Umbilical cannulation optimizes circuit flows in premature lambs supported by the EXTra-uterine Environment for Neonatal Development (EXTEND).

ABSTRACT

KEY POINTS Bronchopulmonary dysplasia is a disease of extreme prematurity that occurs when the immature lung is exposed to gas ventilation. We designed a novel 'artificial womb' system for supporting extreme premature lambs (called EXTEND) that obviates gas ventilation by providing oxygen via a pumpless arteriovenous circuit with the lamb submerged in sterile artificial amniotic fluid. In the present study, we compare different arteriovenous cannulation strategies on EXTEND, including carotid artery/jugular vein (CA/JV), carotid artery/umbilical vein (CA/UV) and umbilical artery/umbilical vein (UA/UV). Compared to CA/JV and CA/UV cannulation, UA/UV cannulation provided significantly higher, physiological blood flows to the oxygenator, minimized flow interruptions and supported significantly longer circuit runs (up to 4 weeks). Physiological circuit blood flow in UA/UV lambs made possible normal levels of oxygen delivery, which is a critical step toward the clinical application of artificial womb technology. ABSTRACT EXTEND (EXTra-uterine Environment for Neonatal Development) is a novel system that promotes physiological development by maintaining the premature lamb in a sterile fluid environment and providing gas exchange via a pumpless arteriovenous oxygenator circuit. During the development of EXTEND, different cannulation strategies evolved with the aim of improving circuit flow. The present study examines how different cannulation strategies affect EXTEND circuit haemodynamics in extreme premature lambs. Seventeen premature lambs were cannulated at gestational ages 105-117 days (term 145-150 days) and supported on EXTEND for up to 4 weeks. Experimental groups were distinguished by cannulation strategy carotid artery outflow and jugular vein inflow (CA/JV; n = 4), carotid artery outflow and umbilical vein inflow (CA/UV; n = 5) and double umbilical artery outflow and umbilical vein inflow (UA/UV; n = 8). Circuit flows and pressures were measured continuously. As we transitioned from CA/JV to CA/UV to UA/UV cannulation, mean duration of circuit run and weight-adjusted circuit flows increased (P < 0.001) and the frequency of flow interruptions declined (P < 0.05). Umbilical vessels generally accommodated larger-bore cannulas, and cannula calibre was directly correlated with circuit pressures and indirectly correlated with flowpressure ratio (a measure of post-membrane resistance). We conclude that UA/UV cannulation in fetal lambs on EXTEND optimizes circuit flow dynamics and flow stability and also supports circuit flows that closely approximate normal placental flow.