Percutaneous Cannulation in Predominantly Venoarterial Extracorporeal Membrane Oxygenation by Intensivists.

Critical Care Medicine www.ccmjournal.org e595 The authors reply: The letter by Burrell et al (1) describing the percutaneous cannulation experience by intensivists at Alfred Hospital provides important information for intensivists involved in extracorporeal life support (ECLS) programs and nicely complements our reported experience (2). Our ECLS program has historically provided predominantly respiratory support, and our cardiac support experience is more limited. Our ECLS experience consists mostly of venovenous cannulation with a smaller venoarterial cohort. Burrell et al (1) describe their cannulation experience, which includes a large venoarterial cohort. Aspects of the Alfred Hospital experience deserve mention. The first is their success with arterial cannulation with a low complication rate. Arterial cannulation entails more inherent risk than venous because arterial injury can lead to substantial more morbidity than venous injury. The 8% arterial complication rate reported, although higher than their venous rate (2%), is not unexpected and would not be considered excessive. Some of their complications required surgical management, so the availability of surgical services would seem advisable. Another aspect is the larger number of intensivists (16) available to perform cannulations. Although this would seem to dilute the experience of each intensivist, the approach of having more than one intensivist cannulating a given patient (as we do) increases the exposure to the procedure and helps maintain cannulation skills. secured. A second patient’s internal jugular venous access cannula was accidentally dislodged and removed but was resecured without major consequence. Overall average ECMO duration was 4 ± 4 days for VA ECMO and 14 ± 4 days for VV ECMO. Survival to hospital discharge was 73 of 122 (60%) for VA ECMO (including extracorporeal CPR) and 32 of 45 (71%) for VV ECMO. In total, seven of 167 patients (4%) had hemorrhagic or ischemic stroke, and there was one blood stream infection, and one confirmed cannula related infection. In conclusion, intensivist-led cannulation is also possible for a predominantly VA ECMO service and is associated with an acceptable complication rate. The authors have disclosed that they do not have any potential conflicts of interest.