Prognosis and outcome of neonates treated either with veno-arterial (VA) or veno-venous (VV) ECMO.

Abstract A comparison was done between neonates requiring veno-arterial (VA) ECMO (too small jugular vein, inability to insert a 12 Fr double lumen catheter or cardio-circulatory instability) and neonates treated with veno-venous (VV) ECMO in the same period of time. From 1991-1995 ECMO was done in 48 neonates after failure of maximum conventional treatments, NO-inhalation and HFOV. 30/48 babies were treated with VV-ECMO, with a switch to VA-ECMO later on in 3 of them. In 18 infants VA-ECMO was installed primarily. Differences between the VA- and VV-ECMO group were: the OI was higher in the VV-treated babies (62 +/- 20 vs. 48 +/- 13, p < 0.03), as were birth weight (3385 +/- 570 vs. 2963 +/- 653 g, p < 0.04), gestational age (39.7 +/- 1.6 vs. 37.9 +/- 2.7 weeks, p < 0.01) and MAP (18.7 +/- 2.2 vs. 17.1 +/- 2.4 cm H2O, p < 0.05). Severe ICH's occurred more frequently in the VA-treated babies (29 vs. 7%, p < 0.05), the rate of other complications was equal. The mortality rates were 43% (VA) and 15% (VV), p < 0.05. About one third of neonatal ECMO candidates will be treated with VA-ECMO, even if the VV-ECMO technique is available. Need for VA-ECMO implies--due to a higher number of preterm babies and a greater severity of illness before ECMO--a higher incidence of ICH's and a higher mortality rate.