VV ECMO is not widely used. A recent collaborative multicenter trial of neonatal ECMO showed that only 25% of the patients received VV ECMO (Lancet 1996;348:75-82). We retrospectively reviewed the charts of all neonatal ECMO candidates admitted to the NICU at the RAH since the introduction of VV ECMO in March 1993 to December 1996. Sixty-five near-term and term neonates were supported with ECMO during this period. Fifty-seven (88%) were initially supported by VV ECMO and 8 (12%) by venoarterial (VA) ECMO. Five (9%) of those initially supported by VV ECMO required conversion to VA ECMO for myocardial dysfunction. Six of the 8 VA ECMO neonates were placed on VA ECMO due to cannulation failure (small jugular vein) and the other two because of myocardial dysfunction. The neonates placed on VV ECMO were unstable with an oxygenation index of 89 ± 40 (mean ± SD), and requiring inotropic support of dopamine at 18 ± 7 mcg/kg/min and epinephrine at 1.2 ± 1 mcg/kg/min. The following table compares survival between the RAH and the Extracorporeal Life Support Organization(ELSO) data (January 1996) according to diagnosis when only VV ECMO was used. Fisher's exact test was employed.
This study shows that our policy of preferentially using VV ECMO does not result in increased mortality and that VV ECMO can be successfully used in unstable newborns.
