DESIGN: Retrospective cohort analysis of children who underwent a systemic-to-pulmonary shunt after excluding those who had it for Norwood or Damus-Kaye-Stansel procedure.
SETTING: The Royal Children's Hospital, Melbourne, VIC, Australia.
PATIENTS: From 2008 to 2015, 201 children who had a systemic-to-pulmonary shunt were included.
INTERVENTIONS: Major adverse event is defined as one or more of cardiac arrest, chest reopening, or requirement for extracorporeal membrane oxygenation. Study outcome is a "composite poor outcome," defined as one or more of acute kidney injury, necrotizing enterocolitis, brain injury, or in-hospital mortality.
MEASUREMENTS AND MAIN RESULTS: Median (interquartile range) age was 12 days (6-38 d) and median (interquartile range) time to major adverse event was 5.5 hours (2-17 hr) after admission. Overall, 36 (18%) experienced a major adverse event, and reasons were over-shunting (n = 17), blocked shunt (n = 13), or other (n = 6). Fifteen (88%) in over-shunting group suffered a cardiac arrest compared with two (15%) in the blocked shunt group (p < 0.001). The composite poor outcome was seen in 15 (88%) in over-shunting group, four (31%) in the blocked shunt group, and 56 (34%) in those who did not experience a major adverse event (p < 0.001). By multivariable analysis, predictors for composite poor outcome were major adverse event due to over-shunting (no major adverse event-reference; over-shunting odds ratio, 18.60; 95% CI, 3.87-89.4 and shunt-block odds ratio, 1.57; 95% CI, 0.46-5.35), single ventricle physiology (odds ratio, 4.70; 95% CI, 2.34-9.45), and gestation (odds ratio, 0.84/wk increase; 95% CI, 0.74-0.96).
CONCLUSIONS: Infants who suffer major adverse event due to over-shunting experience considerably poorer outcomes than those who experience events due to shunt block. A mainly hypoxic event with maintenance of systemic perfusion (as often seen in a blocked shunt) is less likely to result in poorer outcomes than those after a hypoxic-ischemic event (commonly seen in over-shunting).
METHODS AND RESULTS: This was a retrospective review of 198 patients who underwent VA-ECMO after congenital cardiac surgeries at our institute between 2004 and 2023. Patients were divided into pre-ECMO protocol (2004-2017) and post-ECMO protocol (2018-2023) implementation. There were 107 patients in the preprotocol era and 91 in the postprotocol era. We compared weaning from ECMO and survival to hospital discharge between the two eras. An analysis of the factors influencing survival to hospital discharge was also done. ECMO was initiated through the central cannulation technique through median sternotomy in all patients. The median age and weight at initiation were 4 months (interquartile range [IQR] 1-33.5 months) and 4.4 kg (IQR 3.3-10.1 kg), respectively. The successful weaning of the ECMO (n = 67/91, 73.6%) and survival to discharge (n = 43/91, 47.3%) were higher in patients of the postprotocol era. However, it was not statistically significant. Higher risk adjustment for congenital heart surgery-1 >3 and acute kidney injury were independent predictors of poorer survival to hospital discharge.
CONCLUSIONS: A protocol-based ECMO program may improve outcomes of successful weaning and survival to discharge in patients undergoing congenital cardiac surgeries.