State and Regional Variation in Access to Pediatric Extracorporeal Membrane Oxygenation

Extracorporeal membrane oxygenation, or ECMO, is one of the most advanced forms of life support used in pediatric critical care. It serves children with severe cardiac or respiratory failure when conventional therapies are no longer enough. Because ECMO requires specialized equipment, trained teams, and complex hospital systems, it is only offered at select centers. This study, titled State and Regional Variation in Access to Pediatric Extracorporeal Membrane Oxygenation, explores an important health systems question: how evenly is access to pediatric ECMO distributed across the United States?

The authors conducted a geospatial cross-sectional analysis focused on both state boundaries and Pediatric Emergency Referral Regions, or PERRs. This is a valuable design because state policy often drives healthcare organization, while referral regions better reflect where children actually go for specialty care. Using the Extracorporeal Life Support Organization registry, the investigators identified hospitals that had provided pediatric or neonatal ECMO within the previous five years. They then mapped access using ZIP Code Tabulation Areas, population-weighted centroids, and U.S. pediatric population estimates for children younger than 15 years.

The study examined three key access models. First, it measured direct access, defined as being able to reach an ECMO center by car within 60 minutes. Second, it assessed broader access by including indirect access, meaning children could first reach a hospital capable of facilitating timely air transfer to an ECMO center. Third, it evaluated potential access to extracorporeal cardiopulmonary resuscitation, or ECPR, for children with out-of-hospital cardiac arrest, using a much narrower 15-minute travel threshold.

The findings reveal striking geographic disparities in pediatric ECMO access. The researchers identified 258 pediatric ECMO centers and 169 pediatric ECPR centers in the United States. At the state level, the median number of ECMO centers was four, while the median at the regional level was two. Six states and 14 referral regions had no ECMO center at all. This confirms that pediatric ECMO infrastructure is not evenly distributed and remains concentrated in selected metropolitan and tertiary-care areas.

Direct access to ECMO varied substantially. In the median state, about 69.2% of children had direct driving access within 60 minutes. At the regional level, the median was 70.0%. Importantly, 14 states and 27 referral regions had direct access rates below 50%, showing that many children still live far from lifesaving ECMO-capable centers. This pattern is particularly concerning for remote and rural parts of the western United States, where long distances and low population density make the creation and maintenance of high-volume specialty centers difficult.

The picture changed when indirect access through transfer networks was added. Under this broader model, ECMO access became close to universal in many areas. The median state had 98.5% access through direct and/or indirect pathways, and the median region had 99.7%. Even so, some states and regions still remained below 50%, demonstrating that transport systems, while powerful, do not fully eliminate inequities. The study therefore suggests that strengthening transfer infrastructure, including air transport capability, could meaningfully expand access in underserved areas.

Perhaps the most sobering findings involved ECPR access. ECPR is a highly time-sensitive intervention that depends on immediate recognition, rapid transport, and seamless activation of ECMO teams. The study found that only a small percentage of children live close enough to a pediatric ECPR center to make this realistic. In the median state, only 11.2% of children had potential ECPR access, and at the regional level the median was 10.3%. Only two states, the District of Columbia and Rhode Island, exceeded 25% access. This underscores how rare timely pediatric ECPR access remains in the U.S.

From a policy and systems perspective, this article is highly relevant. It shows that state and regional variation in pediatric ECMO access is not just a theoretical issue but a measurable structural disparity. The authors argue that these findings can help policymakers, children’s hospitals, and regional leaders decide where to develop new ECMO programs, when to improve interfacility transfer systems, and how to balance competition versus centralization among existing centers. In regions with no ECMO centers, leaders may need to consider whether a new program is feasible. In regions with many centers, the focus may need to shift toward coordination, case volume, and quality outcomes.

The paper also connects access questions to broader concerns about pediatric healthcare regionalization in the United States. It highlights how children often cross state borders for specialized care and why referral regions may sometimes be more informative than political boundaries. This makes the study useful not just for ECMO specialists, but also for hospital administrators, emergency planners, and health policy researchers interested in equitable access to pediatric critical care.

The study has limitations. The ELSO registry is voluntary, so some ECMO-capable centers may not be captured, while others classified as pediatric centers may primarily care for larger adolescents. The transport assumptions are also modeled rather than observed and may overestimate access when weather, staffing, or infrastructure constraints reduce helicopter availability. Even with these limitations, the work provides an important best-case estimate of pediatric ECMO and ECPR access in the current U.S. system.

Overall, this is a strong and timely health services study. It does not test a new therapy, but it addresses a major determinant of whether children can receive advanced life support at all. Its findings are directly actionable and offer a clear roadmap for improving pediatric ECMO equity across states and regions. 

3
This is a strong geospatial health services study with a clear methodology and meaningful national relevance, but it is observational and model-based rather than a randomized or interventional clinical study. It provides important systems-level evidence, though not the highest level of causal scientific evidence.