RESUMO
BACKGROUND: It is common for patients on venovenous extracorporeal membrane oxygenation (VV ECMO) to require continuous renal replacement therapy (CRRT). This can be done using separate vascular access for the CRRT circuit, by placing the CRRT hemofilter within the ECMO circuit, or through a separate CRRT circuit connected to the ECMO circuit. When a CRRT circuit is connected to the ECMO circuit, the inflow and outflow CRRT limbs can both be placed pre-ECMO pump or the CRRT circuit can span the ECMO pump, with the CRRT inflow post-ECMO pump and the outflow pre-ECMO pump. Both configurations require the CRRT alarms to be inactivated due to high positive pressure experienced post-pump and low negative pressure pre-pump. We describe a novel technique that does not require separate venous access and still allows the CRRT alarms to be activated. TECHNIQUE: The CRRT inflow line is connected to the post-oxygenator de-airing port. The CRRT outflow line is connected to the pre-pump side of the ECMO circuit. Pigtails allow for these connections and act as resistors negating the large range of pressures generated by the ECMO centrifugal pump. RESULTS: We implemented this configuration in 11 patients with 100% success rate allowing for alarms to be maintained in all patients. The median number of interruptions per 100 CRRT days was 11.7. The median CRRT filter lifespan was 2.2 days, and the average blood flow was maintained at 311 mL/min. CONCLUSIONS: This configuration allows for efficient use of CRRT in ECMO patients while maintaining the safety alarms on the CRRT machine.
RESUMO
OBJECTIVES: To determine the safety and efficacy of a rapidly deployed intensivist-led venovenous extracorporeal membrane oxygenation cannulation program in a preexisting extracorporeal membrane oxygenation program. DESIGN: A retrospective observational before-and-after study of 40 patients undergoing percutaneous cannulation for venovenous extracorporeal membrane oxygenation in an established cannulation program by cardiothoracic surgeons versus a rapidly deployed medical intensivist cannulation program. SETTING: An adult ICU in a tertiary academic medical center in Camden, NJ. PATIENTS: Critically ill adult subjects with severe respiratory failure undergoing percutaneous cannulation for venovenous extracorporeal membrane oxygenation. INTERVENTIONS: Percutaneous cannulation for venovenous extracorporeal membrane oxygenation performed by cardiothoracic surgeons compared with cannulations performed by medical intensivists. MEASUREMENTS AND MAIN RESULTS: Venovenous extracorporeal membrane oxygenation cannulation site attempts were retrospectively reviewed. Subject demographics, specialty of physician performing cannulation, type of support, cannulation configuration, cannula size, imaging guidance, success rate, and complications were recorded and summarized. Twenty-two cannulations were performed by three cardiothoracic surgeons in 11 subjects between September 2019 and February 2020. The cannulation program rapidly transitioned to an intensivist-led and performed program in March 2020. Fifty-seven cannulations were performed by eight intensivists in 29 subjects between March 2020 and December 2020. Mean body mass index for subjects did not differ between groups (33.86 vs 35.89; p = 0.775). There was no difference in days on mechanical ventilation prior to cannulation, configuration, cannula size, or discharge condition. There was no difference in success rate of cannulation on first attempt per cannulation site (95.5 vs 96.7; p = 0.483) or major complication rate per cannulation site (4.5 vs 3.5; p = 1). CONCLUSIONS: There is no difference between success and complication rates of percutaneous venovenous extracorporeal membrane oxygenation canulation when performed by cardiothoracic surgeons versus medical intensivist in an already established extracorporeal membrane oxygenation program. A rapidly deployed cannulation program by intensivists for venovenous extracorporeal membrane oxygenation can be performed with high success and low complication rates.