Download Full Text (372 KB)

Presentation date

Summer 8-10-2022

College, Institute, or Department


Faculty Mentor

Amanda Marshall

Research Mentor

Amanda Marshall


The data collected in this project will be used as a part of a multi-center study conducted by Boston Children’s Hospital to assess if the software Etiometry® can be used to predict extubation success in patients in the CICU. After compiling the data from EPIC, the goals of our analysis were to find any associations with extubation failure rates. Trends in this data could help identify potential areas of focus to mitigate extubation failures. Some specific hypotheses that were explored included the associations between extubation failure and delayed sternal closure, increased length of stay, length of post-op stay, ECMO, and death. The similarity between CHMC’s rate of extubation and what is being reported nationally was also considered. Data was collected by reviewing patient charts in EPIC and analyzed using Chi squared and the Student’s T-test where appropriate. This study found that neonatal status and escalated respiratory support in within 48 hours trended towards an association with extubation failure. It was also found that extubation failure trended towards an association with death. Within the collected data, there was not one overarching diagnosis that was more common in failed extubation encounters than successful encounters. While the conclusions from the data collected in this study are limited by the sample size, this data will benefit the multi-center study conducted by Boston Children’s Hospital in assessing the accuracy of Etiometry® in predicting extubation success.


Extubation, Etiometry, Risk Analysis, Respiratory Support, Mechanical Ventilation

Risk Analysis to Predict Extubation Failure in the Cardiac Intensive Care Unit