[Outcome of pediatric extracorporeal membrane oxygenation in a single center]

Objective: To investigate the application and outcome of pediatric extracorporeal membrane oxygenation (ECMO) in a single center. Methods: The clinical data of 52 pediatric patients with cardiopulmonary failure received ECMO support in Bayi Children's Hospital Affiliated to General Hospital of Beijing Military Command of PLA were collected from January 2012 to October 2016. All patients were divided into two stages by time. January 2012 to December 2014 was stage one. January 2015 to October 2016 was stage two. A retrospective analysis was done for these patients between two stages. In addition, all clinical data were compared with the data of extracorporeal life support organization (ELSO). The constituent ratio differences in different groups were tested by chi square test. Results: In 52 cases, there were 40 boys and 12 girls, aging from 1 day to 7 years, weighing from 2 to 20 kg. There were 35 cases who successfully weaned from ECMO (67%), and 25 cases were able to be discharged alive (48%). In stage one, there were 24 ECMO cases, 18 boys and 6 girls. There were 15 cases successfully weaned from ECMO (63%). Nine patients survived until discharge (38%). Complications were found in 15 cases during ECMO support (63%). In stage two, there were 28 ECMO cases, 22 were boys and 6 were girls. There were 20 cases successfully weaned from ECMO (71%). Sixteen patients survived until discharge (57%). Complications were found in 12 cases during ECMO support (43%). There was no significant difference in survival rates between two stages. However, the neonatal survival rate was higher in stage two than in stage one (71% (12/28) vs. 31% (5/24), χ(2)=5.107, P=0.038). The proportion of respiratory support was higher in stage two than in stage one (50% (14/28) vs. 21% (5/24), χ(2)=4.741, P=0.029), while the proportion of extracorporeal cardiopulmonary resuscitation (ECPR) decreased significantly (21% (6/28) vs. 67% (16/24), χ(2)=10.835, P=0.001). Application of peritoneal dialysis treatment in stage two was higher (6 vs. 0 cases, χ(2)=8.097, P=0.025). Mortality of ECMO was still higher than that of ELSO (48% (25/52) vs. 62% (34 655/55 886), χ(2)=4.281, P<0.05). The constituent ratio of different types of support varied between ECMO and ELSO patients (χ(2)=19.562, P<0.001). Conclusions: ECMO technology can provide effective support for severe cardiopulmonary failure in critically ill children. Due to the multidisciplinary nature of ECMO technology, the complexity and characteristics of pediatric patients, it takes long time to improve ECMO management and prognosis.