While the use of the centrifugal vortex ECMO pump as an alternative to
a roller-occlusion pump offers distinct advantages, unacceptable hemo
lysis may occur during its use in newborn infants. The authors studied
87 consecutive neonatal patients with respiratory failure supported w
ith venoarterial ECMO using a centrifugal vortex pump. Baseline mean p
lasma free hemoglobin level for all patients during the first 48 hours
of bypass was 31.3 +/- 3.1 mg/dl. In 51 patients, an abrupt rise in t
he plasma free hemoglobin occurred. In 48 of 51 patients, the pump hea
d alone was changed at 91.9 +/- 6.6 hours after initiation of bypass.
Mean plasma free hemoglobin decreased to baseline values following pum
p head change. The authors could not determine any factors that distin
guished the infants who developed hemolysis from those who did not. Ch
anging only the pump head provides a simple approach to hemolysis enco
untered during use of the centrifugal vortex pump in newborn infants.