ECMO for Adult Respiratory Failure

Current Use and Evolving Applications

Abstract

Extracorporeal membrane oxygenation (ECMO) is increasingly being used to support adults with severe forms of respiratory failure. Fueling the explosive growth is a combination of technological improvements and accumulating, although controversial, evidence. Current use of ECMO extends beyond its most familiar role in the support of patients with severe acute respiratory distress syndrome (ARDS) to treat patients with various forms of severe hypoxemic or hypercapnic respiratory failure, ranging from bridging patients to lung transplantation to managing pulmonary hypertensive crises.

The role of ECMO used primarily for extracorporeal carbon dioxide removal (ECCO₂R) in the support of patients with hypercapnic respiratory failure and less severe forms of ARDS is also evolving. Select patients with respiratory failure may be liberated from invasive mechanical ventilation altogether and some may undergo extensive physical therapy while receiving extracorporeal support. Current research may yield a true artificial lung with the potential to change the paradigm of treatment for adults with chronic respiratory failure.

The use of extracorporeal membrane oxygenation (ECMO) for adult respiratory failure has markedly increased in recent years.¹ Advances in device technology, increased use during the 2009 influenza A (H1N1) pandemic, and the results of a randomized controlled trial that suggested improved outcomes in the group considered for ECMO propelled the world into the modern era of extracorporeal support for adult respiratory failure.^1–3 Despite limited high-quality evidence supporting its use, ECMO has expanded beyond its original application in the acute respiratory distress syndrome (ARDS) to include bridging patients to lung transplantation, managing pulmonary hypertensive crises, and treating patients with other forms of refractory hypoxemic or hypercarbic respiratory failure.^4–6

Principles and Circuitry

Extracorporeal membrane oxygenation works by removing blood from a large central vein, typically from the femoral or internal jugular vein, and pumping it across a gas-exchange device known as an oxygenator (Figure 1). In venovenous ECMO, blood is returned to a central vein, whereas in venoarterial ECMO, blood is returned to an artery, most commonly the femoral artery. Therefore, venovenous ECMO provides respiratory support, whereas venoarterial ECMO provides both respiratory and hemodynamic support. Arteriovenous ECMO, comprised of a pumpless circuit driven by the patient’s femoral arterial pressure, generates low blood flow rates and is primarily used for carbon dioxide (CO₂) removal.⁷

Figure 1: Two-site approach to venovenous extracorporeal membrane oxygenation cannulation. The venous drainage cannula is typically placed in the femoral vein and extends to in the inferior vena cava; the venous return cannula is typically placed in the internal jugular vein and extends to the right atrium. Venous blood is drawn from the femoral vein into the pump and propelled into the oxygenator before being returned into the internal jugular vein. Reprinted with permission of collectedmed.com.

The oxygenator is divided into two chambers, separated by a semipermeable membrane, across which oxygen and CO₂ are exchanged by diffusion. Modern oxygenators are comprised hollow fibers made of polymethylpentene which allow diffusion of gas, but not liquid. Driven by a centrifugal pump, blood flows along one side of the membrane while sweep gas flows along the other side. The composition of the sweep gas is oxygen and ambient air, in proportions controlled by a blender. Carbon dioxide is more soluble than oxygen and diffuses readily across the membrane, so CO₂ clearance is primarily determined by the gradient across the oxygenator membrane. Sweep gas flow removes CO₂ from the gas chamber, thereby maintaining the CO₂ gradient and allowing for continued diffusion, so that higher sweep gas flow rates effectively provide more ventilation.⁸

Oxygenation is primarily determined by the blood flow rate across the oxygenator membrane.⁸ As the blood flow rate is largely limited by the size of the venous drainage and, to a lesser extent, the return cannulae, larger cannula sizes permit greater blood flow rates and allow for a greater percentage of a patient’s cardiac output to be oxygenated by the device. Arterial oxygen delivery varies according to changes in the cardiac output, which is a reflection of a patient’s physiologic demand. At a fixed device blood flow rate, the proportion of total oxygen delivered to the patient by ECMO will decrease as the patient’s cardiac output increases. In the normal patient with healthy lungs, this phenomenon is immaterial. However, this may be significant in a patient with pulmonary disease, as fixed oxygen delivery from the ECMO circuit with increased physiologic demand will be reflected as a decreased partial pressure of arterial oxygen (PaO₂). In addition, recirculation, which occurs when reinfused, oxygenated circuit blood is drawn directly back into the venous drainage cannula and bypasses the patient’s native circulation, does not contribute to systemic oxygenation and will negatively impact systemic oxygen saturation.

Cannulation in venovenous ECMO can be two site or single site. In two-site cannulation, blood is typically removed from a femoral vein and returned to the internal jugular vein (Figure 1). In single-site cannulation, a dual-lumen cannula is positioned in the right or left internal jugular vein (or less commonly in the subclavian vein), with its distal end extending into the inferior vena cava (Figure 2). Venous blood is removed from two drainage ports positioned in the superior and inferior vena cavae, whereas oxygenated blood is returned through the other lumen of the same cannula with its outlet port positioned within the right atrium and the reinfusion jet directed toward the tricuspid valve. The advantages of a dual-lumen cannula are single-site cannulation, minimal recirculation, and improved patient mobility.^9–12 Seldinger technique is used in both approaches; however placement of a dual-lumen cannula typically requires fluoroscopy, echocardiography, or both.¹³ Major risks of ECMO include hemorrhage, hemolysis, infection, and circuit- or cannula-related complications such as clotting, neurovascular damage, or limb ischemia.¹⁴ The rate of these and other complications is markedly lower in the modern era of ECMO.¹⁵

Figure 2: Single-site approach to venovenous extracorporeal membrane oxygenation (ECMO) cannulation. A dual-lumen cannula is typically positioned in the internal jugular vein and terminates in the inferior vena cava. Venous blood from the drainage lumen is drawn into the ECMO circuit from ports positioned in the superior and inferior vena cavae. Blood oxygenated by the ECMO circuit is returned to the second lumen of the same cannula, through a port in the right atrium, with blood flow directed across the tricuspid valve. Reprinted with permission of collectedmed.com.

ECMO for ARDS

Severe ARDS is the most commonly accepted indication for ECMO in adult respiratory failure.¹ Although ECMO was first successfully used in an adult with traumatic ARDS in 1971, its use during the ensuing decades was limited because of high mortality and complication rates.^16–18 Two randomized controlled trials comparing ECMO with mechanical ventilation to mechanical ventilation alone, performed with what is now outdated technology, showed no survival benefit in the group receiving ECMO.^17,18

However, with the advent of modern extracorporeal technology and critical care practices, a group of intensive care units in Australian and New Zealand reported a 73% survival rate in patients with severe ARDS treated with ECMO during the influenza A (H1N1) pandemic.^19,20 Several European centers also reported high survival when using ECMO for H1N1-related ARDS.^21,22 The efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR) trial published in 2009 randomized 180 adult patients with severe but potentially reversible respiratory failure to conventional management versus referral to an ECMO center for consideration of ECMO.²³ The trial showed a 16% absolute reduction in the primary end-point of death or severe disability in the ECMO-referred group. Although impressive in design and execution, the lack of a standardized ventilator management protocol in the control group, among other criticisms, limits the strength of the authors’ conclusions.⁴

Because definitive evidence for its use in ARDS is lacking, ECMO is still considered to be a rescue therapy for this indication.²⁴ An ongoing, international, randomized, controlled trial, ECMO to Rescue Lung Injury in Severe ARDS (www.clinicaltrials.gov [NCT01470703]), seeks to address this by randomizing patients to standard-of-care mechanical ventilation or ECMO, thereby addressing some limitations of the CESAR trial.⁴

There are no universally agreed-upon criteria for when to initiate ECMO for severe ARDS, but reasonable indications include severe hypoxemia with PaO₂ to a fraction of inspired oxygen ratio (FiO₂) of less than 80 mm Hg, uncompensated respiratory acidosis with a pH less than 7.15, or potentially harmful end-inspiratory plateau airway pressures greater than 35 cm H₂O despite optimized sedation and ventilator management.⁴ There are no absolute contraindications to ECMO, although consideration should be given to the patient’s likelihood of recovery with ECMO, especially in patients with advanced age, multiple organ failures, or severe comorbid disease.

Even high-volume centers may care for only a few patients annually with ARDS severe enough to be appropriate for ECMO, so many patients originate at other institutions. Ambulance and air transport of patients on ECMO have been shown to be both feasible and safe.^25,26

The role of ECMO may be most apparent when ARDS is more severe that tissue hypoxia or life-threatening hypercapnia-mediated acidosis result. However, the benefit of ECMO may also lie in limiting the ventilator-associated lung injury resulting from the high-volume, high-pressure ventilation often necessary to maintain adequate gas exchange in the absence of ECMO.

Low-volume, low-pressure mechanical ventilation (ventilation at 6 ml/kg or less of predicted body weight [PBW] with an end-inspiratory plateau pressure of ≤30 cm H₂O) reduces mortality compared with mechanical ventilation at higher pressures and volumes, even when associated with lower systemic oxygenation.^27–29 The survival benefit of this strategy may extend for at least 2 years.³⁰ Despite the benefit of lung-protective ventilation, its consistent use in ARDS may be limited by the poor lung compliance associated with severe lung injury and the unacceptable acidosis resulting from lowering tidal volumes sufficiently to achieve an acceptable end-inspiratory plateau pressure. In this circumstance, lung-protective ventilation is often forfeited to maintain pH.^31,32 By correcting the acidosis while simultaneously oxygenating the blood, ECMO permits low tidal volume ventilation while maintaining an acceptable pH.

The benefit of ECMO may extend beyond facilitating low tidal volume ventilation. Enhanced lung-protective ventilation, sometimes referred to as “lung rest,” targets tidal volumes and end-inspiratory plateau pressures lower than the currently accepted standard of care (often <20–25 cm H₂O) and may be beneficial. Similar to other applications of extracorporeal technology, this concept is not new and has undergone resurgence in the modern era of ECMO.^33–36 Achieving such low end-inspiratory plateau pressures in the setting of poor pulmonary compliance may require tidal volumes of 1–3 ml/kg of PBW, relying on extracorporeal CO₂ removal (ECCO₂R) to correct the hypercapnia-mediated respiratory acidosis seen at lower minute ventilation. As opposed to the higher blood flow needed for oxygenation, enhanced lung-protective ventilation in patients who are not profoundly hypoxemic may be achieved with lower blood flow rates and smaller cannulae that target ECCO₂R alone.^37,38

Accumulating evidence suggests that the modern application of enhanced lung-protective ventilation strategies may attenuate lung injury and result in improved clinical outcomes.^30,37–39 Lower tidal volume ventilation in ARDS has been prospectively associated with lower mortality, even at tidal volumes lower than 6 ml/kg of PBW.³⁰ Retrospective analysis of the original ARDSnet low tidal volume data shows that lower day 1 plateau pressure is associated with lower mortality, even at an end-inspiratory plateau pressure less than 30 cm H₂O.³⁹ Reduction in plateau pressures in patients with ARDS from the currently accepted levels of 28–30 to 25–28 cm H₂O with the use of ECCO₂R to manage the ensuing hypercapnia-mediated acidosis decreased inflammatory cytokines known to be associated with ARDS.³⁷ A recently published trial using ECCO₂R with very low tidal volumes (3 ml/kg PBW) compared with low tidal volume ventilation (6 ml/kg IBW) showed a trend toward the primary outcome of ventilator-free days. Post hoc analysis demonstrated a significant difference in the group with more severe hypoxemia (PaO₂/FIO₂ <150 mm Hg).³⁸ Additional trials are needed to further evaluate the benefit of very low tidal volume ventilation in patients with ARDS.

ECMO as a Bridge to Lung Transplantation

In patients with end-stage lung disease awaiting transplantation, invasive mechanical ventilation and ECMO have both been associated with high posttransplant mortality and have been considered relative contraindications to transplantation.^40–42 For the thousands of patients who may wait 12–24 months or longer for a suitable organ, decompensation requiring either often represents a tipping point that significantly reduces their likelihood of long-term survival.⁴³ Many of these patients experience complications of critical illness or mechanical ventilation that make them ineligible for transplantation.^40,44

The use of ECMO to bridge patients to lung transplantation has become increasingly accepted as recent reports of experienced centers using ECMO in this population report improved outcomes.^42,45–49 Notably, if extracorporeal support can provide adequate gas exchange, in select patients, ECMO can replace mechanical ventilation and eliminate the risk of ventilator-associated complications.^47–51 Furthermore, patients receiving aggressive physical therapy while supported with ECMO may be able to increase their pretransplant conditioning, as ECMO may compensate for the exertional hypoxemia or hypercapnia that would typically limit physical activity.^48–51 Small case series at high-volume centers suggest that this may translate into better outcomes compared with that of the general lung-transplant population.^47–49,52,53

Pulmonary Hypertension

Patients with pulmonary arterial hypertension who experience an acute crisis or progressive right ventricular failure have high morbidity and mortality.^54,55 In an acute crisis, this decompensation often occurs too rapidly for effective optimization of pulmonary vasodilator therapy and progressive cardiac dysfunction and multi-organ failure may ensue. In patients with refractory disease awaiting transplantation, mechanical offloading of the right ventricle may be the only remaining treatment modality. In these circumstances, venoarterial ECMO has been used successfully as either a bridge to lung transplantation or a bridge to recovery.^45,56–59 By bypassing the failing right heart and delivering blood directly to the arterial circulation, venoarterial ECMO allows organ function to be preserved. Patients with acute decompensation can have pulmonary vasodilator medications optimized with ECMO as a bridge to recovery, whereas patients awaiting lung transplantation may have pulmonary vasodilators weaned to permit a greater proportion of the patient’s cardiac output to flow through the low-resistance ECMO circuit.⁵⁹

Cannulation in venoarterial ECMO is traditionally performed via the femoral vein and artery, although this configuration has several notable limitations. In femoral venoarterial ECMO, oxygen delivery to the upper body is dependent on retrograde aortic blood flow and may be impeded by anterograde aortic blood flow from the native cardiac output. An alternative approach includes branching an internal jugular reinfusion cannula off the arterial cannula thereby splitting returned blood between the femoral artery and internal jugular vein and improving oxygen delivery to the upper body in a hybrid configuration known as venoarterial-venous ECMO. Because of the limited mobility and surgical and infectious risks incurred with multiple cannulation sites, this configuration is not ideal, particularly in patients awaiting lung transplantation.

Venoarterial ECMO provided through the upper body may address the limitations of femoral venoarterial cannulation. A novel configuration in which venous blood is removed from the internal jugular vein and returned to the right subclavian artery has been guarantees oxygen delivery to the upper body while optimizing mobility and minimizing infectious risk.⁵⁶ To reduce the risk of limb ischemia, the arterial cannula may be inserted into a graft sewn in an end-to-side manner positioned in the distal end of the subclavian artery.⁶⁰

Taking advantage of a patient’s anatomy, a dual-lumen cannula may be used to effectively provide venoarterial, upper-body support in patients with an atrial septal defect or patent foramen ovale.^61,62 Positioning the cannula so that the reinfused blood is directed across the defect allows an oxygenated right-to-left shunt to be created, thereby delivering oxygenated blood to the systemic circulation while offloading the right ventricle via a single cannula.^61,62 An atrial septostomy may provide the same opportunity in the absence of a preexisting intracardiac shunt.^63,64

Chronic Obstructive Pulmonary Disease

The use of ECCO₂R in acute exacerbations of chronic obstructive pulmonary disease (COPD) is an evolving and potential future application of extracorporeal technology.^65–69 Patients requiring invasive mechanical ventilation for COPD exacerbations have an in-hospital mortality rate of up to 30% and survivors have high morbidity and healthcare costs.^70,71 Although invasive mechanical ventilation can typically support the physiologic demands of a patient with COPD, it can lead to dynamic hyperinflation, high intrinsic positive end-expiratory pressure (PEEP), and ventilator-associated pneumonia, require higher levels of sedation, and result in deconditioning and long-term disability in an often older and sicker population.^71–75

Extracorporeal CO₂ removal has been used successfully as an adjunct to, or in lieu of, mechanical ventilation in patients with severe exacerbations of COPD and impending respiratory failure (Figure 3).^65–69 The hypercapnic respiratory failure seen in acute exacerbations of COPD may be ameliorated with an ECCO₂R device that functions at low blood flow rates via a smaller, potentially safer cannula and may obviate the need for invasive mechanical ventilation.⁶⁵ As ECCO₂R may compensate for the exertional dyspnea that would typically limit physical activity, physical rehabilitation can be enhanced relative to that possible with mechanical ventilation. When coupled with an upper-body, dual-lumen cannula, ECCO₂R may allow for reliable and safe ambulation in this population.⁶⁵ This strategy, which conceives the extracorporeal circuit as an artificial lung, has the potential to alter the paradigm for the treatment of patients with severe COPD exacerbations. Larger randomized, clinical trials are in the planning stages.

Figure 3: A patient receiving venovenous extracorporeal carbon dioxide removal as part of a pilot study designed to facilitate early endotracheal extubation and ambulation in patients with severe exacerbations of chronic obstructive pulmonary disease.

Alternative Applications

Extracorporeal support has been used to treat patients with other underlying etiologies of acute respiratory failure, such as status asthmaticus.^76–80 The acute hypercapnia-mediated respiratory acidosis in status asthmaticus is associated with dynamic hyperinflation and high levels of intrinsic PEEP which may result in pneumothorax, hemodynamic compromise, or death.⁸¹ The use of ECCO₂R in this setting is similar to its use in acute exacerbations of COPD.

Venovenous ECMO has been used to manage patients with diffuse alveolar hemorrhage attributed to vasculitis, collagen vascular disease, and other causes.^82–86 Bleeding, particularly alveolar hemorrhage, was typically considered a contraindication to ECMO with older technology because of the high levels of anticoagulation required to maintain the patency of the circuit, as well as the degree of hemolysis that necessitated frequent transfusions.^87–90 However, newer, more biocompatible circuits have a reduced risk of hemorrhage and hemolysis, require only low-dose anticoagulation and, at times, may be run with no anticoagulation at all.^4,91,92

Venoarterial ECMO may be used in patients with massive pulmonary embolism when thrombolysis is contraindicated or ineffective.^93–95 It can be used to support these patients as a bridge to recovery—providing physiologic support while allowing time for systemic anticoagulation to take effect—or as a bridge to surgical embolectomy or possibly catheter-based therapies.⁹⁴ Similar to patients with acute pulmonary embolism, venoarterial ECMO has been used to support women with amniotic fluid emboli after a Cesarean section.^96,97

The Future

Despite a paucity of high-quality evidence, the use of ECMO for the management of adult respiratory failure is increasing throughout the world.¹ More evidence is clearly needed to define the proper role of ECMO and ECCO₂R in patients with severe respiratory failure. Studies are ongoing or in the planning stages for a wide variety of applications.

It is possible that the use of extracorporeal support may extend to those with milder forms of ARDS, using ECCO₂R to facilitate enhanced lung-protective ventilation. Extracorporeal membrane oxygenation either as a bridge to recovery for acute forms of respiratory failure or as a bridge to transplantation for chronic, irreversible respiratory failure may permit more patients to be managed in an awake or extubated manner with reduced sedative use and increased physical therapy, which may have long-term physical, psychiatric, and neurocognitive benefits.⁹⁸

A notable limitation of this technology, however, is that patients must be managed in an intensive care unit at a specialized center capable of performing ECMO. For patients with acute respiratory failure unable to be weaned from ECMO or for those with chronic respiratory failure ineligible for transplantation, no long-term or destination therapy exists. There is currently no pulmonary equivalent to a ventricular-assist device, no total artificial lung that meets a patient’s physiologic needs outside of the intensive care unit, or no lung-replacement center, akin to a dialysis center, where patients could receive intermittent, recurrent treatments. Efforts to decrease the size and improve the efficiency of existing extracorporeal technology, as well as create destination therapies, are underway, and in the future may be able to help those with acute and chronic respiratory failure on both a short- and long-term basis (Figure 4).

Figure 4: Evolving approach to extracorporeal membrane oxygenation and extracorporeal carbon dioxide removal for adult respiratory failure. BTR, bridge to recovery; BTT, bridge to lung transplantation.

Extracorporeal membrane oxygenation and ECCO₂R may demonstrate considerable, continued expansion in the coming years. The use of these devices as an artificial lung may liberate some patients from mechanical ventilation altogether, whereas the prospect of a pulmonary destination device holds the promise of transforming care for patients with chronic or irreversible respiratory failure.