Hemolytic and thrombocytopathic characteristics of extracorporeal membrane oxygenation systems at simulated flow rate for neonates.

Use of extracorporeal membrane oxygenation (ECMO) in the treatment of acute respiratory and cardiac failure has become an accepted modality in many institutions for patients failing conventional therapy. It has been shown to significantly reduce mortality and morbidity in critically ill infants and adults (1-6). The success of ECMO in providing gas exchange and circulatory support has led to an increase in its use in increasingly complex medical and surgical patients (4, 7). Recently, new technologies for ECMO in terms of centrifugal pumps and hollow-fiber oxygenators have become available in the marketplace. Although practitioners have been slow to adopt these newer ECMO components as a result of past concerns of severe hemolysis, many centers are now adopting these newer ECMO components. According to a 2008 survey of North American active ECMO centers, >80% routinely used roller pumps for neonatal ECMO (8). This study also showed that these centers primarily used the classic silicone membrane oxygenators (67%) in comparison to the centers using microporous hollow-fiber oxygenators (19%) and polymethylpentene hollow-fiber oxygenators (14%). A follow-up survey in 2010 found that the majority of centers had switched to hollow-fiber oxygenators, although centrifugal pump use remains less than roller pumps. The impact of these new systems on hemolysis and safety has not been thoroughly addressed. Studies have shown that the release of cell-free plasma hemoglobin, a marker of hemolysis; generated during ECMO is associated with renal dysfunction, increased blood transfusions, clot formation in the circuit, and death in children on ECMO after cardiac surgery (9, 10). A 2004 Extracorporeal Life Support Organization registry analysis revealed that hemolysis occurred in 13.6% of all ECMO procedures with a higher incidence of dangerous hemolysis in centrifugal pumps (11). Prediction of severe hemorrhage during pediatric ECMO is challenging, because there is no currently acceptable specific clinical or laboratory marker (12, 13). A recent study showed that ECMO components can induce platelet microaggregates associated with impaired neurophysiological function (14) and can trigger microhemorrhage and inflammatory reactions. Previous published data comparing ECMO components at low flow rates simulating neonatal ECMO conditions is limited. In 1996, a study by Moon et al (15) showed in a low-flow environment that there was no significant difference in hemolysis between previously used pumps. Our hypothesis is that at low flow rates, differences in hemolysis and platelet aggregation are negligible when comparing a centrifugal pump/hollow-fiber oxygenator system with a traditional roller-pump/silicone membrane system.