Newborns placed on ECMO for respiratory failure survive to hospital discharge roughly 68% of the time, making neonatal respiratory support the most successful category of ECMO use across all age groups. That number comes from the Extracorporeal Life Support Organization (ELSO) international registry, which tracks outcomes for over 100,000 survivors worldwide. But that overall figure masks enormous variation. A baby with meconium aspiration syndrome may have better than 90% odds of going home, while an infant needing ECMO after cardiac arrest faces survival rates closer to one in three. What drives the gap is primarily the underlying diagnosis, though gestational age, the type of ECMO circuit used, the treating hospital’s experience, and complications during the run all shift the odds in ways that matter to families waiting for news.
Why the Underlying Diagnosis Matters Most
The single biggest predictor of whether a baby survives ECMO is why they needed it in the first place. Neonatal ECMO cases fall broadly into respiratory and cardiac categories, and the survival gap between them is substantial.
Among respiratory diagnoses, meconium aspiration syndrome (MAS) carries the best prognosis. Babies who aspirate meconium during or before delivery can develop severe respiratory failure, but their lungs are structurally normal and often recover well once given time on the circuit. Survival for MAS on ECMO exceeds 94%, and these patients tend to have fewer complications during their run than babies with other diagnoses.1PubMed. Extracorporeal membrane oxygenation: use in meconium aspiration syndrome MAS patients also show fewer complications regardless of which type of ECMO circuit is used.2ASAIO Journal. ECMO for Meconium Aspiration Syndrome: Support for Relaxed Entry Criteria
Persistent pulmonary hypertension of the newborn (PPHN) sits in the middle range. When babies with PPHN fail other treatments like inhaled nitric oxide and high-frequency ventilation, ECMO serves as a rescue therapy with a survival rate around 67%.3PubMed. International treatment outcomes of neonates on extracorporeal membrane oxygenation (ECMO) with persistent pulmonary hypertension of the newborn (PPHN): a systematic review The broader picture for PPHN has improved dramatically over the decades. Before inhaled nitric oxide and ECMO were available, only about 64% of PPHN babies survived at all. After these therapies became standard, 30-day survival climbed to roughly 88%.4PubMed Central. Mortality Risk Factors and Survival Outcomes in Infants with Persistent Pulmonary Hypertension of the Newborn
Congenital diaphragmatic hernia (CDH) is the respiratory diagnosis where outcomes are most guarded. In CDH, the baby’s abdominal organs push up through a hole in the diaphragm, crowding the lungs and often causing severe underdevelopment. Among all infants with CDH, overall mortality is around 5%, but that number jumps to about 44% when the baby is sick enough to need ECMO.5PubMed Central. ECMO utilization in infants with congenital diaphragmatic hernia in the USA A recent single-center study found survival to discharge after ECMO for CDH was around 60%, a figure that held steady even as the center became more selective about which CDH babies went on the circuit.6Journal of Pediatric Surgery. Less is more: ECMO utilization and outcomes in congenital diaphragmatic hernia Whether the baby’s repair happens before or during ECMO also changes the math: an older study found 72% survival when repair preceded ECMO, but only 48% when the repair had to be done while the baby was still on the circuit.7PubMed. Survival of patients with congenital diaphragmatic hernia during the ECMO era: an 11-year experience
Cardiac indications carry the lowest survival rates. When ECMO is deployed after cardiac surgery or during cardiac arrest in an infant (sometimes called extracorporeal cardiopulmonary resuscitation, or ECPR), overall survival to discharge for neonates is about 39%.8World Journal for Pediatric and Congenital Heart Surgery. Extracorporeal Cardiopulmonary Resuscitation (ECPR) in Infants and Children: A Single-Center Retrospective Study Some single-center reports are grimmer still, with one showing only 28% overall survival in children with congenital heart disease on ECMO.9PubMed Central. Risk Factors and Outcomes of Children with Congenital Heart Disease on Extracorporeal Membrane Oxygenation—A Ten-Year Single-Center Report Babies with single-ventricle heart defects face especially steep odds, with one analysis identifying single-ventricle physiology as a strong independent predictor of death.10PubMed Central. Extracorporeal Cardiopulmonary Resuscitation in Infants: Outcomes and Predictors of Mortality
The Changing Profile of Babies Who Go on ECMO
One reason survival numbers have drifted downward over time, from about 76% in earlier decades to around 67% more recently for neonatal respiratory ECMO, is that the babies being treated have changed.11PubMed Central. Neonatal respiratory extracorporeal membrane oxygenation and primary diagnosis: trends between two decades Advances in neonatal care, especially inhaled nitric oxide and gentler ventilation strategies, now save many of the babies who previously would have needed ECMO. What remains are the more complex cases. Since 2017, more than a third of neonatal respiratory ECMO runs in the ELSO registry are classified as “Other,” a catch-all for babies whose conditions don’t fit neatly into classic categories like MAS or CDH.12PubMed. Changing populations being treated with ECMO in the neonatal period – who are the others? These cases are harder to treat and harder to predict outcomes for. ECMO runs are getting longer, and family-requested withdrawal is becoming more common as some diagnoses are recognized as incompatible with meaningful recovery.
In practical terms, this means that comparing today’s survival statistics to those from the 1990s or early 2000s is not straightforward. A lower headline number doesn’t mean ECMO has gotten worse; it reflects a sicker, more complex patient population.
Venovenous Versus Venoarterial ECMO
The way the ECMO circuit connects to the baby’s body also affects survival. In venovenous (VV) ECMO, blood is drained from and returned to the venous system, supporting the lungs while the baby’s own heart still does the pumping. In venoarterial (VA) ECMO, the circuit takes over both heart and lung function by draining from the venous side and returning to an artery. VA is necessary when the heart itself is failing, but it comes with trade-offs.
In a large study of pediatric sepsis patients on ECMO, VV survival was 79% compared with 64% for VA, and VA ECMO was associated with roughly double the odds of dying even after accounting for how sick the babies were at the start.13Journal of Pediatric Surgery. Venoarterial versus venovenous extracorporeal membrane oxygenation in pediatric sepsis Neonates in that analysis fared better than older children across the board, with 73% survival compared with just 31% for adolescents. A separate neonatal-focused study of 244 babies found overall survival of 88%, with VA cannulation linked to roughly four times the risk of dying on ECMO compared to VV, even after adjusting for illness severity. The difference was especially striking for CDH and PPHN.14Journal of Perinatology. Cannulation approach and mortality in neonatal ECMO
This doesn’t mean VA ECMO is the wrong choice when it’s chosen. Babies placed on VA tend to be sicker, with more cardiac compromise, which means some of the survival gap reflects patient selection rather than the circuit itself. Still, the trend in neonatal care has been toward using VV whenever the baby’s heart can handle it.
Gestational Age and Birth Weight Thresholds
ECMO has traditionally been reserved for babies born at 34 weeks of gestation or later and weighing at least 2 kilograms (about 4.4 pounds). Those cutoffs were established decades ago because early studies found unacceptable rates of brain bleeding and death in smaller, more premature infants.15PubMed. Reconsidering ECMO in Premature Neonates The worry was straightforward: the blood thinners required to keep the ECMO circuit running put fragile developing brains at high risk of hemorrhage.
More recent data has nudged the boundary. Babies born at 32 to 34 weeks now show improved survival and lower rates of brain hemorrhage compared with earlier eras, prompting some centers to consider ECMO on a case-by-case basis for moderately premature infants.16PubMed. Expanding neonatal ECMO criteria: When is the premature neonate too premature Below 32 weeks, though, the picture remains bleak. An analysis of the ELSO registry found that neonates born at 30 to 31 weeks with birth weights of 1.5 to 2.0 kilograms had in-hospital mortality of 70 to 75%. Both gestational age and birth weight showed a clear linear relationship with survival: every additional week and every additional gram mattered.17PubMed. How low should we go? Outcomes of ECMO in neonates with low gestational age or birth weight
Complications During the ECMO Run
ECMO keeps babies alive when nothing else can, but it is far from gentle on the body. The circuit requires continuous blood thinners to prevent clotting in the tubing, which creates a constant tug-of-war between bleeding and clotting. Both complications cut survival significantly. In a large registry analysis, bleeding was associated with a 41% lower likelihood of survival, and blood clots with a 33% reduction.18PubMed Central. Association of bleeding and thrombosis with outcome in Extracorporeal Life Support
Brain injury is the complication families fear most, and it occurs in about one in five neonates on ECMO. In a review of over 7,000 neonates from the ELSO registry, 20% developed neurologic complications, and mortality among those babies was 62% compared with 36% in those without brain injury.19PubMed. Neurologic complications in neonates supported with extracorporeal membrane oxygenation. An analysis of ELSO registry data These complications include brain bleeds, strokes, and seizures, and they are a major reason ECMO teams perform regular head ultrasounds during the run.
Kidney injury is another frequent problem. One 14-year cohort study found that babies who developed the most severe category of kidney failure on ECMO had only 35% survival to ICU discharge, compared with 78% for babies without kidney injury.20PubMed Central. Acute kidney injury is a frequent complication in critically ill neonates receiving extracorporeal membrane oxygenation: a 14-year cohort study A multicenter study confirmed that kidney injury, measured either by rising creatinine levels or the need for dialysis on the circuit, was independently tied to longer ECMO runs and higher odds of death at discharge.21PubMed Central. The Incidence of Acute Kidney Injury and its Effect on Neonatal and Pediatric ECMO Outcomes: A multicenter report from the KIDMO Study Group
Hospital Volume and Where You Get Treated
Not all ECMO centers are equal, and the hospital where a baby is treated can meaningfully change survival odds. A large analysis of pediatric and neonatal ECMO found that high-volume centers had significantly lower odds of death compared with low-volume centers, even after adjusting for how sick the patients were. The threshold most associated with improved outcomes was about 22 cases per year.22PubMed Central. Pediatric and Neonatal Extracorporeal Membrane Oxygenation; Does Center Volume Impact Mortality? Pediatric-specific ECMO centers that handled high volumes had the lowest risk-adjusted mortality among all hospital types, at about 17%, compared with 38% at high-volume non-pediatric centers.23PubMed. Pediatric Extracorporeal Membrane Oxygenation Mortality Is Related to Extracorporeal Membrane Oxygenation Volume in US Hospitals
For CDH specifically, the volume effect is even sharper. Low-volume centers, defined as those handling fewer than two CDH ECMO cases per year, had 54% higher odds of death compared with high-volume centers.24Pediatric Critical Care Medicine. Center Volume and Survival Relationship for Neonates With Congenital Diaphragmatic Hernia Treated With Extracorporeal Life Support This isn’t surprising given the complexity of managing these patients, but it’s worth knowing if families have any say in where their baby is transferred.
Long-Term Outcomes for Survivors
Surviving ECMO is the first hurdle. The second is what comes after. Babies who make it off the circuit face real risks of long-term health problems, particularly related to the brain and lungs.
Neurodevelopmentally, ECMO survivors tend to look reasonably well in the first couple of years, with many hitting early milestones on time. But problems surface as the demands on the brain increase. By school age, many children show difficulties with processing speed, spatial reasoning, and memory, even when overall IQ scores are in the normal range.25PubMed. Long-term outcome of children treated with neonatal extracorporeal membrane oxygenation: increasing problems with increasing age A review of brain development in ECMO survivors emphasized the potential for lower composite scores in cognitive and motor function, as well as language and learning difficulties, and recommended multidisciplinary follow-up with regular neurological and psychological assessments.26PubMed Central. Brain development in newborns and infants after ECMO
Lung problems are also common among survivors. In a single-center study, 76% of ECMO survivors developed chronic lung disease, and a quarter had severe forms. CDH babies were particularly affected, with 94% developing chronic lung disease versus 60% of babies with other diagnoses.27PubMed Central. Chronic Lung Disease Following Neonatal Extracorporeal Membrane Oxygenation: A Single-Center Experience An older but carefully done pulmonary follow-up study found that ECMO survivors showed airway obstruction, hyperinflated lungs, lower oxygen levels during exercise, and higher rates of exercise-induced wheezing compared with healthy controls. The severity of lung problems correlated with how much oxygen and ventilator pressure the babies needed after coming off ECMO.28Pediatrics. Long-Term Pulmonary Sequelae in Children Who Were Treated With Extracorporeal Membrane Oxygenation for Neonatal Respiratory Failure
None of this means ECMO survivors are destined for poor quality of life. Many do well, especially those with diagnoses like MAS where the lungs were structurally normal to begin with. But families should expect that follow-up care will likely extend years beyond the NICU stay and may include developmental therapy, pulmonary monitoring, and educational support.
Predicting Individual Outcomes
Because survival varies so widely, researchers have developed scoring tools to estimate an individual baby’s risk before and during ECMO. The Neo-RESCUERS score, designed specifically for neonates going on respiratory ECMO, uses variables available at the time of cannulation to predict in-hospital mortality. In validation testing, babies in the lowest-risk group had about 7% mortality, while those in the highest-risk group had about 66% mortality.29PubMed Central. Development and Validation of the Neonatal Risk Estimate Score for Children Using Extracorporeal Respiratory Support For CDH specifically, a separate model that included complications occurring during the ECMO run improved predictive accuracy considerably, with the highest-risk group facing mortality above 85%.30PubMed Central. Development and Validation of Extracorporeal Membrane Oxygenation Mortality-Risk Models for Congenital Diaphragmatic Hernia
A broader tool called the Pediatric ECMO Prediction model works across all indications, not just respiratory failure. The strongest predictors of death in that model were bloodstream infection and the need for ECPR (ECMO started during cardiac arrest), each of which carried roughly four to five times the odds of dying.31PubMed Central. Development of the Pediatric Extracorporeal Membrane Oxygenation Prediction Model for Risk-Adjusting Mortality These tools aren’t meant to tell a family whether their baby will survive, but they help clinical teams frame conversations about likely outcomes and guide decisions about when to continue or redirect care.
What ECMO Costs
ECMO is among the most resource-intensive therapies in all of medicine. The median hospital cost for children supported with ECMO is around $230,000, though bills can easily exceed $400,000 for complex cases. Cost climbs significantly when complications like kidney failure or blood clots develop, and smaller hospitals tend to have higher costs than large academic centers.32PubMed. Hospital Costs for Neonates and Children Supported with Extracorporeal Membrane Oxygenation National trend data show costs increasing by roughly $14,000 per year, reflecting both longer stays and rising medical expenses overall.33PubMed Central. National trends in neonatal extracorporeal membrane oxygenation in the United States
For CDH families specifically, longer ECMO runs are associated with higher lifetime costs but also more quality-adjusted life years gained. An economic analysis estimated that staying on ECMO for more than three weeks for CDH cost about $1 million but yielded roughly 5 quality-adjusted life years, compared with about 1.8 quality-adjusted life years for runs under two weeks.34PubMed. Cost-effectiveness analysis of extended extracorporeal membrane oxygenation duration in newborns with congenital diaphragmatic hernia in the United States This analysis suggests that extended runs, while expensive, can be worthwhile for the right patients. These costs are largely absorbed by insurance and hospital budgets in the U.S., and families rarely see the full figure directly, but the resource intensity of ECMO is one reason it remains concentrated at specialized centers.
Whether Race and Socioeconomic Factors Affect Outcomes
Given the well-documented disparities in neonatal care more broadly, it’s a reasonable question whether a baby’s racial background or family income level predicts ECMO outcomes. The existing data is somewhat reassuring on this point. A study examining ECMO outcomes in relation to social determinants of health found that while the underlying diagnoses differed by group (meconium aspiration was more common in lower-opportunity communities and among Black infants, while CDH was more common among White infants), NICU mortality itself was not significantly different across racial or socioeconomic groups.35Journal of Perinatology. Neonatal extracorporeal membrane oxygenation outcomes according to social determinants of health The researchers interpreted this as suggesting some resilience to social risk factors once a baby is already at an ECMO-capable center. Whether disparities exist in who gets referred or transported to those centers in the first place is a different question and one that is harder to study.