Multisite Venovenous Extracorporeal Membrane Oxygenation in Pediatric Patients Under 15 kg: Three-Center, Analysis of Surgical Versus Percutaneous Approach and Thrombosis Risk, 2017-2024.

OBJECTIVES

An increasing number of pediatric centers use the femoral vein in neonates and smaller children to provide venovenous multisite (VVMS) extracorporeal membrane oxygenation (ECMO), but there are no studies comparing surgical vs. percutaneous approaches. We investigated the thrombosis risk associated with VVMS, comparing outcomes between the surgical and percutaneous approach.

DESIGN

Retrospective data analysis.

SETTING

Evelina London Children's Hospital, Heim Pal Hospital and Semmelweis University, 2017-2024.

PATIENTS: We included 58 neonatal and pediatric patients weighing less than 15 kg, supported with VVMS ECMO for respiratory disease.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN

RESULTS

We collected patient and venovenous ECMO details. Thrombosis of cannulated vessels, as determined by a pediatric radiologist via follow-up vascular ultrasound, was the primary outcome. In total, 58 patients received VVMS, survived and were assessed for thrombosis. There were 34 surgically cannulated patients with median (interquartile range [IQR]) weight 3.6 kg (IQR, 2.8-7 kg) vs. 24 percutaneously cannulated patients (median weight, 8 kg [IQR, 3.7-12.3 kg]) who were cannulated by pediatric anesthetists or intensive care physicians. Surgical placement of cannulas, in comparison with percutaneous placement, was associated with greater odds of thrombosis of internal jugular and femoral veins (odds ratio, 37; 95% CI, 7-266), although the surgical group were younger and of lower weight ( p < 0.05). For any given weight, percutaneously placed cannulas were smaller than those placed surgically, yet they still provided adequate ECMO flow.

CONCLUSIONS

This retrospective case series of VVMS in neonates and small children (< 15 kg) shows that surgical cannulation was associated with greater odds of vascular thrombosis at vessel sites, but a potential confounder may be that the surgical group were younger and of smaller weight. Overall, for any given weight, percutaneous cannulas were smaller than those placed surgically.