Are Extracorporeal Membrane Oxygenation Circuits That Are Primed with Plasmalyte and Stored a Likely Source of Infection

The use of extracorporeal membrane oxygenation (ECMO) has become a standard technique to provide temporary respiratory and cardiovascular support to mature neonate and pediatric patients. Cardiopulmonary bypass allows blood to be oxygenated extra corpus, circumventing the need for gas by the lungs. Each ECMO circuit comprises a pump, an oxygenator, and a warming unit linked by plastic tubing to cannulas (2). Nosocomial infection is recognized as a complication of ECMO, principally due to multiple portals of entry in the circuit (5). We have previously reported (20th Annual Children’s National Medical Centre Symposium ECMO & Advanced Therapies for Respiratory Failure, February 2004, Keystone, Colo.) that ECMO circuits with hollow-fiber oxygenators can be stored without loss of function for up to 2 weeks, allowing rapid access to this complex and expensive life support. Our study also showed that the sterility of circuits could be maintained over this time; however, our microbiology was restricted to culture-dependent techniques. The possible existence of uncultivatable microorganisms could threaten quality assurance (6). We describe the use of universal oligonucleotide primers to amplify, by PCR, highly conservative genetic loci in both bacteria and fungi to validate aseptic storage of ready-to-use ECMO circuits. Ten ECMO circuits were primed under sterile conditions with 1 liter of plasmalyte (i.e., “wet primed”) and stored for either 7 (n 5) or 14 (n 5) days at 8°C with a pump speed of 10 rpm and an oxygenator gas flow of 0.2 liters/min. Each day, 50 ml of plasmalyte was aseptically removed for microbiological and pharmaceutical analyses. Previously described methods for the extraction and amplification of bacterial (3, 4) and fungal (1) DNA were used to recover any contaminating nucleic acids from 1-ml aliquots of each 50-ml sample. The PCR mixture was added directly after this extraction following standard parameters for a commercial polymerase (7). Despite repeated attempts, using variable annealing temperatures, no amplification products could be detected from any of the ECMO samples. Clinical isolates obtained from several London teaching hospitals were grown in plasmalyte and used as positive controls to test the PCR methodologies. Amplification products could be detected by using DNA templates from each isolate; these isolates were identified as Enterococcus faecalis, Escherichia coli, Staphylococcus aureus (methicillin resistant), Candida albicans, Candida glabrata, and Candida krusei. Taken together, our results from culture-dependent and independent assessment of sterility strongly suggest that circuits stored “wet primed” are not a likely source of infection for patients receiving ECMO life support.