inflated to a static volume, and then oscillated around the mean airway pressure.

Neonatal and Pediatric ECMO

For infants who cannot be successfully ventilated using these modes of ventilation, extracorporeal life

support (ECLS) can become an option. ECLS is indicated in cases of acute cardiopulmonary failure with

expected organ recovery which has been unresponsive to maximal therapy.51 The use of ECLS in

neonates is based on the idea that with maximal support native function improves, through natural

development or improvement, by application of medical and/or surgical treatments, or transplantation.

Three clinical measurement systems have been identified to identify these patients all of which are

associated with high mortality risk. However, the decision to initiate this therapy is based on clinical

judgment and the patient’s response to maximal medical therapy.

Dr. Robert Bartlett and his group had been studying extracorporeal support in laboratory animals for

10 years before the first neonatal extracorporeal membrane oxygenation (ECMO) patient was supported

in 1975; this clinical application signified a radical shift in the treatment paradigm for neonatal

respiratory failure.52 The neonate, Esperanza, was suffering from respiratory failure secondary to

meconium aspiration that was recalcitrant to conventional ventilator support. She was successfully

supported and now is 37 years old.53

The currently accepted treatment of neonatal cardiopulmonary failure includes low-volume protective

ventilation,54 inhaled nitric oxide (iNO),55,56 surfactant therapy,57,58 and HFOV.59 If the cardiac or

pulmonary failure is refractory to maximal medical therapy then ECMO should be considered.60,61 The

standard application of ECMO is primarily limited to neonates ≥34 weeks EGA and ≥2 kg. Clinical

experience and laboratory work suggest extracorporeal support may be effective at lower gestational

ages by using “Premie ECMO” or the “Artificial Placenta (AP) (Table 99-1).”

Indications

ECLS is indicated for acute cardiopulmonary failure with high mortality/morbidity unresponsive to

maximal medical treatment and with expected organ recovery. The premise is that patients can be

supported with ECLS while native function improves either by allowing natural development or

improvement (lung maturation and surfactant development), application of medical or surgical

therapies, or by transplantation. Identification of the patients that can benefit from ECLS can be

challenging.

Three clinical measurement systems have been developed and tested to assist in identifying patients

that will benefit from ECLS support.

1. Oxygenation Index (OI) = (MAP × FiO2 × 100) / PaO2 Where MAP = mean airway pressure. This

index has been evaluated62,63 and shown that an OI >40 in three to five postductal gases was

predictive of a mortality risk ≥80%.64

2. Postductal Alveolar-Arterial Oxygen Gradient [(A-a)DO2] An (A-a)DO2 of 610 Torr or greater despite 8

hours of maximal medical therapy predicted a mortality of 79%.63

3. Ventilation Index = (Respiratory Rate × PaCO2 × Peak Inspiratory Pressure)/1,000

Rivera et al. found that a ventilation index >40 and OI >40 were associated with a 77% mortality

risk. They also found that the combination of peak inspiratory pressure ≥40 cm H2O and an (A-a)DO2

>580 mm Hg was associated with a mortality of 81%.65

These clinical measurement systems are useful to quantitate the degree of cardiopulmonary

derangement, and subsequently categorize patients into candidates for ECLS therapy or continued

maximal medical therapy. However, the decision to initiate ECLS therapy is often a clinical decision

based upon clinical judgment and the patient’s individual response to maximal medical therapy. Patients

are commonly started on ECLS therapy when they have failed maximal medical support, significant

barotrauma is imminent, and are felt to have good potential for complete organ recovery.

Classic Contraindications and Possible Treatment Expansion

The classic contraindications to ECLS therapy are listed below. As ECLS treatment evolves and

technology advances, many of the classic contraindications to ECLS are being challenged.

1. EGA less than 34 weeks: The higher incidence of intracranial bleeding in premature infants has

historically precluded the use of ECLS in neonates less than 34 weeks EGA.66,67 However, recent data

indicate that ECLS can potentially be used in infants as low as 29 weeks EGA with acceptable survival

2854

and intracranial hemorrhage (ICH) rates. Ideally, development of nonthrombogenic coating of circuit

components would obviate the need for systemic heparinization and decrease the risk of using ECLS

in premature infants.68–71

2. ICH greater than grade II: Neonates with ICH of higher grades are at increased risk of extension of

their hemorrhage with systemic heparinization. This remains true today, but the development of

technologies that obviate the need for heparinization may allow the use of ECLS in neonates with preexisting ICH in the future.69,71–75 In addition, our experience has suggested that ECMO can be applied

when expected mortality is higher in neonates with grade II ICH. In that setting, lower levels of

anticoagulation are cautiously applied.

3. Mechanical ventilation for longer than 7 to 10 days: Classically, mechanical ventilation has been

associated with higher incidence of bronchopulmonary dysplasia and irreversible fibroproliferative

lung disease. The duration of pre-ECMO mechanical ventilation is being challenged; data from the

Extracorporeal Life Support Organization (ELSO) registry demonstrate survival of 50% to 60% after

pre-ECLS mechanical ventilation of up to 14 days.76

4. Cardiac arrest which requires cardiopulmonary resuscitation (CPR): Many centers now consider patients

who suffer pre-ECLS cardiac arrest candidates for support. Survival rates up to 60% have been

demonstrated in neonates who suffer cardiac arrest prior to or during cannulation.77,78 Predictably,

good outcomes are associated with effective CPR during the resuscitation.

5. Conditions incompatible with meaningful life after therapy – profound neurologic impairment, congenital

anomalies, or other conditions: With improvement in medical and surgical care, conditions once thought

to be nonsurvivable require constant reassessment.

Outcomes

Today, ECMO is a part of routine management in the neonatal ICU. Overall survival is 85%, with 98%

survival for meconium aspiration and 55% survival for diaphragmatic hernia.52 Overall survival after

ECLS for neonatal respiratory failure has recently declined. There are likely a few reasons for this

decrease in survival. Since its peak in 1992 of 1,500 cases, ECLS has been used with less frequency for

critically ill neonates. The fewer number of cases is due to an improvement in other modalities of

support such as iNO, HFOV, and surfactant therapy. To some extent, these improvements have been

realized as a result of lessons learned from early ECMO experience.52,79–81 As a result, the neonatal

patients receiving ECLS as the initial form of therapy for cardiopulmonary failure are declining, and the

patients that ultimately require ECLS are probably more ill and further along the timeline of their

illness.

Table 99-1 ECLS Modality by Age Group

Modalities

ECLS aims to provide perfusion of warmed, arterialized blood into the patient.82,83 Traditionally,

venovenous (VV) support has been used for respiratory failure while venoarterial (VA) support has been

utilized in cases of cardiac or combined cardiopulmonary failure. VV-ECLS can be performed through

one-site or two-site cannulation. Traditionally, blood was drained through a cannula placed in the right

internal jugular (IJ) vein and returned through the femoral vein, although this approach proved

problematic in newborns with small femoral veins. More recently one-site or venovenous double-lumen

(VVDL) cannulation has emerged as the preferred cannulation technique. In this mode, a single cannula

is placed in the right IJ vein. Deoxygenated blood is drained from one port, pumped through the ECLS

circuit where gas exchange occurs, and returned through a separate port on the same cannula into the

right atrium. VA-ECLS provides complete cardiopulmonary support. Typically a drainage cannula is

placed in the right IJ. Deoxygenated blood is pumped through the ECMO circuit and returned through a

cannula in the right carotid artery. Historically, VA-ECLS has been used more than VV-ECLS in the

support of neonates. However, data from the ELSO registry demonstrate that the use of VV-ECLS is

increasing for neonatal cardiopulmonary failure.83

ECMO II

2855

The first generation of ECMO support devices (known as ECMO I) was used from 1975 to 2008.

However, problems with the oxygenator membrane, rotational pump, and cannulae limited expansion of

the technology. Recently there has been substantial technologic improvement in ECMO system

components and circuitry, defined as ECMO II.84

Modern hollow fiber oxygenators have several advantages over the original silicon membrane

oxygenators.85–87 The hollow fiber design provides a much lower resistance across the oxygenator

membrane, allowing for the use of centrifugal pumps in the ECMO circuit. The centrifugal pumps used

now consist of a rotating impeller that spins on a small bearing or is magnetically suspended.88 Some of

the newest pumps available provide pulsatile flow.89 Additionally, the advent of the double-lumen

Avalon cannula (Avalon Elite, Avalon Laboratories, Rancho Dominguez, CA) allows the percutaneous

placement of a single cannula for VV-ECMO support that both drains and reinfuses blood to allow for

effective circuit flow.90,91 Further, there has been significant research effort put forth to improve the

biocompatibility of circuit surface–blood interfaces.72,73 Finally, the simplification of operating the

ECMO II system makes it such that a trained ICU nurse can manage the circuit. This decreases

manpower utilization, as a single ECMO technician may now oversee several patients at once.92 These

advancements provide translational development in the application and expansion of ECMO support for

neonatal cardiopulmonary failure.

ECMO in Premature Neonates

Previous studies have shown poor survival with the use of ECMO in premature neonates. Bartlett’s early

report of ECLS in 45 moribund neonates, revealed a 55% survival rate for neonates overall. Nine of

these patients were less than 35 weeks EGA – two survived (22%). All six who died had ICH.93 A later

report showed similarly worrisome survival for premature neonates less than 35 weeks EGA – 33%

(3/15). One of these survived with an ICH; the other 12 patients with ICH died.63 A study examining

ICH in patients on ECLS showed that all eight of the patients less than 35 weeks EGA had ICH and that

six of these died immediately after termination of ECLS while two died within 1 year of age.66 Given

these poor survival rates (22% to 33%) and the extremely high rate of ICH (75% to 100%), Cilley

recommended that the use of ECLS be contraindicated in neonates of less than 35 weeks EGA. However,

examination of the ELSO database between 1988 and 1991, found that survival had improved to 63%

and ICH rates decreased to 37% for premature neonates born at 34 weeks EGA or younger.66,67

We recently performed a comprehensive analysis of the ELSO database from 1976 to 2008. This

analysis showed no statistically significant difference in the risk of IVH in infants 30 to 33 weeks EGA

(21%) when compared to those 34 weeks EGA (17%), p = 0.195.94 The overall survival rate of 54%

and ICH rate of 18% for the entire cohort of neonates on ECLS at 34 weeks EGA or younger suggest

that the incidence of both mortality and ICH have decreased over time, with a more dramatic reduction

observed in ICH. Although the survival for the 29- to 33-week EGA group was significantly decreased

(48%) when compared to those patients with EGA 34 weeks (58%, p = 0.011), one could argue that

these rates are clinically acceptable. Patients born at younger EGA are expected to have a more difficult

hospital course. These data suggest that ECLS in the modern era should be explored at EGA as low as 29

weeks with reasonable survival and acceptable rates of ICH.

Rationale for an Artificial Placenta – Applying Fetal Physiologic Principles to the Treatment of

Extreme Prematurity. ECMO is increasingly effective for late preterm to term infants (34 to 40 weeks

EGA), and it might be beneficial for some infants who are 29 to 34 weeks EGA. However, a different

paradigm is required for ECLS in extremely low gestational age newborns (ELGANs, 22 to 28 weeks

EGA). ELGANs are at the greatest risk for death and poor long-term outcomes.95–97 In particular,

respiratory failure in premature neonates contributes to significant mortality and long-term disability.

Conventional mechanical ventilation is often inadequate to provide gas exchange and can cause trauma

to the under-developed lungs. Even more advanced strategies developed to minimize the trauma of

conventional ventilation, including surfactant replacement therapy, nitric oxide inhalation, and HFOV,

may at times be insufficient to support these most fragile infants.98,99 In theory, the most effective

solution is to create an AP that maintains fetal circulation and obviates the need for mechanical

ventilation and resultant devastating barotrauma by completely bypassing the lungs.

Definition of the Artificial Placenta

1. Maintenance of fetal circulation and the intrauterine environment

2. No mechanical ventilation

2856

3. Simulated fetal breathing with fluid-filled lungs

4. A novel form of a pump-driven VV or arteriovenous (AV) ECLS/ECMO with outflow via the right

jugular vein or umbilical artery, respectively, and inflow via the umbilical vein

A VV-ECLS AP offers several potential advantages over an AV-ECLS system. First, it eliminates the use

of arterial vessels that are prone to spasm. Placing a cannula in the right atrium through the jugular

vein allows for a passive drainage process that reduces potential negative pressure and cavitation in the

circuit. Additionally, VV-ECLS uses the subject’s own venous system as a blood reservoir, instead of

requiring an external blood reservoir. Another potential advantage relates to the drainage cannula itself.

Although it is feasible to advance a large cannula (10 to 12 Fr) through the umbilical artery into the

aorta in preterm sheep, this is not possible in premature human infants due to umbilical artery size and

vessel anatomy. It would be very difficult to place even a 6-Fr cannula through the umbilical artery of a

premature human baby. On the other hand, a larger cannula (6 to 8 Fr) can be inserted into the right IJ

vein in premature infants to allow for adequate drainage. Finally, a VV-ECLS AP operates in parallel

with the systemic circulation. Similar to traditional VV-ECMO circuits, the fetal heart would experience

little increased resistance and afterload from the parallel circuit.83,100 Conversely, an AV-ECLS system

with an oxygenator placed in series has the potential to increase stress on the fetal heart and contribute

to cardiac failure.101

Clinical Prognostic Factors

The ideal patient population to benefit from the AP are ELGANs (<28 weeks EGA) who fail the most

aggressive pulmonary therapies, such as low peak pressure ventilation, nitric oxide inhalation,

surfactant administration, and HFOV. Early identification of those with the highest mortality risk has

been validated in this population using mortality prediction tools (CRIB II and SNAPPE II) based on data

obtained in the first 12 hours of life.102–104

Potential Complications and Current Limitations

There are several potential complications and questions that must be addressed in the laboratory before

the AP can be taken to the clinical setting. The first, and most pressing, is the issue of anticoagulation

and the risk of intracerebral or intraventricular hemorrhage (IVH). Issues with IVH were the main

driving force for limiting conventional ECMO to late-gestation newborns.67,105 However, many of these

reports were from a time before activated clotting time (ACT) was routinely followed in systematically

anticoagulated neonates, leading to excessive anticoagulation. More recent studies on the pathogenesis

of IVH suggest that many causes may be iatrogenic. The mechanism of injury may be either due to a

disruption of cerebral blood flow or injury to the relatively weak endothelial tissue of the germinal

matrix.106,107 Disruption of cerebral blood flow can be caused by transfusion or rapid volume expansion

that leads to increases in arterial pressure and cerebral blood flow, or through positive pressure

ventilation that may obstruct venous outflow and result in increased cerebral perfusion pressures.106,108

Avoiding ventilation and providing stable hemodynamics and gas exchange might afford the same

protection against IVH. Nevertheless, to further minimize the risk of IVH, we propose the eventual use

of nonthrombogenic surfaces in the form of biomaterials that release NO to completely eliminate the

need for anticoagulation.69,71,74,75

Prematurity

Extremely premature infants are still “fetuses” living outside the womb and the physiology,

pathophysiology, and treatment should be viewed through the lens of fetal physiology. Premature birth

(<37 weeks EGA) occurs in 11% of births in the United States and remains the leading cause of

morbidity and mortality in industrialized nations accounting for 60% to 80% of deaths in infants

without congenital abnormalities.109 Prematurity affects most organ systems and includes temperature

instability, respiratory failure and CLD, gastrointestinal problems including necrotizing enterocolitis

(NEC), ICH, and sepsis. ICH occurs in approximately 12,000 premature infants every year with 50% to

75% of the survivors developing cerebral palsy, mental retardation, or hydrocephalus and another 25%

of the nondisabled survivors developing psychiatric disorders or problems with executive function.110

Prevention, early recognition, and timely management are crucial in decreasing long-term morbidity

and mortality.

The more premature the infant, the greater the risk for pulmonary complications with ELGANs born

before 28 weeks EGA at greatest risk. As mentioned earlier, these infants are born at the transitional

period between the canalicular and saccular phases which is when the air–blood barrier is just beginning

2857