Medical Director, Advanced Cardiac Therapies Program Children's National Hospital, Washington DC Bethesda, Maryland, United States
Abstract:
Background: Utilization of ECMO worldwide is rising however outcomes are still suboptimal. Example, survival in cardiac ECMO for infants is at less than 50%. However, it is challenging to compare data, center performance, outcomes due to variations in practice and conduct of ECMO. We therefore conducted a comprehensive survey of ECMO centers around ECMO program setup, equipment, program protocols and ECMO management to better understand the variability.
Methods: Survey of ECMO centers listed in the ELSO registry database.
Results: 101 ECMO centers from 25 countries responded. Program setup: Primary target population 45 adults, 30 neonatal / pediatric, 5 pediatric, 19 serving all ages. ECMO threshold was variable: some with well defined oxygen index , Murray score, cardiac index while others defined it by ELSO guidelines or subjective clinical assessment. Indications: cardiac failure (50), respiratory failure (18), ARDS (24), PPHN (5) and meconium aspiration (3). Contraindications: neurologic injury was the commonest, followed by age including < 36 weeks gestational age, weight < 2 kg and medical futility. Circuits: unspecified pump in 47 centers, Cardiohelp in 29, centrifugal in 18, rollerhead in 6 and Livanova system in 4 centers. 36 centers used blood prime and crystalloids by rest. Primed circuits were stored for 30 days (range 24 hours to 90 days) by majority. ECMO Management: Adequacy of ECMO support was assessed by clinical assessment and blood gases . Specific monitoring: SVO2 (23 centers) and NIRS (28 centers). Adequacy of CO2 removal was assessed using blood gases by all with 15 centers also using ETCO2. Suction pressures: Neonatal / pediatric centers -20 to-50 mmHg , adults -80 to -150mmHg. Outlet pressures: majority do not follow. Bubble detector were used by 21 centers while others relied on closed loop, bedside monitoring and prevention. Hematologic parameters: Following were ranges reported: Hemoglobin 7-10 (median 8) gm/dl by adult centers and 8-13 (median 10) gm/dl. Platelet count 50-80k/mcl for nonbleeding and >100k/mcl for bleeding patients. AT3 was not routinely measured by 30 centers while 47 measured it daily and remaining centers adopted as needed. Ventilation on ECMO: Rest settings were well defined by centers for VA ECMO while some stating ‘lung protective ventilation’ and others not choosing rest settings. On VV ECMO, 37 centers chose lung collapse strategy while 58 centers maintained lung volumes. Antibiotics prophylaxis outside of cannulations is rare. Weaning trials true clamping 48 centers, rest perform low flow trials. Routine use of bridge during wean was done by 25 centers. ECMO withdrawal for futility is accepted universally however definition of the same is challenging as acknowledged by majority of the responders. Brain death, multi organ failure, poor prognosis and family wishes are some of the reasons mentioned.
Conclusion: In this comprehensive study of ECMO centers, we documented clear and significant variability in ECMO setup, practice , parameters and management protocols. These make comparative analysis of outcomes very challenging and studies directed at identifying appropriate basic parameters and management protocols are necessary.
