressure should not exceed 350-500 mmHg. If it does, a high SVR, high resistance in outflow cannula, tubing or membrane oxygenator could be the cause.
Before pump, inflow line pressure can be used to monitor negative suction pressure by the pump.
Oxygenation is measured both before and after the oxygenator. PO2 levels before the oxygenator indicate the degree of recirculation between inflow and outflow cannula in V-V ECMO. The PO2 levels levels measured after the oxygenator reflects the function of the oxygenator and should be greater than 150mmHg.
Veno-venous cannulation consists of one or two large central venous "inflow"/"inlet" cannulas to pump/oxygenator and one "outflow"/"outlet" cannula delivering oxygenated blood to the patient, positioned in or close to right atrium. Veno-arterial cannulation consists of a large central venous "inflow" cannula and a major arterial "outflow" cannula.
Types of cannulation
Central surgical cannulation: usually refers to arterial cannula or graft connected directly the aorta, using either an open or closed chest approach. This is common after cardiac surgery (i.e. following failure to wean from CPB).
Peripheral surgical cannulation: cut down and cannulation of major vessel. Femoral vein is preferable to jugular vein. Right femoral vein is preferable to left. Femoral artery cannulation may require distally placed "backflow" cannula to the superficial femoral artery, especially in patients with limited collateral flow (i.e. young patients).
Peripheral percutaneous cannulation: As above, using Seldinger technique. Percutaneous subclavian cannulation should be avoided secondary to bleeding complications.
After guide wire insertion, just before cannula placement, a bolus of heparin (5000-10,000 units or 50-100 units/kg) is given.
Transesophageal echocardiography should be considered to evaluate cannula positioning.
V-V ECMO inflow venous cannula tip should be in the IVC at the level of diaphragm when introduced through the femoral vein.
When introduced through the jugular vein, the cannula tip should be in the subclavian vein. Outflow venous cannula tip should be in right atrium.
To achieve the required flow rates, the cannulas should be at least 20Fr, and ideally 23-26Fr.
Large double lumen cannula are now available that allow for single vessel cannulation.
ECMO flow targets
V-A ECMO (for cardiac failure): 50-60cc/kg/min. The flow is limited by vascular access, length and size of cannula, and pump properties.
V-V ECMO (for respiratory failure): 60-80cc/kg/min. CO2 removal always exceeds O2 delivery. Besides blood flow, oxygenator/membrane properties and gas sweep determine O2 and CO2 levels.
A-V ECMO (for hypercapnic respiratory failure): 25% of cardiac output.
After heparinization and cannulation, blood flow is initially increased to determine the maximum flow capacity of the circuit.
Flow is limited by venous pressure, resistance of the tube/cannula and patient right or left circulation resistance. Thus hypovolemia, kinking of tubing, clot in oxygenator or partial cannula obstruction by clot or vessel has to be considered.
The pump flow is then decreased to the lowest level which provides adequate support for specific type of ECMO.
V-V ECMO goal
Arterial saturation 85-90% (PaO2 55-60mmHg) with the lowest possible CVP. With current cannulation strategies, it is neither possible nor necessary to obtain complete venous drainage. Hence, efficiency of V-V ECMO depends on the ECMO flow relative to patient's total venous return and cardiac output. As long as cardiac output and hemoglobin concentration are normal in a sedated normothermic patient, an arterial saturation of 85-90% is adequate for systemic oxygen delivery. Higher SaO2 targets (>90%) would require higher ECMO flows, which can predispose to volume overload, venous congestion, pulmonary edema and hemolysis.
Arterial saturation below 80% (PaO2 <50 mmHg) in a setting of reasonable ECMO flows and postoxygenator PO2 above 150mmHg or minimal improvement in saturation with higher flows suggests a high degree of recirculation through the ECMO circuit (i.e. a decrease in ECMO/patient's total venous return ratio). This would be confirmed by a decreased difference between pre- and post-oxygenator blood saturations. Recirculation can be decreased by adjusting or distancing the inflow and outflow cannula. This can be guided by monitoring SaO2, pre- and post-oxygenator saturation, echocardiography or fluoroscopy.
Even with optimal cannula positioning, recirculation can still occur secondary to increased pulmonary vascular resistance or right heart failure, leading to preferential flow through ECMO circuit. Other causes of decreased forward flow, including pneumothorax, pleural effusions, hemothorax, pulmonary embolism, abdominal compartment syndrome and left heart failure, should be ruled out. Only large pneumothoraces and pleural effusions causing hemodynamic compromise should be treated because of high risk for bleeding.
Pulmonary vasodilator and right heart inotropic support may be needed.
Low arterial saturation can also be caused by high patient venous return/flow relative to ECMO flow. Diuresis, if possible, or additional inflow cannula may be required.
V-A ECMO goal
Adequate oxygen delivery is the goal. Echocardiographic evaluation of the left ventricle (LV) should follow V-A ECMO initiation. In non-ejecting or minimally ejecting left ventricles blood from bronchial, thebesian veins and right sided circulation will gradually over several hours increase left atrial pressure leading to pulmonary edema. Inotropic support and vasodilator therapy should be the first intervention.
Systemic flow pulsatility, around 5-10 mmHg, generated by the heart and pulmonary artery pressure trends, can help guide the need for further evaluation. Loss of pulsatility and elevated PA pressures require echocardiography and CXR to rule out intra-thoracic processes leading to empty LV, such as pleural effusion, hemothorax, pneumothorax, versus worsening LV function.
Refractory LV distention in the setting of a closed aortic valve, moderate to severe mitral regurgitation and/or estimated high left atrial pressure, requires LV decompression to avoid progressive pulmonary edema. Additional inflow cannula insertion, LV vent insertion or transeptal left atrial decompression might be required. In patients with a reasonable chance of LV recovery (e.g. myocarditis), LV should be kept as decompressed as possible and sinus rhythm, if tachyarrhythmia is present, should be recovered as soon as possible.
Cardiac function recovery in a patient with severely compromised lung function and with peripheral V-A ECMO has to be closely monitored as the possibility exists of delivering critically low partial pressure O2 blood from native heart to coronaries, brain and the right upper extremity.
An arterial line or pulse oximetry probe in right upper extremity should be monitored, although pulse oximetry may be unreliable because of diminished pulsatility. Brain oximetry may provide an informative trend to guide the frequency of right radial artery sampling. In patients with primary lung pathophysiology leading to acute cardiac hypoxic or metabolic dysfunction, early switch to V-V ECMO should be considered in this case.
After a bolus (50-100 units/kg) of heparin for cannulation an infusion is started at rate of 10units/kg/h. The heparin is monitored either by activated clotting time (ACT) or activated partial thromboplastin time (aPTT). A usual iniital target for the non-bleeding patient with a platelet count over 70,000 is ACT 180-220 or aPTT 60-90.
The target for anticoagulation has to be individualized based on signs of hypo- or hypercoagulability, ECMO flow rates (flows <2L/min require greater anticoagulation) and end organ injury (liver and kidney dysfunction).
Centrifugal pump head, the inflow side of the oxygenator and connectors are the most likely sites of clot formation. A flashlight can be used to examine the ECMO circuit for evidence of clots. Small clots (<5mm) do not necessarily effect the oxygenator function and do not require circuit/oxygenator change.
Hypo-coagulopathic patients may require heparizination targets as low as an ACT 150 or aPTT 45.
Severe bleeding requires temporarily stopping heparinization and reversal of coagulopathy.
In patients who require V-A ECMO after cardiopulmonary bypass, heparin infusion is started once they are normothermic, chest tube output is low (less than 100cc/hr), and coagulation parameters are close to normal (INR <1.6, Plt>70,000). Heparin is usually held until 12 to 24 hours post-operatively.
Ventilator settings should not be changed until adequate ECMO flows and gas exchange are established. Deep sedation and even paralysis to avoid high intrathoracic pressure may be required in the first 24-48 hours.
Typical "resting" ventilatory settings include an FiO2 below 60%, respiratory rate below 10, peak airway pressure below 35cmH2O, mean airway pressure below 25 cmH2O and positive end expiratory pressure below < 15 cmH2O.
High airway, high intrathoracic pressures, very low airway pressures and lung derecruitment can all increase PVR and may negatively impact the hemodynamics and/or gas exchange in patients on V-V ECMO.
Large air leaks (i.e. broncho-pleural fistulas) may require very low (< 10cm H20) continuous positive airway pressures until the leak closes. After the leak is sealed airway pressure is gradually increased to normal ventilatory settings. The recruitment of the completely atelectatic lungs may take days and may not occur until spontaneous ventilation is resumed.
Free plasma hemoglobin (Hb) is usually less than 10mg/dl. A plasma free Hb above 50mg/dl is significant.
Clinical signs of hemolysis include red or dark brown urine, elevated potassium, unxplained new onset of renal failure, icterus.
Causes of hemolysis secondary to ECMO include clot within a circuit or around cannulas, inadequate preload causing very low suction pressure and intermittent inflow cannula occlusion (chattering), or inappropriately high pump speed.
Management: Increase preload, adjustment of pump speed, echocardiography to rule out cannula obstruction, circuit change if necessary.
Most often secondary to anticoagulation. Bleeding sites include those for vascular access, surgical sites or remote (e.g gastrointestinal or intracranial).
Surgical bleeding shouldbe investigated. High blood pressure should be controlled. Further managment depends on severity of bleeding, ranging from maintaining low ACT around 150-180 to reversing anticoagulation completely. Fresh frozen plasma transfusion if INR is elevated and platelets transfusion if less than 70,000U may be required.
White (platelet) or dark clots can be seen in low flow parts of the ECMO circuit, e.g. connectors and oxygenator. A flashlight is used for clot detection. Large clots (> 0.5 cm) or rapidly growing clots should be removed by changing the whole or the part of the circuit.
Circuit air bubbles
Observed or detected by a bubble detector. Most commonly found in inflow, venous, part of the circuit. The pump has to be stopped, lines have to be clamped near the patient and bubbles must be evacuated or the whole circuit has to be changed.
V-A ECMO: Signs of cardiac recovery should occur early (days) depending on pathophysiology, and include rhythm recovery, less inotropic support required to maintain increased pulsatility and improved contractility on echocardiography.
Weaning: After optimization of inotropic and blood pressure support, a bolus of heparin (2000-5000 units) is administered to prevent ECMO circuit clot formation from low flow and the ECMO flow is gradually decreased to around 1 L/min. Hemodynamic stability can be followed by SVO2, CVP, PAP and CO, and cardiac contractility recovery can be assessed by echocardiogram. Adequate hemodynamics, an improved echocardiogram and low to moderate inotropic support suggests readiness for decannulation.
V-V ECMO: Recovery of pulmonary function can take weeks. Signs of lung recovery include improved CXR and improved static compliance with or without recruitment.
Weaning: Maintaining circuit flow, normal (i.e. non-resting) ventilatory settings are established. ECMO FIO2 and flow is turned gradually down and if tolerated then off. In borderline patients a trial of weaning - cannulas clamped but not removed up to several hours - can be performed. The trial requires frequent or continuous cannulas flushed with heparinized saline to avoid thrombosis.
Arterial cannulation sites need to be surgically repaired. Venous cannulation sites require manual compression for at least 20 minutes following cannula removal.
2. Emergency Management
Special considerations for nursing and allied health professionals.
What's the evidence?
Kolobow, T, Zapol, W, Pierce, JE. "Partial extracorporeal gas exchange in alert newborn lambs with a membrane artificial lung perfused via an A-V shunt for periods up to 96 h". Trans Am Soc Artif Intern Organs. vol. 14. 1968. pp. 328-34.(The first design of membrane oxygenator for prolonged use.)
Hill, JD, De Leval, MR, Fallat, RJ. "Acute respiratory insufficiency: treatment with prolonged extracorporeal oxygenation". J Thorac Cardiovasc Surg. vol. 64. 1972. pp. 551-62.(The first favorable patient outcome from ECMO application for respiratory distress.)
Peek, M, Mugford, M, Tiruvoipati, R. "Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomized controlled trial". The Lancet. vol. 374. 2009. pp. 1351-63.(The most recent randomized trial of ECMO use for ARDS.)
Combes, A, Pellegrino, V. "Extracorporeal membrane oxygenation for 2009 influenza A (H1N1)-associated acute respiratory distress syndrome". Semin Respir Crit Care Med. vol. 32. 2011. pp. 188-94.(A report on the most successful use of ECMO for ARDS so far.)
Brodie, D, Bacchetta, M. "Extracorporeal Membrane Oxygenation for ARDS in Adults". N Engl J Med. vol. 365. 2011. pp. 1905-14.(A recent review article on ECMO for ARDS.)
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
Endocrinology Advisor Articles
- Two Phases of C-Peptide Decline Identified in Type I Diabetes
- Dulaglutide Effective for Patients With T2D, Moderate to Severe CKD
- Incidence of Diabetes Influenced by Endocrine-Disrupting Chemicals in the Environment
- Guidelines for Management of Hypothalamic-Pituitary, Growth Disorders in Childhood Cancer Survivors
- Assessment of Low-Dose Aspirin As Chemoprevention in Type 2 Diabetes
- Using Latent Class Trajectory Analysis to Determine Glucose Response Curve Patterns
- First CGM System With Implantable Glucose Sensor Approved
- Adjunctive Metformin for Insulin Resistance in T1D: A Clinical Perspective
- Risk for Below Knee Amputations With Canagliflozin vs Other Antihyperglycemic Agents
- Empagliflozin, Linagliptin Combination Therapy vs Linagliptin Monotherapy for Type 2 Diabetes
- Type 2 Diabetes and Alzheimer Disease: What's the Connection?
- Government and Industry Lead the Way in Funding USPSTF Systematic Reviews
- New Romosozumab BLA Includes Data from Additional Osteoporosis Trials
- Impact of Maternal Diabetes on Outcomes in Very Preterm, Very Low-Birth-Weight Infants
- Age-Related Differences in Manifestation of Cushing Disease Identified