Updates in Pediatric Extracorporeal Membrane Oxygenation

Updates in Pediatric Extracorporeal Membrane Oxygenation

ARTICLE IN PRESS Journal of Cardiothoracic and Vascular Anesthesia 000 (2019) 115 Contents lists available at ScienceDirect Journal of Cardiothorac...

784KB Sizes 0 Downloads 22 Views

ARTICLE IN PRESS Journal of Cardiothoracic and Vascular Anesthesia 000 (2019) 115

Contents lists available at ScienceDirect

Journal of Cardiothoracic and Vascular Anesthesia journal homepage: www.jcvaonline.com

Review Article

Updates in Pediatric Extracorporeal Membrane Oxygenation Eleonore Valencia, MD, Viviane G. Nasr, MD

1

Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, Harvard Medical School, Boston, MA

Extracorporeal membrane oxygenation is an increasingly used mode of life support for patients with cardiac and/or respiratory failure refractory to conventional therapy. This review provides a synopsis of the evolution of extracorporeal life support in neonates, infants, and children and offers a framework for areas in need of research. Specific aspects addressed are the changing epidemiology; technologic advancements in extracorporeal membrane oxygenation circuitry; the current status and future direction of anticoagulation management; sedative and analgesic strategies; and outcomes, with special attention to the lessons learned from neonatal survivors. Ó 2019 Elsevier Inc. All rights reserved. Key Words: pediatric; extracorporeal membrane oxygenation; pharmacokinetic; coagulation

FIRST REPORTED on in 1972, extracorporeal membrane oxygenation (ECMO) is an advanced form of life support that has been used to support more than 100,000 patients with refractory cardiac and/or respiratory failure.1,2 Neonates have comprised the majority of patients, with almost 50% of reported cases.3 In conjunction with the rising incidence of ECMO use, information sharing and research also have flourished internationally. These efforts largely have been facilitated by the Extracorporeal Life Support Organization (ELSO). Even though notable progress has been made to ECMO technology, the rates of complications and mortality remain considerable. Given the increasing use of ECMO support, in particular mobile ECMO, all providers, including pediatric anesthesiologists, should have a thorough understanding of this support modality. This publication reviews the indications, contraindications, and cannulation strategies of ECMO and coagulation and ventilation management in neonates, infants, and children supported with ECMO.

1Address reprint requests to Viviane G. Nasr, MD, Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, 300 Longwood Ave., Boston, MA 02115. E-mail address: [email protected] (V.G. Nasr).

https://doi.org/10.1053/j.jvca.2019.09.006 1053-0770/Ó 2019 Elsevier Inc. All rights reserved.

Epidemiology ELSO has had a pivotal role in chronicling the trends in ECMO use over time. In the span of almost 3 decades, the number of neonatal and pediatric centers reporting to the ELSO registry has more than doubled, with more than 227 centers represented globally in 2015.3,4 Nearly 60,000 neonates (0-28 days) and children (29 days to 17 years) supported with ECMO have been reported on since 1989. Table 1 illustrates the most common indications for ECMO.58 The overall incidence of neonatal ECMO has declined over time. Respiratory failure remains the leading indication in this population; however, the distribution of underlying etiologies has evolved. Congenital diaphragmatic hernia (CDH) has become the most common diagnosis, superseding meconium aspiration syndrome (MAS) and persistent pulmonary hypertension of the newborn, likely attributable to the advent of high-frequency oscillatory ventilation, inhaled nitric oxide, and surfactant.9,10 Venoarterial ECMO (VA ECMO) is the primary mode used to support this group.4 Despite the decreasing trend in ECMO for neonatal respiratory failure, its use for primary cardiac failure has increased. Congenital heart disease is the most common indication. The most common subtype specified in the ELSO registry is hypoplastic left heart syndrome.11

ARTICLE IN PRESS 2

E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

Table 1 Relative Indications for ECMO Support According to Age Group ECMO Type

Age Group

Relative Indications

Cardiac5

Neonatal,* pediatricy

Structural congenital heart disease  Perioperative support Nonstructural congenital or acquired heart disease  Cardiomyopathy  Myocarditis  Myocardial infarction  Arrhythmias  Acute right-sided heart failure secondary to pulmonary hypertensive crisis  Hemodynamic deterioration secondary to intoxication  Heart failure  Bridge to decision regarding heart transplantation  Bridge to ventricular assist device  Bridge to heart transplantation  Postheart transplantation support  High-risk cardiac catheterization  Periprocedural support Impaired gas exchange as defined by OI  Recommended threshold OI 40  Consider transfer to ECMO center for OI 25 Impaired gas exchange as defined by OI or respiratory acidosis  No threshold value for oxygenation; however, consider if OI not improving by day 2 of illness  Recommended threshold pH < 7.0 Consider if sustained in-hospital cardiac arrest intractable to conventional CPR and if arrest occurred at an experienced ECMO center

Respiratory6,7

Neonatal*

Pediatricy

ECPR8

Neonatal,* pediatricy

Abbreviations: CPR, cardiopulmonary resuscitation; ECMO, extracorporeal membrane oxygenation; ECPR, extracorporeal cardiopulmonary resuscitation; OI, oxygenation index. * Neonatal: 0 to 28 days. y Pediatric: 29 days to 17 years.

The pediatric population has experienced a rapid upsurge in ECMO use for both respiratory and cardiac support. Primary lung infection (bronchiolitis, pneumonia, pertussis) continues to be the leading diagnosis for those with respiratory failure. Venovenous (VV) ECMO has become the predominant mode to support children with respiratory failure over the past decade.3,4 Congenital heart disease also is the most common ECMO indication for children with cardiac failure, with leftto-right shunts being the most frequently indicated subtype.11

Extracorporeal cardiopulmonary resuscitation (ECPR) is an increasingly used therapy to reestablish circulation in patients who have sustained cardiac arrest intractable to conventional cardiopulmonary resuscitation (CPR). During ECPR, VA ECMO cannulation is performed emergently while CPR is ongoing with the goal of achieving return of circulation via extracorporeal flows. This trend has been particularly notable in the pediatric population compared with neonates and adults. Importantly, the majority of arrests have been witnessed, have occurred in acute care units, and have led to ECMO cannulation within 60 minutes.4 Circuitry An ECMO circuit is composed of the following 3 main components: a mechanical pump, a membrane lung, and a heat exchanger. These devices are interconnected via circuit tubing that is linked to at least 2 cannulae, providing inflow to and outflow from the patient (Fig 1). Additional safety and monitoring components also are interspersed throughout the circuit. A bridge may be incorporated between the proximal inflow and outflow limbs as a means to maintain circuit patency during low-flow states, such as during weaning. The machinery represents an area of considerable evolution in the field of ECMO. Pump The pump propels blood through the circuit, which allows for gas exchange via the membrane lung, and supports hemodynamics via varying flow rates in the case of VA ECMO. There are 2 primary types of pumps—roller and centrifugal. Roller pumps actively propel blood forward via a rotating roller head that compresses blood-containing tubing against a plate. A bladder (reservoir) is added to the circuitry, in between the outflow cannula and the pump, to provide a safety mechanism against vascular or red blood cell injury induced by the negative suction pressure. The bladder also mitigates trauma to red blood cells by decreasing the pump speed when the drainage volume is inadequate. An important issue with the roller pump is that it does not decelerate when there is increased distal resistance, placing the circuit at risk of rupture.12 The centrifugal pump uses a vortex technology to actively pull blood in and then push it forward via the centrifugal forces generated. Because forward flow from the pump is passive, there is no risk of circuit rupture in the presence of a distal obstruction. Considerable advancements have been made to the centrifugal pump technology to lessen the risk of complications, such as thrombosis and hemolysis with resultant renal injury. Newer generation designs have addressed such issues via the incorporation of magnetic levitation or single pivot bearings in place of multiple bearings, shafts, and seals to minimize friction and heat generation, as well as washout holes to prevent stagnant blood flow.12 In a laboratory-based study comparing 2 new generation centrifugal pumps with a conventional centrifugal pump and a conventional roller pump, the conventional centrifugal pump had a nearly 2-fold increased rate of hemolysis. There was no difference in the degree of

ARTICLE IN PRESS E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

3

Fig 1. Schematic of a neonatal extracorporeal membrane oxygenation circuit. Deoxygenated blood drains via an outflow cannula into a pump, by which it is propelled forward through the membrane oxygenator for gas exchange and heat exchanger for warming. Oxygenated blood subsequently is returned to the patient via the inflow cannula. A bladder (optional) can be implemented to mitigate red blood cell and vascular injury induced by the pump, a particular consideration for roller pumps. A bridge (optional) also can be used in between the proximal limbs to allow for ongoing flow within the circuit at an adequate rate to prevent thrombosis, should the patient be weaning or need to be disconnected temporarily. CO2, carbon dioxide; O2, oxygen.

hemolysis among the newer generation centrifugal pumps and the roller pump.13 However, several studies have since presented conflicting data, with some demonstrating an increased incidence of hemolysis with novel centrifugal pumps compared with roller pumps.1416 It is challenging to make a definitive conclusion, given the retrospective nature of these studies in addition to the multifactorial etiology of hemolysis. Finally, in addition to these improvements, centrifugal pumps, such as the Maquet Cardiohelp (Maquet Cardiovascular, Rastatt, Germany) have become more compact, allowing for smaller priming volumes, portability, and patient rehabilitation.17 Since the evolution of the centrifugal pump, roller pump use has decreased. The neonatal population is the only group that predominantly uses roller pumps in the present, and its use has declined over the past 2 decades as well.4,1820 Membrane Lung The membrane lung, also referred to as the membrane oxygenator, provides gas exchange for the patient. It has

undergone substantial innovation since the initially reported Bramson lung, a cumbersome parallel plate system composed of several conduits of blood enveloped by concentric silicone membranes that facilitate oxygenation. Hollow-fiber technology, originating from hemodialysis, was taken up shortly thereafter and has since been used and adapted further to provide a safer and more efficient membrane. These membranes are composed of thousands of hollow fibers, through which gas traverses and around which blood flows. The multitude of fibers affords a large surface area within a compact device. Gas exchange occurs via diffusion across the microporous fiber interface. Polyprolene initially supplanted the silicone material as a more efficient means of gas exchange. However, plasma leakage into the gas-filled Polyprolene fiber impeded the diffusion of gases. Polymethylpentene (PMP) now predominates the use of Polyprolene and silicone membranes because it is both impermeable to plasma and proficient at gas exchange.4,21 Serial surveys by Lawson et al. demonstrated that at least 70% of neonatal ELSO centers used PMP membrane oxygenators in 2011 compared with 14% in 2008.1920

ARTICLE IN PRESS 4

E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

Several of the newer PMP membrane oxygenators also contain the heat exchanger, allowing for a more efficient ECMO circuit.17 Surface Material The prosthetic circuit elicits a systemic inflammatory response, subsequently leading to a hypercoagulable state, upon contact with a patient’s blood.22,23 Various surface coatings with heparin, albumin, and assorted polymers have been developed in an effort to simulate the hemostatic balance typically imparted by the vascular endothelium.24 Heparin-bonded circuits were introduced decades ago in an attempt to lessen the hemorrhagic risks associated with systemic anticoagulation therapy. Although studies have demonstrated decreased in vivo levels of inflammatory markers, heparin-bonded circuits have had a variable effect on clinical outcomes and have not eliminated the need for systemic anticoagulation therapy.25,26 Phosphorylcholine is a hydrophilic, nonthrombogenic polymer that is commercially available for use as ECMO circuit tubing. Adult cardiopulmonary bypass (CPB) studies suggest that phosphorylcholine-coated circuits may allow for a lower level of systemic heparin administration.27 However, it has otherwise not been shown to have an advantage in the degree of inflammation or clinical outcomes compared with heparinbonded circuits.2829 Poly-2-methoxyethyl acrylate is another commercially available, less hydrophilic, polymeric material that has been shown to have antiinflammatory and antithrombogenic properties in pediatric CPB.30 However, interestingly, it also has been associated with a postoperative systemic inflammatory response syndrome in a small cohort of infants who were randomly assigned to undergo CPB with poly-2methoxyethyl acrylatecoated versus heparin-coated circuits.31 Importantly, CPB is the predominant model under which these surface modifications have been studied and therefore does not account for important ECMO characteristics that influence thrombosis, like longer run duration. Cannulation Cannula selection and cannulation technique are determined by the mode of ECMO support, patient age and size,

underlying diagnosis, and institution preference (Table 2). ECMO cannulae are commonly made of polyurethane and a wire mesh to provide infrastructure against kinking. Similar to the circuit tubing, the cannulae also can be heparin-bonded to mitigate the inflammatory and prothrombotic stimuli of the synthetic material. Single lumen cannulae with 1 hole at the end of the tip or multiple fenestrations at the tip are used to provide arterial and venous access, respectively, via the carotid and jugular or femoral vessels. Alternatively, single-site VV ECMO support can be used via the use of a jugular double lumen cannula, which incorporates drainage and reinfusion ports. VV dual-lumen (VVDL) cannulae offer several advantages, including a decrease in the risks associated with multiple access points, such as infection and surgical site bleeding; ability to provide the VV mode to children whose femoral vessels are otherwise too small; increased patient mobility; and a decrease in the amount of recirculation when well seated.32 Recirculation, the phenomenon in which oxygenated blood from the inflow cannula is pulled into the outflow cannula before reaching systemic circulation, leading to desaturation and decreased flow, is more commonly an issue with multiple site VV ECMO. Despite these potential benefits, VVDL cannulae must be well positioned so as not to impede flow. This can be particularly challenging in neonates and small children and result in significant morbidity, including vascular trauma and myocardial perforation with resultant tamponade. For instance, the bicaval Avalon cannulae (Getinge AB, Gothenburg, Sweden) must be situated with the drainage ports in the superior and inferior venae cavae and the reinfusion port in the right atrium directed at the tricuspid valve. The use of echocardiography and fluoroscopy has been shown to decrease the need for bicaval VVDL cannula repositioning and the associated complications.33 Another type of VVDL cannula, the OriGen (OriGen Biomedical, Austin, TX), must be positioned with its drainage and reinfusion ports in the right atrium. Although seemingly less complicated, a recent small study demonstrated no need for repositioning in neonatal and pediatric patients who underwent echocardiography with initial OriGen cannula placement compared with 20% who did not have echocardiographic guidance and ultimately required repositioning. Interestingly, the rate of complications, including atrial perforation

Table 2 Modes of ECMO Support Mode

Indication

Outflow Cannula(e)

Inflow Cannula(e)

Venoarterial (VA) Venovenous (VV) Venovenous arterial (VVA) Venoarterial venous (VAV)

Primary cardiac failure § respiratory failure Primary respiratory failure without refractory cardiovascular compromise Primary respiratory failure with subsequent refractory cardiovascular compromise Primary cardiac and respiratory failure with subsequent North-South syndrome*

Venovenous venous (VVV)

Inadequate venous drainage Inadequate venous drainage

V V V V V, V V, V

A V V, A A, V A V

Abbreviation: ECMO, extracorporeal membrane oxygenation. * North-South syndrome: phenomenon in which there is hypoxemia of the upper body, including the brain, owing to recovery of native myocardial function that competes with retrograde inflow from the femoral cannula. Implementation of an additional inflow cannula to the internal jugular vein can improve regional oxygenation.

ARTICLE IN PRESS E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

and cannula site bleeding, was similar in each group.34 The application of fluoroscopic guidance may minimize the adverse events associated with VVDL cannulation, although the incidence of such events and whether fluoroscopy would be an efficient use of resources has yet to be studied.33 The site for VA cannulation is primarily dependent on patient age and indication. In central (transthoracic) cannulation, which is primarily used for failure to wean from CPB, the aorta and right atrium are accessed. In peripheral VA ECMO, access for infants and young children typically is via the common carotid artery and internal jugular vein because the femoral vessels are too small to accommodate cannulation. The optimal time at which to transition from neck to femoral cannulation is unknown. In a 2018 survey of the American Pediatric Surgical Association, age, in contrast to weight, was the predominant determining factor. Five years was the age at which more surgeons would consider femoral cannulation; however, neck cannulation remained the preference for all pediatric age groups.35 Selection of the neck versus femoral vessels in older children has been debated and largely driven by the concern for adverse acute and long-term neurologic sequelae after carotid artery cannulation and ligation. The risk of neurologic morbidity reported in the literature is inconsistent. A recent large evaluation of the ELSO registry did not demonstrate a significant difference in the incidence of neurologic complications (ischemic stroke, intracranial hemorrhage, seizures, brain death) between carotid versus femoral arterial cannulation in neonatal and pediatric patients after correcting for age, severity of illness, or mode of ECMO support.36 In contrast, several other analyses of neonatal and pediatric patients from the ELSO registry showed a significantly increased rate of stroke with carotid cannulation, compared with femoral or aortic cannulation, that was independent of age.37,38 These studies are limited by their retrospective nature. Femoral access is not without consequence, given the risk of ischemic injury to the distal limb with the subsequent potential need for vascular reconstruction, amputation, or other surgical intervention. Extrapolated from the adult experience, additional reperfusion cannulae are commonly placed to provide blood flow to the ipsilateral limb, either as a preventive strategy or therapeutic measure after ischemia has occurred. The true incidence of limb ischemia in pediatric patients supported with VA ECMO via femoral cannulation is unknown, although 2 case series reported incidences of >50%.39,40 In one of these case series, 4 of 11 patients (36%) with ischemic limbs required interventions, including 1 vascular reconstruction, 2 amputations, and 1 fasciotomy. Two of these patients had undergone prior reperfusion cannula placement. That study did not identify any risk factors associated with the development of limb ischemia, including the ratio of cannula size-tobody surface area.39 In the second case series, 4 of 7 patients (57%) who sustained ischemic injury and had reperfusion cannulae in place required vascular reconstruction, 1 of whom also ultimately required an amputation.40 There has been a shift from the VA to VV mode to provide primary respiratory support in pediatric patients. The proposed

5

benefits of VV ECMO include the elimination of the aforementioned arterial complications and preservation of more physiological hemodynamics. Multiple retrospective, registrybased studies have demonstrated improved survival and outcomes in pediatric patients supported with VV versus VA ECMO for various indications, such as sepsis and acute respiratory failure.41,42 There also is an evolving body of literature, largely limited to the adult population, suggesting that a decreased level of anticoagulation may be safe in VV ECMO, which could be especially impactful for patients in whom preexisting hemorrhage prohibits cannulation.4345 In contrast to the pediatric population, neonatal VV ECMO use has not followed the same trend. This may be explained partially by the limitation in the size of venous cannulae commercially available.17 Nonetheless, the application of VV ECMO in neonates remains under study. An international survey of pediatric surgeons’ opinions regarding the management of neonates with CDH showed a preference for VV ECMO.46 A recent propensity score-matching study comparing neonates with CDH who were cannulated for VA versus VV ECMO found no overall difference in mortality. However, there was a decreased incidence of neurologic morbidity in the subgroup that was treated with VV ECMO before CDH repair.47 The advent of percutaneous equipment has provided the opportunity for nonsurgical clinicians to perform cannulations. This has been predominantly explored and found to be successful in the adult ECMO population.48 In contrast, in the American Pediatric Surgical Association’s survey by Garcia et al., nearly all pediatric cannulations were performed by general or cardiac surgeons.35 The training of anesthesiologists, intensivists, cardiologists, interventional radiologists, and other nonsurgical physicians may enhance the accessibility of ECMO support. Left Heart Decompression In patients requiring VA ECMO support for refractory cardiogenic shock, in particular those without a preexisting intracardiac shunt (eg, atrial septal defect), left heart decompression can be performed to facilitate myocardial recovery by unloading the distended left ventricle and therefore relieving wall stress. In addition, decompression can mitigate the complications of left atrial hypertension, including pulmonary edema and pulmonary hemorrhage. Decompression can be achieved percutaneously via atrial septostomy (balloon, blade, stent) or placement of a venting cannula in the atrium or ventricle under fluoroscopy or echocardiography in the catheterization laboratory or at the bedside. Surgical atrial septostomy or atrial or ventricular cannulation also can be performed, most commonly at the time of central VA ECMO cannulation. The rate at which these interventions are used in the neonatal and pediatric populations is unknown, although single institution experiences have reported incidences of 10.55% to 12.9%.49,50 Frequently cited underlying etiologies include cardiomyopathy, myocarditis, and low cardiac output postcardiotomy. No survival benefit has been demonstrated in patients who have undergone left heart decompression,

ARTICLE IN PRESS 6

E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

although these studies were likely underpowered.49,50 However, a recent multicenter study showed a significant reduction in ECMO duration with early (<18 h) decompression.51 In addition to the aforementioned strategies, the percutaneous Impella (Abiomed, Danvers, MA) ventricular assist device has been evaluated as an alternative means to decompress the left side of the heart. The advantage of the Impella is the potential for reduction in afterload, an inherent issue with peripheral VA ECMO secondary to retrograde flow from the femoral artery to the aorta. A recent retrospective study demonstrated a significant survival benefit in adults with cardiogenic shock who were treated with peripheral VA ECMO plus Impella device placement within 24 hours of cannulation compared with those who underwent conventional or no decompression.52 A small case series of 8 children (22.1 kg) with cardiogenic shock reported effective left heart decompression with the Impella; however, the pediatric literature regarding its benefit over conventional decompression is otherwise lacking, likely in part owing to the limited availability of Impella catheters to accommodate pediatric sizes.53

high PEEP approach. Keszler et al. showed that neonates with primary respiratory failure who were randomly assigned to a high-PEEP (12-14 cmH2O) strategy had a significantly decreased duration of ECMO and improved lung compliance compared with those treated with low PEEP (3-5 cmH2O). Notably, although the study was insufficiently powered to detect a significant difference in mortality, all 4 patients who died were in the low-PEEP group.59 A more recent retrospective analysis of neonates with acute respiratory failure from the ELSO database demonstrated similar findings and, importantly, no significant difference was found in the rates of pulmonary or neurologic complications, including air leak, seizures, intracranial hemorrhage (ICH), or infarction, between the groups.60 Finally, the concept of near apneic ventilation, in which the rate is decreased to subphysiological levels (eg, 5) in combination with adequate PEEP and a minimal driving pressure, remains under study. Although ECMO offers a promising opportunity for lung recovery, further research is needed to more explicitly guide lung protective ventilator management with particular attention to the pathophysiological differences between neonates and children.

Mechanical Ventilation During ECMO Weaning ECMO There are no specific guidelines regarding the method by which to mechanically ventilate patients supported with ECMO for cardiac or respiratory failure. ELSO recommends a “lung rest” approach to lessen the risk of ventilator-induced lung injury until the acute inflammatory phase has defervesced.54 Important considerations include minimizing plateau pressure with target tidal volumes of 6 mL/kg (predicted body weight); titrating positive end-expiratory pressure (PEEP) to mitigate atelectasis while balancing the potential for adverse hemodynamic effects; decreasing the respiratory rate; and allowing permissive hypoxemia to minimize pulmonary toxicity while maintaining adequate end-organ oxygen delivery. In a 2013 international survey of 141 neonatal, pediatric, and adult ECMO centers, only 27% reported using a specific protocol to manage mechanical ventilation in patients with severe acute respiratory distress syndrome (ARDS) on VV ECMO. More generally, 77% indicated using a “lung rest” approach with approximately 75% titrating tidal volumes to 6 mL/kg for both children and adults alike. Whereas the majority of centers used a target PEEP of 6 to 10 cmH2O, there was a 3-fold greater use of PEEP >11 cmH2O in adult patients.55 There is a lack of pediatric-specific research on mechanical ventilation strategies during ECMO. Some of the recommendations have been extrapolated from the adult ARDS experience.56 In the 2013 propensity-matched analysis of adults with ARDS secondary to influenza A treated with or without ECMO, a subgroup analysis of the ECMO patients demonstrated a significantly decreased rate of mortality in those managed with lower plateau pressures.57 In addition, a large, multi-institutional retrospective analysis of adult patients supported with ECMO for severe ARDS found a significant survival benefit in patients treated with higher PEEP.58 A few neonatal studies also have demonstrated an advantage for a

In the case of weaning ECMO for primary respiratory failure, compliance and oxygenation can be assessed routinely with reasonable ventilator settings (peak pressure 25 cmH2O, fraction of inspired oxygen <50%) during the ECMO run. The methods by which to wean include decreasing sweep gas flow rate, the fraction of delivered oxygen, and/or the pump flow rate. According to the 2013 survey by Marhong et al., the majority of centers reported decreasing the sweep gas as the preferential method to wean VV ECMO.55 For primary cardiac failure, improvement in pulse pressure, end-tidal carbon dioxide (if no cardiac shunt), inotropic need, and function as measured using echocardiography can signify myocardial recovery. Importantly, echocardiographic assessment should be performed under optimal loading conditions, necessitating decreased flow rates of the ECMO circuit and the decompression device, if present. The pump flow rate is the only method by which to wean VA ECMO for cardiac failure. Anticoagulation Systemic anticoagulation is an essential but challenging component of ECMO management. Upon contact with the foreign ECMO circuitry, the body elicits a complex inflammatory reaction, which subsequently induces a prothrombotic condition that could lead to serious morbidity in the patient and compromise the viability of the circuit.61 However, the risk of thrombosis must be balanced carefully with the potential for hemorrhage because many of these patients already may have acquired coagulopathy from their critical illness. In the case of excessive bleeding, the transfusion thresholds at which platelets, fresh frozen plasma, and cryoprecipitate are administered can be lowered; the dose of systemic anticoagulation can be titrated; and antifibrinolytic therapy can be given, depending

ARTICLE IN PRESS E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

on the severity. Recombinant activated factor VII has been evaluated for refractory hemorrhage during ECMO; however, the overall experience is limited, and pediatric case reports have described variable efficacy.6264 Its use should be considered with caution, given the risk of precipitating, detrimental thrombosis and the lack of supporting evidence. Unfractionated Heparin Unfractionated heparin (UFH) remains the gold standard therapy for systemic anticoagulation in neonatal and pediatric patients. In an international survey of nearly 120 ELSO centers, the majority of which were neonatal or pediatric, 100% reported use of continuous infusions of UFH for ECMO anticoagulation management. Furthermore, 72% indicated that they had no upper threshold to the maximum dose of UFH needed to achieve their targeted anticoagulation goal.65 UFH is practical given the extensive experience with its use and the availability of protamine as a rapid reversal agent; however, it also presents important challenges. One of the primary issues is UFH’s dependence on antithrombin (AT) III for effect, particularly in neonates who have lower physiological levels than children and adults. Neonates have an even more pronounced AT III deficiency in the setting of critical illness and ECMO.66,67 The benefit of AT III repletion has yet to be proven. Several retrospective analyses of neonates who received AT III for variably low plasma levels (<65%-100%) have inconsistently showed a decreased transfusion requirement. Otherwise, there is no evidence to indicate that AT III administration improves circuit viability or decreases the dose of UFH required to meet anticoagulation goals. Importantly, there was no increased incidence of hemorrhage with AT III administration in any of these studies.6870 Additional disadvantages to UFH include the heterogeneity of its molecular weight leading to inconsistent pharmacologic properties and the risk of heparin-induced thrombocytopenia.71,72 Direct Thrombin Inhibitors There is an early but evolving ECMO experience with direct thrombin inhibitors (DTIs), a more novel class of anticoagulants. DTIs are a promising alternative because they do not rely on any cofactor for effect, have more predictable pharmacologic characteristics, and have affinity for fibrin-bound thrombin. There is no proven reversal agent, although some subtypes, such as bivalirudin, have a rapid half-life, can be dialyzed, and potentially may be reversed by recombinant factor VII.7375 According to the aforementioned survey, as of 2011, 8% of ECMO centers reported using DTIs as a non-UFH means of anticoagulation.65 There since have been increasing reports of DTI use in neonatal and pediatric ECMO patients, in particular with bivalirudin.76 Several case series regarding the use of bivalirudin and CPB and ECMO have been published.77,78 One such case series compared 2 groups of postcardiotomy ECMO patients, including 4 neonates and 6 children, who received either UFH or bivalirudin as the first-line anticoagulant.

7

Boluses were not administered before the start of the infusion, and initial dosing and titration of the infusions were standardized. The UFH group experienced significantly more bleeding and transfusion requirements than did the bivalirudin group. Moreover, the daily ECMO cost was greater in the UFH group, and this difference was particularly pronounced in the pediatric patients. The overall mortality rate was high (76%), although there was no difference between the groups.79 A second case series reported the use of bivalirudin for the anticoagulation management of 12 neonatal and pediatric patients undergoing VV or VA ECMO after the development of heparininduced thrombocytopenia, heparin resistance, or significant thrombosis. Bolus administration and dosing and adjustment of the continuous infusion varied among patients; however, there was no correlation of dosing with age. Two of the patients experienced bleeding from their chest tubes; otherwise, there was no identified ICH, and the observed mortality rate was equivalent to that reported by the ELSO registry.80 The literature for bivalirudin is promising; however, more rigorous research is needed to better understand its pharmacologic properties. Antiplatelet Agents Approximately 10% of institutions indicated adjunctive use of antiplatelet therapy.65 However, the pediatric literature regarding the effectiveness and adverse effects of antiplatelet agents during ECMO is lacking. Monitoring There is no standard assay to monitor the anticoagulant effect of UFH in neonatal and pediatric patients supported with ECMO. In the survey by Bembea et al., 97% of centers reported the use of activated clotting time (ACT), 94% activated partial thromboplastin time (aPTT), and 65% anti-factor Xa.65 Even though ACT offers the benefit of point-of-care testing, elevated levels are not specific to the anticoagulant effects of UFH but rather can indicate alternative derangements, including platelet dysfunction, thrombocytopenia, hypofibrinogenemia, coagulation factor deficiency, and hemodilution. aPTT is less equivocal; however, it also can be prolonged by low levels of the intrinsic pathway coagulation factors and hemodilution. In addition, the normal aPTT range is age dependent, with neonates and young children having longer baseline levels. Anti-factor Xa is the most direct measure of UFH that is clinically available; however, high levels can be confounded by hyperbilirubinemia and increased plasma-free hemoglobin.81 Several retrospective and prospective studies have evaluated the association among ACT, aPTT, and anti-factor Xa and UFH dose, as well as clinical outcomes, in neonatal and pediatric ECMO. ACT has been shown consistently to have no correlation with UFH dose and, in fact, underestimates the degree of anticoagulation with resultant hemorrhagic complications.8284 aPTT has better correlation with UFH dose, although it has

ARTICLE IN PRESS 8

E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

been associated with increased thrombotic complications in both the circuitry and patient, despite therapeutic levels.82 Anti-factor Xa has superior correlation to UFH dose, and furthermore, low levels have been associated with increased thrombosis.85 A more wide-ranging approach may be necessary to distinguish the anticoagulant effect of UFH from other etiologies that contribute to a patient’s coagulopathy, allowing for more appropriately targeted interventions (eg, UFH titration v repletion of coagulation factors). A study by Northrop et al. demonstrated significantly decreased bleeding and thrombotic complications after implementation of a comprehensive monitoring protocol for UFH titration, in which antiXa was used as the primary assay in conjunction with ACT, aPTT, AT III, and thromboelastography.86 Sedation and Analgesia The management of sedation and analgesia for neonatal and pediatric patients supported with ECMO requires a deliberate approach in order to optimize comfort and safety while minimizing adverse effects. A key challenge in the delivery of sedation and analgesia is the alteration of pharmacokinetics during ECMO, specifically increased volume of distribution, decreased elimination, and drug adsorption by the circuit.87 Several in vitro studies, including a recently published study that incorporated currently used equipment (PMP membrane oxygenators), have demonstrated an increased amount of sequestration associated with more lipophilic drugs, such as fentanyl and midazolam.8890 Despite this known and replicated phenomenon, a multi-institutional survey of pediatric ECMO centers in the United States revealed that 60% of centers use fentanyl as a first-line medication.91 Another retrospective analysis of a single center’s sedative and analgesic practice in more than 150 neonatal and pediatric ECMO patients also demonstrated a preference for fentanyl. Interestingly, the authors found that different subsets of patients had disparate needs. In particular, children with primary respiratory failure were more difficult to manage because they required larger doses of opiates and benzodiazepines and had more frequent use of adjuncts such as dexmedetomidine.92 The RESTORE trial (Randomized Evaluation of Sedation Titration for Respiratory Failure) was a prospective, multiinstitutional trial that specifically focused on neonatal and pediatric patients with ARDS and randomly assigned them to receive sedation per a goal-directed, nursing-driven protocol versus usual care. In a secondary analysis, that study found that the majority of ECMO patients were intentionally deeply sedated or chemically paralyzed compared with their nonECMO, mechanically ventilated counterparts. As a result, the analgesic and sedative doses more than doubled by decannulation.93 These studies also highlight the adverse effects associated with increased doses of opiates and benzodiazepines, including withdrawal, longer duration of mechanical ventilation, and increased length of stay in the intensive care unit.92,93 There is no literature to date that describes the use of dexmedetomidine as a primary analgesic and sedative agent in pediatric patients supported with ECMO.

Outcomes Survival Sixty-one percent of neonates, infants, and children supported with ECMO have survived since the original ELSO report in 1988. Currently, neonates with respiratory failure have the highest rate of survival followed by children with respiratory failure, with MAS and asthma demonstrating the best prognosis in each age group, respectively. Both neonatal and pediatric patients who undergo ECPR have the lowest survival rates.4 The plateau in neonatal and pediatric mortality over time is complex and likely related to multiple patient and institutional factors. Notably, the number of children with 1 or more underlying comorbid conditions who have undergone ECMO therapy has increased substantially over time, with one study reporting a 28% increase from 1993 to 2007. These children were found to have a greater risk of mortality, in particular those with liver failure, cancer, immunodeficiency, or renal failure.94 A longer duration of conventional therapy before initiation of ECMO also has been identified as a risk factor.95 From an institutional standpoint, a significant association between a hospital’s ECMO case volume and rate of mortality has been described multiple times. Centers with a higher volume of ECMO cases per year have demonstrated a lower rate of mortality, in particular those that are pediatric specific, in which >90% of patients are 18 years old.9699 Moreover, a recent analysis of neonatal and pediatric cardiac surgical patients from the Pediatric Health Information System identified a stronger association between cardiac surgical case volume and mortality compared with cardiac ECMO volume.100 Large hospitals, classified according to number of beds, also have been correlated with a decreased mortality rate.95 These studies provide convincing evidence that neonatal and pediatric ECMO may benefit from greater regionalization, particularly in the United States, where there are well over 100 neonatal and/or pediatric centers. The early published Italian experience has demonstrated that neonatal and pediatric patients requiring ECMO can be cannulated safely at and transported from non-ECMO centers via the implementation of a protocolled referral process and a trained multidisciplinary team.101 Mobile ECMO is a globally evolving and promising domain that may allow for such regionalization. Certain diagnostic categories, such as pediatric respiratory failure, have demonstrated a substantial improvement in survival over time. The pediatric sepsis experience represents a particular noteworthy example, for which survival now is projected to be >50%, despite an increasing number of children with multiorgan dysfunction and comorbidities at time of cannulation.4,102,103 Single-institution experiences have provided some elucidation as to why the survival rate may have improved. In a small retrospective study by Sole et al. that spanned a 15-year period, survival increased from 14.3% during the first 7 years to 57.2% during the subsequent 8 years. One particular notable and significant finding between the earlier and later subgroups was a decreased median time to

ARTICLE IN PRESS E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

cannulation from the initial presentation. The authors also identified a significantly increased rate of mortality in children who had a greater severity of vasoplegia.104 Importantly, a case series by MacLaren et al. described a considerably improved survival rate of 74% in a cohort of children who underwent central cannulation, allowing for higher flow rates to combat severe vasoplegia.105 A few tools have been developed recently to predict the risk of mortality in children undergoing ECMO for primary respiratory failure, including the Pediatric Risk Estimate Score for Children Using Extracorporeal Respiratory Support (PedRESCUERS) and the Pediatric Pulmonary Rescue with Extracorporeal Membrane Oxygenation Prediction (P-PREP) score (Table 3). The Ped-RESCUERS model is a 9-variable score that was derived and internally validated using data from the ELSO registry. Its predictive ability was fair in discriminating mortality on receiver operating characteristic analysis. It has yet to undergo external validation.106 The P-PREP score is a 7-variable model that was derived and internally validated using the ELSO registry dataset and subsequently externally validated using the Pediatric Health Information System

9

dataset. Comparable to its performance on internal validation, the P-PREP score had fair performance on retrospective external validation.107 Both of these risk prediction tools are limited by the availability of information from such databases. The ability to accurately predict which children have a higher risk of mortality with ECMO compared with conventional therapy could improve outcomes and resource allocation substantially. Morbidity ECMO-related complications have been associated with an increased risk of mortality and secondary morbidity that can have long-lasting effects on survivors. According to the 2016 Pediatric ELSO Report, ICH is the most frequently occurring complication in neonates with primary respiratory failure. Surgical site bleeding is the most common complication in the remainder of the neonatal and pediatric populations. There are up to 39% and 12% increases in mortality for these patients who sustain ICH and surgical site bleeding, respectively. Gastrointestinal hemorrhage also results in a considerably increased rate of mortality of up to 35%.4

Table 3 Pre-ECMO Mortality Prediction Tools for Pediatric Patients with Primary Respiratory Failure

Use Variables

Discriminative ability

Ped-RESCUERS106

P-PREP107

Pre-ECMO probability of mortality in children 29 d to <18 y with primary respiratory failure pH PCO2 Hours from admission to ECMO center to cannulationHours from intubation to ECMO cannulationType of mechanical ventilation  CMV  HFOV Presence of specific pulmonary diagnosis  Pertussis  Asthma  Bronchiolitis Presence of comorbid malignancyMean arterial pressure  CMV  HFOV Administration of milrinone before ECMO initiation

Pre-ECMO probability of mortality in children >7 d to <18 y with primary respiratory failure pH  <7.11 v 7.11-7.34  >7.34 v 7.11-7.34 Mode of ECMO  VA  VV Duration of mechanical ventilation  14 d  >14 d PaO2/FiO2 ratio  300 mmHg  201-300 mmHg  101-200 mmHg  100 mmHg Year of ECMO* Presence of specific pulmonary diagnosis  Asthma  Aspiration  RSV  Sepsis-induced ARDS  Pertussis Presence of any specific comorbidity  Cardiac arrest before ECMO cannulation  Cancer  Acute renal failure  Acute liver necrosis AUC 0.69 (95% CI 0.67-0.71)y AUC 0.66 (95% CI 0.63-0.69)z

AUC 0.635 (95% CI 0.594-0.651)y

Abbreviations: ARDS, acute respiratory distress syndrome; AUC, area under the curve; CI, confidence interval; CMV, conventional mechanical ventilation; ECMO, extracorporeal membrane oxygenation; FiO2, fraction of inspired oxygen; HFOV, high-frequency oscillatory ventilation; PaO2, partial pressure of arterial oxygen; PCO2, partial pressure of carbon dioxide; RSV, respiratory syncytial virus; VA, venoarterial; VV, venovenous. * Association of decreased rate of mortality with more recent ECMO year. y Internal validation. z External validation.

ARTICLE IN PRESS 10

E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

The neurologic sequelae of ECMO can be detrimental with the potential for long-term morbidity. Among the most severe complications are seizures, ICH, ischemic infarction, and brain death. Frequently described risk factors include age <30 days; prematurity; pre-ECMO severity of illness (eg, acidosis, inotropic support); diagnostic indication; and CPR before or during (ECPR) cannulation.108110 Given their portability, head computed tomography, head ultrasound, and electroencephalography are the primary modalities used to assess for neurologic injury. The use of head ultrasound is limited by the presence of an adequately sized fontanel. Brain magnetic resonance imaging is obtained variably post-ECMO. No standard practice in regard to timing, indication, or other patient characteristics has been described. Although it obviates the risk of radiation, head ultrasound has been reported to have low specificity for ischemic infarction compared with computed tomography.111112 Cerebral near-infrared spectroscopy (NIRS) has been evaluated as an alternative modality to detect the occurrence of neurologic events during ECMO. Small, retrospective adult studies have demonstrated that patients with acute, sustained desaturation as measured using cerebral NIRS were more likely to have sustained a cerebrovascular accident.113114 The pediatric literature is more limited. One study assessed the utility of cerebral frequency-domain NIRS to detect intracranial events in a series of 7 neonatal and pediatric patients undergoing VV and VA ECMO. That study demonstrated that frequency-domain NIRS may provide a signal to volume loss in the frontoparietal regions; however, it was not sensitive for volume loss in the deeper parenchymal tissues. Furthermore, its value in the diagnosis of hemorrhage could not be determined because none of the patients sustained an ICH.115 Although NIRS provides the benefits of noninvasive and real-time monitoring, it has yet to be proven to have adequate sensitivity or specificity for the detection of important neurologic complications. The long-term neurologic outcomes of pediatric ECMO survivors have been widely studied over the past 2 decades. Several groups have identified gross motor delays in elementary school age children and ongoing neuropsychological deficits, such as spatial ability, memory, and working speed, through adolescence.116118 It is imperative to note that the neurodevelopmental assessments of children in these follow-up studies were compared with normal reference ranges. The United Kingdom (UK) Randomized Trial of Neonatal Extracorporeal Membrane Oxygenation is a valuable study regarding the effect of ECMO on neurodevelopment. Two cohorts of infants who were 35 weeks gestational age and had similar types (CDH, MAS, pulmonary hypertension of the newborn, sepsis, respiratory distress syndrome) and severities (oxygenation index  40) of illness were randomly assigned to undergo conventional therapy versus ECMO. The trial was aborted early after an identified survival benefit in the ECMO group.119 The neurodevelopmental outcomes of the survivors have since been followed-up through 7 years of age to date. No significant differences have been identified between the surviving groups of ECMO versus conventional therapy patients in the domains of cognition, neuromotor skills, behavior, vision, and

hearing.120 That study was particularly important because the long-term outcomes of infants who underwent ECMO were compared with a similar group of infants, rather than normative data for age. In a single institution analysis of 95 neonatal and pediatric patients with cardiac failure who underwent VA ECMO for low cardiac output, failure to wean from CPB, or in cardiac arrest (ECPR), 75% of those who survived to discharge experienced normal to mildly disabled outcomes per the Pediatric Overall Performance Category Score and the Pediatric Cerebral Performance Category Scales. Moreover, an additional 6% who had been classified as having moderate to severe disability ultimately improved to have mild or no disability at the 2-year follow-up. Interestingly, cerebral infarction was the only complication that significantly differed between the normal to mildly disabled group (0%) versus moderately to severely disabled group (15%). Of note, the average run time in this cohort of patients was <2 days compared with the average of 6 to 7 days reported by ELSO; it is plausible that the study’s patients had lesser disease severity and may have sustained more complications with longer run times, which consequently may have affected their outcomes.4,121 In contrast, in a smaller cohort of neonatal and pediatric patients with similar underlying cardiac pathologies, 50% had mental disability, defined as a score of <70 on the standardized intelligence tests (Bayley Scales of Infant Development II, the Wechsler Preschool and Primary Scale of Intelligence) at >6month follow-up. The survival rate in this group was 41% versus 66% in the prior study, suggesting that these patients had a greater severity of illness.122 Large multi-institutional studies that use a standardized neurodevelopmental assessment and compare survivors with matched control patients are necessary to more accurately differentiate the long-term effect of ECMO versus the underlying disease process on cognitive and functional status. Acute kidney injury (AKI) and fluid overload during ECMO also have been implicated as independent risk factors for mortality.123126 Per the ELSO registry, over the past 3 decades, the incidence of AKI was 7.8% in neonates and 12.9% in children. These incidences are likely underestimated because AKI is defined as a serum creatinine of 1.5 mg/dL rather than according to the glomerular filtration rate. Furthermore, at least 15.9% of neonates and 23.2% of children have required renal replacement therapy (RRT), including hemofiltration or hemodialysis, to date.127 The risk of AKI appears to increase with duration of ECMO. A retrospective analysis of noncardiac neonatal and pediatric patients included in the ELSO registry demonstrated a sequentially increasing risk of mortality with the need for RRT, presence of AKI, and combination of both, after controlling for severity of illness, mode of ECMO, and patient characteristics, such as comorbidities.124 In regard to the pediatric cardiac ECMO population, a retrospective analysis of a single institution’s experience also demonstrated increased odds of mortality in patients with AKI. Although the study was small, it showed increased incidences of AKI (71.7%) and RRT (58.7%). Notably, AKI was defined according to glomerular filtration rate (<35 mL/min/1.73 m2) rather than serum creatinine. Single ventricle anatomy was not

ARTICLE IN PRESS E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

associated with an increased risk of the development of AKI.125 AKI during ECMO has not been shown to lead to chronic renal failure, although the literature is limited.127 Several groups have shown long-term pulmonary morbidity in neonatal ECMO survivors, in particular obstructive lung disease.128,129 However, the direct effect of ECMO on the development of chronic lung disease cannot be deduced from many of these studies because the lung function of the ECMO patients was compared with healthy subjects rather than conventionally treated control patients matched according to diagnosis and severity of illness. The UK Collaborative Trial did not demonstrate a significant difference in pulmonary outcome at age 7 in matched cohorts of neonates who had been randomly assigned to ECMO versus conventional therapy.119 There is need for the assessment of the long-term pulmonary outcomes in pediatric patients, given a lack of reported literature and differences in the pathophysiology of the underlying diagnostic indications for ECMO compared with neonates. Ethical Considerations ECMO is a high-risk but potentially life-saving therapy, for which patient candidacy is complex and controversial. Futile cases can have serious consequences, such as the moral distress of caregivers and providers during prolonged runs, the terminal discontinuation of ECMO, and the use of extraordinary resources. However, the determination of acceptable versus unacceptable risk is obscure in many cases. Furthermore, gained and shared experience and technologic advancement have led to improved outcomes in situations previously or currently deemed to be contraindications, such as pediatric sepsis and cardiac and respiratory failure in those with severe chromosomal disorders.104,130 ELSO has established guidelines regarding absolute contraindications to ECMO, all of which are deemed to be fatal, irreversible conditions. These include lethal chromosomal anomalies (trisomy 13 and 18), severe neurologic injury (“intracranial hemorrhage with mass effect”), malignancies for which there is no cure, and recipients of allogeneic bone marrow transplantantion with pulmonary disease.6,7 However, despite ELSO’s recommendations, a 2009 survey of neonatal ECMO centers in North America and Europe revealed that 9% of providers would consider ECMO for patients with trisomy 13, 10% for neonates with trisomy 18, 27% for those with grade III or IV intraventricular hemorrhage, and 52% for patients with severe hypoxic ischemic encephalopathy.131 Once ECMO is initiated, the determination of the duration after which survival or a functional outcome is not possible presents another important challenge. The longest ECMO run published occurred in a 26-year-old man who drowned, was cannulated for 117 days (VA/venovenous arterial for 87 days, VV for 30 days), and ultimately survived with mild pulmonary fibrosis and otherwise no morbidity attributable to ECMO.132 There are case series of neonatal and pediatric patients who survived VV and VA ECMO runs lasting up to 48 days with favorable neurologic outcomes.133,134 However, in a single institution case series describing 22 neonatal and pediatric patients who underwent prolonged (>28 d) ECMO for cardiac

11

or respiratory failure, only 14% survived and, of those, 9% experienced severe neurologic morbidity.135 Although research has allowed for the identification of risk factors for morbidity and mortality, it still is not possible to precisely predict who will survive with a favorable outcome and the duration of time needed for such recovery. A report of a 6-year-old with acute myeloid leukemia, a high-risk diagnosis, and acute respiratory failure secondary to bacteremia, who ultimately survived a prolonged 38-day run of VV ECMO without morbidity, highlights the dilemma with which clinicians are faced.136 Conclusions The field of ECMO has advanced considerably over the past 50 years, in large part secondary to the neonatal and, more recently, adult experiences. ELSO has had a substantial effect on the evolution of ECMO via international collaboration and the collection and dissemination of information that have propelled research forward. The machinery, in particular, represents an area of significant achievement with the transformations of the pumps and membrane lungs. Furthermore, there is an improved understanding of the long-term implications of ECMO by way of neonatal survivors. Still, there are important problems to be solved and crucial questions to be answered. Anticoagulation remains one such unsolved problem, given the incidence of hemorrhage and its effect on outcomes. Finally, we have yet to accurately predict which children can be saved by ECMO with minimal to no morbidity. Conflicts of Interest The authors declare no conflicts of interest. References 1 Hill JD, O’Brien TG, Murray JJ, Dontigny L, et al. Prolonged Extracorporeal Oxygenation for Acute Post-Traumatic Respiratory Failure (ShockLung Syndrome): Use of the Bramson Membrane Lung. N Engl J Med 1972;286:629–34. 2 Extracorporeal Life Support Organization. ECLS Registry Report: International Summary. 2019. 3 Thiagarajan RR, Barbaro RP, Rycus PT, Mcmullan DM, Conrad SA, Fortenberry JD, Paden ML, ELSO member centers. Extracorporeal Life Support Organization Registry International Report 2016. ASAIO J. 2017;63:60–7. 4 Barbaro RP, Paden ML, Guner YS, Raman L, et al. Pediatric Extracorporeal Life Support Organization Registry International Report. ASAIO J. 2017;63:456–63. 5 Ryerson L, McMullan DM. Indications and Contraindications for ECLS in Children with Cardiovascular Disease. In: Brogan TV, editor. Extracorporeal Life Support: The ELSO Red Book, 5th Edition, Ann Arbor, MI: Extracorporeal Life Support Organization; 2017. p. 339–46. 6 Coleman RD, Goldman J, Moffett B, Guffy D, et al. Extracorporeal Membrane Oxygenation Mortality in High Risk Populations: An Analysis of the Pediatric Health Information System Database. ASAIO J 2019. https://doi.org/10.1097/MAT.0000000000001002;[Epub ahead of print]. 7 Suttner DM. Indications and Contraindications for ECLS in Neonates with Respiratory Failure. In: Brogan TV, editor. Extracorporeal Life Support: The ELSO Red Book, 5th Edition, Ann Arbor, MI: Extracorporeal Life Support Organization; 2017. p. 239–54.

ARTICLE IN PRESS 12

E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115

8 American Heart Association. https://eccguidelines.heart.org/wp-content/ themes/eccstaging/dompdf-master/pdffiles/part-12-pediatric-advancedlife-support.pdf. Accessed May 8, 2019. 9 Paden ML, Conrad SA, Rycus PT, Thiagarajan RR, ELSO Registry. Extracorporeal Life Support Organization Registry Report 2012. ASAIO J. 2013;59:202–10. 10 Hintz SR, Suttner DM, Sheehan AM, Rhine WD, et al. Decreased Use of Neonatal Extracorporeal Membrane Oxygenation (ECMO): How New Treatment Modalities Have Affected ECMO Utilization. Pediatrics 2000;106:1339–43. 11 Haines NM, Rycus PT, Zwischenberger JB, Bartlett RH, et al. Extracorporeal Life Support Registry Report 2008: Neonatal and Pediatric Cardiac Cases. ASAIO J 2009;55:111–6. 12 Toomasian JM, Vercaemst L, Bottrell S, Horton SB. The Circuit. In: Brogan TV, editor. Extracorporeal Life Support: The ELSO Red Book, 5th Edition, Ann Arbor, MI: Extracorporeal Life Support Organization; 2017. p. 49–80. 13 Lawson DS, Ing R, Chiefetz IM, Walczak R, et al. Hemolytic characteristics of three commercially available centrifugal blood pumps. Pediatr Crit Care Med 2005;6:573–7. 14 Byrnes J, McKamie W, Swearingen C, Prodhan P, et al. Hemolysis During Cardiac Extracorporeal Membrane Oxygenation: A Case-Control Comparison of Roller Pumps and Centrifugal Pumps in a Pediatric Population. ASAIO J 2011;57:456–61. 15 Barrett CS, Jaggers JJ, Cook F, Graham DA, et al. Pediatric ECMO Outcomes: Comparison of Centrifugal Versus Roller Blood Pumps Using Propensity Score Matching. ASAIO J 2013;59:145–51. 16 O’Brien C, Monteagudo J, Schad C, Cheung E, Middlesworth W. Centrifugal pumps and hemolysis in pediatric extracorporeal membrane oxygenation (ECMO) patients: An analysis of Extracorporeal Life Support Organization (ELSO) registry data. J Pediatr Surg 2017;52:975–8. 17 Lequier L, Horton SB, McMullan DM, Bartlett RH. Extracorporeal Membrane Oxygenation Circuitry. Pediatr Crit Care Med 2013;14:S7–12. 18 Lawson DS, Walczak R, Lawson AF, Shearer IR. North American Neonatal Extracorporeal Membrane Oxygenation (ECMO) Devices: 2002 Survey Results. J Extra Corpor Technol. 2004;36:16–21. 19 Lawson DS, Lawson AF, Walczak R, McRobb C. North American Neonatal Extracorporeal Membrane Oxygenation (ECMO) Devices and Team Roles: 2008 Survey Results of the Extracorporeal Life Support Organization (ELSO) Centers. J Extra Corpor Technol. 2008;40:166–74. 20 Lawson S, Ellis C, Butler K, McRobb C, Mejak B. Neonatal Extracorporeal Membrane Oxygenation Devices, Techniques and Team Roles: 2011 Survey Results of the United States’ Extracorporeal Life Support Organization Centers. J Extra Corpor Technol. 2011;43:236–44. 21 Yeager T, Roy S. Evolution of Gas Permeable Membranes for Extracorporeal Membrane Oxygenation. Artif Organs 2017;41:700–9. 22 Millar JE, Fanning JP, McDonald CI, et al. The inflammatory response to extracorporeal membrane oxygenation (ECMO): a review of the pathophysiology. Crit Care Med 2016;20(1):387. 23 Annich GM. Extracorporeal life support: the precarious balance of hemostasis. J Thromb Haemost 2015;13(Suppl 1):S336–42. 24 Ontaneda A, Annich GM. Novel Surfaces in Extracorporeal Membrane Oxygenation Circuits. Front Med 2018;5:32. 25 Sohn N, Marcoux J, Mycyk T, Krahn J, Meng Q. The impact of different biocompatible coated cardiopulmonary bypass circuits on inflammatory response and oxidative stress. Perfusion 2009;24:231–7. 26 Mangoush O, Purkayastha S, Haj-Yahia S, Kinross J, et al. Heparinbonded circuits versus nonheparin-bonded circuits: an evaluation of their effect on clinical outcomes. Eur J Cardiothorac Surg 2007;31:1058–69. 27 Ranucci M, Isgro G, Soro G, Canziani A, et al. Reduced systemic heparin dose with phosphorylcholine coated closed circuit in coronary operations. Int J Artif Organs 2004;27:311–9. 28 Thiasa AS, Andersen VY, Videm V, Mollnes TE. Comparable biocompatibility of Phisio- and Bioline-coated cardiopulmonary bypass circuits indicated by the inflammatory response. Perfusion 2010;25:9–16. 29 Jacobs S, De Somer F, Vandenplas G, Van Belleghem Y, et al. Active or passive bio-coating: does it matter in extracorporeal circulation? Perfusion 2011;26:496–502.

30 Suzuki Y, Daitoku K, Minakawa M, Fukui K, et al. Poly-2-methoxyethylacrylate-coated bypass circuits reduce activation of coagulation and inflammatory response in congenital cardiac surgery. J Artif Organs 2008;11:111–6. 31 Itoh H, Ichiba S, Ujike Y, Douguchi T, et al. A prospective randomized trial comparing the clinical effectiveness of heparin-coated circuitd and PMEA-coated circuits in pediatric cardiopulmonary bypass. Perfusion 2016;31:247–54. 32 Wang-Zwische Double Lumen Cannula – Toward a Percutaneous and Ambulatory Paracorporeal Artificial Lung. ASAIO J 2008;54:606–11. 33 Jarboe MD, Gadepalli SK, Church JT, Arnold MA, et al. Avalon catheters in pediatric patients requiring ECMO: Placement and migration problems. J Pediatr Surg 2018;53:159–62. 34 Salazar PA, Blitzer D, Dolejs SC, Parent JJ, et al. Echocardiographic Guidance During Neonatal and Pediatric Jugular Cannulation for ECMO. J Surg Res 2018;232:517–23. 35 Garcia AV, Jeyaraju M, Ladd MR, Jelin EB, Bembea MM, Alaish S, Rhee D. Survey of the American Pediatric Surgical Association on cannulation practices in pediatric ECMO. J Pediatr Surg 2018;53:1843–8. 36 Johnson K, Jarboe MD, Mychaliska GB, Barbaro RP, et al. Is there a best approach for extracorporeal life support cannulation: a review of the extracorporeal life support organization. J Pediatr Surg 2018;53:1301–4. 37 Teele SA, Salvin JW, Barrett CS, Rycus PT, et al. The association of carotid artery cannulation and neurologic injury in pediatric patients supported with venoarterial extracorporeal membrane oxygenation. Pediatr Crit Care Med 2014;15:355–61. 38 DiGennaro JL, Chan T, Farris RWD, Weiss NS, McMullan DM. Increased Stroke Risk in Children and Young Adults on Extracorporeal Life Support with Carotid Cannulation. ASAIO J 2019. 39 Gander JW, Fisher JC, Reichstein AR, Gross ER. Limb ischemia after common femoral artery cannulation for venoarterial extracorporeal membrane oxygenation: an unresolved problem. J Pediatr Surg 2010;45:2136–40. 40 Andraska EA, Jackson T, Chen H, Gallagher KA, et al. Natural History of Iatrogenic Pediatric Femoral Artery Injury. Ann Vasc Surg 2017;42: 205–13. 41 Pettignano R, Fortenberry JD, Heard ML, Labuz MD. Primary use of the venovenous approach for extracorporeal membrane oxygenation in pediatric acute respiratory failure. Pediatr Crit Care Med 2003;4:291–8. 42 Skinner SC, Iocono JA, Ballard HO, Turner MD, et al. Improved survival in venovenous versus venoarterial extracorporeal membrane oxygenation for pediatric noncardiac sepsis patients: a study of the Extracorporeal Life Support Organization registry. J Pediatr Surg 2012;47:63–7. 43 Kruger K, Schmutz A, Zieger B, Kalbhenn J. Venovenous Extracorporeal Membrane Oxygenation With Prophylactic Subcutaneous Anticoagulation Only: An Observation Study in More Than 60 Patients. Artif Organs 2017;41:186–92. 44 Herbert DG, Buscher H, Nair P. Prolonged Venovenous extracorporeal membrane oxygenation without anticoagulation: A case of Goodpasture syndrome-related pulmonary hemorrhage. Crit Care Resusc 2014;16:69–72. 45 Muellenback RM, Kredel M, Kunze E, Kranke P, et al. Prolonged heparin-free extracorporeal membrane oxygenation in multiple injured acute respiratory distress syndrome patients with traumatic brain injury. J Trauma Acute Care Surg 2012;72:1444–7. 46 Abdulhai S, Glenn IC, McNinch NL, Ponsky TA, et al. Current Practices in the Management of Congenital Diaphragmatic Hernia Patients Requiring Extracorporeal Membrane Oxygenation: Results of an International Survey of Pediatric Surgeons. J Laparoendosc Adv Surg Tech A 2018;28:606–9. 47 Guner YS, Harting MT, Fairbairn K, Delaplain PT, et al. Outcomes of infants with congenital diaphragmatic hernia treated with venovenous versus venoarterial extracorporeal membrane oxygenation: A propensity score approach. J Pediatr Surg 2018;53:2092–9. 48 Conrad SA, Grier LR, Scott K, Green R, Jordan M. Percutaneous Cannulation for Extracorporeal Membrane Oxygenation by Intensivists: A Retrospective Single-Institution Case Series. Crit Care Med 2015;43:1010–5. 49 Kotani Y, Chetan D, Rodrigues W, Sivarajan VB, et al. Left atrial decompression during venoarterial extracorporeal membrane oxygenation for left ventricular failure in children: current strategy and clinical outcomes. Artif Organs 2013;37:29–36.

ARTICLE IN PRESS E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115 50 Eastaugh LJ, Thiagarajan RR, Darst JR, McElhinney DB, et al. Percutaneous Left Atrial Decompression in Patients Supported With Extracorporeal Membrane Oxygenation for Cardiac Disease. Pediatr Crit Care Med 2015;16:59–65. 51 Zampi JD, Alghanem F, Yu S, Callahan R, et al. Relationship Between Time to Left Atrial Decompression and Outcomes in Patients Receiving Venoarterial Extracorporeal Membrane Oxygenation Support: A Multicenter Pediatric Interventional Cardiology Early-Career Society Study. Pediatr Crit Care Med 2019. https://doi.org/10.1097/PCC.0000000000001936; [Epub ahead of print]. 52 Patel SM, Lipinski J, Al-Kindi SG, Patel T, et al. Simultaneous Venoarterial Extracorporeal Membrane Oxygenation and Percutaneous Left Ventricular Decompression Therapy with Impella is Associated with Improved Outcomes in Refractory Cardiogenic Shock. ASAIO J 2019;65:21–8. 53 Parekh D, Jeewa A, Tume SC, Dreyer WJ, et al. Percutaneous Mechanical Circulatory Support Using Impella Devices for Decompensated Cardiogenic Shock: A Pediatric Heart Center Experience. ASAIO J 2018;64:98–104. 54 ELSO Guidelines for Cardiopulmonary Extracorporeal Life Support. Extracorporeal Life Support Organization, Version 1.4 August 2017. Ann Arbor, MI, USA. www.elso.org. 55 Marhong JD, Telesnicki T, Munshi L, Del Sorbo L, et al. Mechanical Ventilation during Extracorporeal Membrane Oxygenation. Annals ATS 2014;11:956–61. 56 The Acute Respiratory Distress Syndrome Network, Brower RG, Matthay MA, Morris A. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000;342:1301–8. 57 Pham T, Combes A, Roze H, Chevret S, et al. Extracorporeal Membrane Oxygenation for Pandemic Influenza A(H1N1)-induced Acute Respiratory Distress Syndrome: A Cohort Study and Propensity-matched Analysis. Am J Respir Crit Care Med 2013;187:276–85. 58 Schmidt M, Stewart C, Bailey M, Nieszkowska A, et al. Mechanical Ventilation Management During Extracorporeal Membrane Oxygenation for Acute Respiratory Distress Syndrome: A Retrospective International Multicenter Study. Crit Care Med 2015;43:654–64. 59 Keszler M, Rychman FC, McDonald JV Jr, Sweet LD, Moront MG, Boegli MJ, et al. A prospective, multicenter, randomized study of high versus low positive end-expiratory pressure during extracorporeal membrane oxygenation. J Pediatr 1992;120:107–13. 60 Alapati D, Aghai ZH, Hossain MJ, Dimberger DR, et al. Lung Rest during Extracorporeal Membrane Oxygenation for Neonatal Respiratory Failure-Practice Variations and Outcomes. Pediatr Crit Care Med 2017;18:667–74. 61 Millar JE, Fanning JP, McDonald CI, McCauley DF, Fraser JF. The inflammatory response to extracorporeal membrane oxygenation (ECMO): a review of the pathophysiology. Crit Care Med 2016;20:387. 62 Preston TJ, Olshove VF Jr, Ayad O, Nicol KK, et al. Novoseven Use in a Non-Cardiac Pediatric ECMO Patient With Uncontrolled Bleeding. J Extra Corpor Technol 2008;40:123–6. 63 Dominguez TE, Mitchell M, Friess SH, Huh JW. Use of recombinant factor VIIa for refractory hemorrhage during extracorporeal membrane oxygenation. Pediatr Crit Care Med 2005;6:348–51. 64 Long MT, Wagner D, Maslach-Hubbard A, Pasko DA, et al. Safety and efficacy of recombinant activated factor VII for refractory hemorrhage in pediatric patients on extracorporeal membrane oxygenation: a single center review. Perfusion 2014;29:163–70. 65 Bembea MM, Annich G, Rycus P, Oldenburg G, Berkowitz I, Pronovost P. Variability in anticoagulation management of patients on extracorporeal membrane oxygenation: an international survey. Pediatr Crit Care Med 2013;14:e77–84. 66 Manco-Johnson MJ. Neonatal antithrombin III deficiency. The American Journal of Medicine 1989;87:S49–52. 67 Arnold P, Jackson S, Wallis J, Smith J. Coagulation factor activity during neonatal extra-corporeal membrane oxygenation. Intensive Care Med 2001;27:1395–400.

13

68 Niebler RA, Christensen M, Berens R, Wellner H, et al. Antithrombin Replacement During Extracorporeal Membrane Oxygenation. Artif Organs 2011;35:1024–8. 69 Stansfield BK, Wise L, Benson Ham P III, Patel P, et al. Outcomes following routine antithrombin III replacement during neonatal extracorporeal membrane oxygenation. J Pediatr Surg 2017. 70 Perry R, Stein J, Young G, Ramanathan R, et al. Antithrombin III administration in neonates with congenital diaphragmatic hernia during the first three days of extracorporeal membrane oxygenation. J Pediatr Surg 2013;48:1837–42. 71 McLean J. The thromboplastic action of cephalin. Am J Physiol 1916;41:250–7. 72 Hirsh J, Raschke R. Heparin and Low-Molecular-Weight Heparin. Chest 2004;126:S188–203. 73 DiNisio M, Middeldorp S, Buller HR. Direct Thrombin Inhibitors. N Engl J Med 2005;353:1028–40. 74 Young G, Yonekawa KE, Nakagawa PA, Blain RC, Lovejoy AE, Nungent DJ. Recombinant activated factor VII effectively reverses the anticoagulant effects of heparin, enoxaparin, fondaparinoux, argatroban, and Bivalirudin ex vivo as measured using thromboelastography. Blood Coagul Fibrinolysis 2007;18:547–53. 75 Vavra KA, Lutz MF, Smythe MA. Recombinant Factor VIIa to Manage Major Bleeding from Newer Parenteral Anticoagulants. Ann Pharmacother 2010;44:718–26. 76 Buck ML. Bivalirudin as an Alternative to Heparin for Anticoagulation in Infants and Children. J Pediatr Pharmacol Ther 2015;20:408–17. 77 Sanfilippo F, Asmussen S, Maybauer DM, et al. Bivalirudin for Alternative Anticoagulation in Extracorporeal Membrane Oxygenation: A Systematic Review. J Intensive Care Med 2017;32:312–9. 78 Zaleski K, DiNardo JA, Nasr VG. Bivalirudin for Pediatric Procedural Anticoagulation: A Narrative Review. Anesth Analg 2019;128:43–55. 79 Ranucci M, Ballotta A, Kandil H, Isgro G, et al. Bivalirudin-based versus conventional heparin anticoagulation for postcardiotomy extracorporeal membrane oxygenation. Crit Care 2011;15:R275. 80 Nagle EL, Dager WE, Duby JJ, Roberts AJ, et al. Bivalirudin in Pediatric Patients Maintained on Extracorporeal Life Support. Pediatr Crit Care Med 2013;14:e182–8. 81 Barton R, Ignajatovic V, Monagle P. Anticoagulation during ECMO in neonatal and paediatric patients. Thrombosis Research 2019;173:172–7. 82 Maul TM, Wolff EL, Kuch BA, Rosendorff A, et al. Activated partial thromboplastin time is a better trending tool in pediatric extracorporeal membrane oxygenation. Pediatr Crit Care Med 2012;13:e363–71. 83 Bembea MM, Schwartz JM, Shah NM, Colantuoni E, et al. Anticoagulation monitoring during pediatric extracorporeal membrane oxygenation. ASAIO J 2013;59:63–8. 84 Liveris A, Bello RA, Friedmann P, Duffy MA. Anti-Factor Xa Assay Is a Superior Correlate of Heparin Dose Than Activated Partial Thromboplastin Time or Activated Clotting Time in Pediatric Extracorporeal Membrane Oxygenation. Pediatr Crit Care Med 2014;15:e72–9. 85 Moynihan K, Johnson K, Straney L, Stocker C, et al. Coagulation monitoring correlation with heparin dose in pediatric extracorporeal life support. Perfusion 2017;32:675–85. 86 Northrop MS, Sidonio RF, Phillips SE, Smith AH, et al. The Use of an Extracorporeal Membrane Oxygenation Anticoagulation Laboratory Protocol Is Associated with Decreased Blood Product Use, Decreased Hemorrhagic Complications, and Increased Circuit Life. Pediatr Crit Care Med 2015;16:66–74. 87 Shekar K, Fraser JF, Smith MT, Roberts JA. Pharmacokinetic changes in patients receiving extracorporeal membrane oxygenation. J Crit Care 2012;27:741e.9–741e.18. 88 Willdschut ED, Ahsman MJ, Allegaert K, Mathot RAA, Tibboel D. Determinants of drug absorption in different ECMO circuits. Intensive Care Med 2010;36:2109–16. 89 Bhatt-Mehta V, Annich G. Sedative clearance during extracorporeal membrane oxygenation. Perfusion 2005;20:309–15. 90 Nasr VG, Meserve J, Pereira LM, Faraoni D, et al. Sedative and Analgesic Drug Sequestration After a Single Bolus Injection in an Ex Vivo

ARTICLE IN PRESS 14

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108 109

E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115 Extracorporeal Membrane Oxygenation Infant Circuit. ASAIO J 2019;65:187–91. DeBerry BB, Lynch JE, Chernin JM, Zwischenberger JB, Chung DH. A survey for pain and sedation medications in pediatric patients during extracorporeal membrane oxygenation. Perfusion 2005;20:139–43. Anton-Martin P, Modem V, Taylor D, Potter D, Darnell-Bowens C. A retrospective study of sedation and analgesic requirements of pediatric patients on extracorporeal membrane oxygenation (ECMO) from a single-center experience. Perfusion 2017;32:183–91. Schneider JB, Sweberg T, Asaro LA, et al. Sedation management in children supported on extracorporeal membrane oxygenation for acute respiratory failure. Crit Care Med 2017;45:e1001–10. Zabrocki LA, Brogan TV, Statler KD, Poss WB, Rollins MD, Bratton SL. Extracorporeal membrane oxygenation for pediatric respiratory failure: Survival and predictors of mortality. Crit Care Med 2011;39:364–70. Nasr VG, Faraoni D, DiNardo JA, Thiagarajan RR. Association of Hospital Structure and Complications With Mortality After Pediatric Extracorporeal Membrane Oxygenation. Ped Crit Care Med 2016;17:684–91. Karamlou T, Vafaeezadeh M, Parrish AM, Cohen GA, Welke KF, Permut L, McMullan DM. Increased extracorporeal membrane oxygenation center case volume is associated with improved extracorporeal membrane oxygenation survival among pediatric patients. J Thorac Cardiovasc Surg 2013;145:470–5. Freeman CL, Bennett TD, Casper TC, Larsen GY, Hubbard A, Wilkes J, Bratton SL. Pediatric and neonatal extracorporeal membrane oxygenation: does center volume impact mortality? Crit Care Med 2014;42:512–9. Barbaro RP, Folafoluwa OO, Kidwell KM, Paden ML, Bartlett RH, Davis MM, Annich GM. Association of Hospital-Level Volume of Extracorporeal Membrane Oxygenation Cases and Mortality: Analysis of the Extracorporeal Life Support Organization Registry. Am J Resp Crit Care Med 2015;191:894–901. Gonzalez DO, Sebastiao YV, Cooper JN, Minneci PC, Deans KJ. Pediatric Extracorporeal Membrane Oxygenation Mortality Is Related to Extracorporeal Oxygenation Volume in US Hospitals. J Pediatr Surg 2019;236:159–65. Barrett CS, Chan TT, Wilkes J, et al. Association of Pediatric Cardiac Surgical Volume and Mortality After Cardiac ECMO. ASAIO J 2017;63:802–9. Di Nardo M, Lonero M, Pasotti E, Cancani F, et al. The first five years of neonatal and pediatric transports on extracorporeal membrane oxygenation in the center and south of Italy: The pediatric branch of the Italian “Rete Respira” network. Perfusion 2018;33:24–30. Coleman R, Moffett B, Loftis L, Shekerdemian L. Mortality in Pediatric ECMO: An Analysis of the Pediatric Health System Information Database. Pediatr Crit Care Med 2013;41:A8. Ruth A, McCracken CE, Fortenberry JD, Hebbar KB. Extracorporeal therapies in pediatric severe sepsis: findings from the pediatric healthcare information system. Crit Care 2015;19:397. Sole A, Jordan I, Bobillo S, Moreno J, et al. Venoarterial extracorporeal membrane oxygenation support for neonatal and pediatric refractory septic shock: more than 15 years of learning. Eur J Pediatr 2018;177:1191–200. MacLaren G, Butt W, Best D, Donath S. Central extracorporeal membrane oxygenation for refractory pediatric septic shock. Pediatr Crit Care Med 2011;12:133–6. Barbaro RP, Boonstra PS, Paden ML, Roberts LA, et al. Development and Validation of the Pediatric Risk Estimate Score for Children Using Extracorporeal Respiratory Support (Ped-RESCUERS). Intensive Care Med 2016;42:879–88. Bailly DK, Reeder RW, Zabrocki LA, Hubbard AM, et al. Development and Validation of a Score to Predict Mortality in Children Undergoing ECMO for Respiratory Failure: Pediatric Pulmonary Rescue with Extracorporeal Membrane Oxygenation Prediction (P-PREP) Score. Crit Care Med 2017;45:e58–66. Hardart GE, Fackler JC. Risk factors of intracranial hemorrhage during neonatal extracorporeal membrane oxygenation. J Pediatr 1999;134:156–9. Cengiz P, Seidel K, Rycus PT, Brogan TV, Roberts JS. Central nervous system complications during pediatric extracorporeal life support: Incidence and risk factors. Crit Care Med 2005;33:2817–24.

110 Polito A, Barrett CS, Wypij D, Rycus PT, Netto R, Cogo PE, Thiagarajan RR. Neurologic complications in neonates supported with extracorporeal membrane oxygenation. An analysis of ELSO registry data. Intensive Care Med. 2013;39:1594–601. 111 Bulas D, Glass P. Neonatal ECMO: neuroimaging and neurodevelopmental outcome. Semin Perinatol 2005;29:58–65. 112 LaRovere KL, Vonberg FW, Prabhu SP, Kapur K, et al. Patterns of Head Computed Tomography Abnormalities During Pediatric Extracorporeal Membrane Oxygenation and Association With Outcomes. Pediatr Neurol 2017;73:64–70. 113 Wong JK, Smith TN, Pitcher HT, Hirose H, et al. Cerebral and Lower Limb Near-Infrared Spectroscopy in Adults on Extracorporeal Membrane Oxygenation. Artif Organs 2012;36:659–67. 114 Pozzebon S, Blandino OA, Franchi F, Cristallini S, et al. Cerebral NearInfrared Spectroscopy in Adult Patients Undergoing Veno-Arterial Extracorporeal Membrane Oxygenation. Neurocrit Care 2018;29:94–104. 115 Tian F, Jenks C, Potter D. Regional Cerebral Abnormalities Measured by Frequency-Domain Near-Infrared Spectroscopy in Pediatric Patients During Extracorporeal Membrane Oxygenation. ASAIO 2017;63:e52–9. 116 Glass P, Wagner AE, Papero RH, Rajasingham SR, et al. Neurodevelopmental status at age five years of neonates treated with extracorporeal membrane oxygenation. J Pediatr 1995;127:447–57. 117 Ijsselstijn H, van Heijst AFJ. Long-term outcome of children treated with neonatal extracorporeal membrane oxygenation: Increasing problems with increasing age. Semin Perinatol 2014;38:114–21. 118 Madderom MJ, Schiller RM, Gischler SJ, van Heijst AFJ, et al. Growing Up After Critical Illness: Verbal, Visual-Spatial, and Working Memory Problems in Neonatal Extracorporeal Membrane Oxygenation Survivors. Pediatr Crit Care Med 2016;44:1182–90. 119 UK Collaborative ECMO Trial Group. UK collaborative randomised trial of neonatal extracorporeal membrane oxygenation. Lancet 1996;348:75–82. 120 McNally H, Bennett CC, Elbourne D, et al. UK Collaborative ECMO Trial Group: United Kingdom collaborative randomized trial of neonatal extracorporeal membrane oxygenation: Follow-up to age 7 years. Pediatrics 2006;117:e845–54. 121 Chrysostomou C, Maul T, Callahan PM, Nguyen K, et al. Neurodevelopmental outcomes after pediatric cardiac ECMO support. Front Prediatr 2013;1:1–6. 122 Lequier L, Joffe AR, Robertson CMT, Dinu IA, et al. Two-year survival, mental, and motor outcomes after cardiac extracorporeal life support at less than five years of age. J Thorac Cardiovasc Surg 2008;136:976–83. 123 Swaniker F, Kolla S, Moler F, Custer J, et al. Extracorporeal life support outcome for 128 pediatric patients with respiratory failure. J Pediatr Surg 2000;35:197–202. 124 Askenazi DJ, Ambalavanan N, Hamilton K, Cutter G, et al. Acute kidney injury and renal replacement therapy independently predict mortality in neonatal and pediatric noncardiac patients on extracorporeal membrane oxygenation. Pediatr Crit Care Med 2011;12:e1–6. 125 Smith AH, Hardison DC, Worden CR, Fleming GM, Taylor MB. Acute Renal Failure During Extracorporeal Support in the Pediatric Cardiac Patient. ASAIO J 2009;44:412–6. 126 Fleming GM, Sahay R, Zappitelli M, King E, et al. The Incidence of Acute Kidney Injury and its Effect on Neonatal and Pediatric ECMO Outcomes: A multicenter report from the KIDMO Study Group. Pediatr Crit Care Med 2016;17:1157–69. 127 Paden ML, Warshaw BL, Heard ML, Fortenberry JD. Recovery of renal function and survival after continuous renal replacement therapy during extracorporeal membrane oxygenation. Pediatr Crit Care Med 2011;12:153–8. 128 Boykin AR, Quivers ES, Wagenhoffer KL, Sable CA, et al. Cardiopulmonary outcome of neonatal extracorporeal membrane oxygenation at ages 10-15 years. Crit Care Med 2003;31:2380–4. 129 Hamutcu R, Nield TA, Garg M, Keens TG, Platzker ACG. Long-Term Pulmonary Sequelae in Children Who Were Treated With Extracorporeal Membrane Oxygenation for Neonatal Respiratory Failure. Pediatrics 2004;114:1292–6. 130 Robinson SG. Indications and Contraindications for ECLS in Children with Respiratory Failure. Brogan TV, editor. Indications and Contraindications

ARTICLE IN PRESS E. Valencia and V.G. Nasr / Journal of Cardiothoracic and Vascular Anesthesia 00 (2019) 115 for ECLS in Children with Respiratory Failure. Extracorporeal Life Support: The ELSO Red Book 2017:239–54. 131 Chapman RL, Peterec SM, Bizzarro MJ, et al. Patient Selection for neonatal extracorporeal membrane oxygenation: beyond severity of illness. J Perinatol 2009;29:606–11. 132 Wang CH, Chou CC, Ko WJ, Lee YC. Rescue a drowning patient by prolonged extracorporeal membrane oxygenation support for 117 days. Am J Emerg Med 2010;28:750.e5–7. 133 Lee WA, Kolla S, Schreiner RJ Jr, Hirschl RB. Prolonged extracorporeal life support (ECMO) for varicella pneumonia. Crit Care Med 1997;25:977–82.

15

134 Vida VL, Rubino M, Stellin G. Prolonged ECMO support for virusinduced cardiorespiratory failure early after cardiac surgery. Pediatr Cardiol 2006;27:122–3. 135 Gupta P, McDonald R, Chipman CW, Stroud M, et al. 20-Year Experience of Prolonged Membrane Oxygenation in Critically Ill Children with Cardiac or Pulmonary Failure. Ann Thorac Surg 2012;93:1584–90. 136 Ong J, Ngiam N, Aye WMM, Maclaren G. Prolonged venovenous extracorporeal membrane oxygenation in a child with leukemia and persistent bacteremia. Pediatr Crit Care Med 2011;12:e395–7.