End-Tidal Carbon Dioxide in Pediatric Cardiac Arrest: Association of Temporal Trends and ROSC

Tuesday, March 24, 2026

10:00 AM - 11:00 AM Central Time

Location: Connections Central - RST 18

Jill L. Sorcher, MD

Children's Hospital of Philadelphia, Pennsylvania

Disclosure(s): No relevant financial relationship(s) to disclose.
RR

Ron W. Reeder, PhD

Professor of Pediatrics
University of Utah

Disclosure information not submitted.
Lindsay Shepard, MD (she/her/hers)

Children's Hospital of Philadelphia

Disclosure(s): No relevant financial relationship(s) to disclose.
KG

Kathryn Graham, MLAS

Research Project Manager
Childrens Hospital of Philadelphia

Disclosure information not submitted.
AT

Alexis Topjian, MD, FCCM, MSCE

Professor
Childrens Hospital of Philadelphia, Pennsylvania

Disclosure information not submitted.
RB

Robert Berg, MD

Professor
Children's Hospital of Philadelphia

Disclosure information not submitted.
RS

Robert Sutton, MD, FCCM

Professor, Anesthesiology and Critical Care Medicine
Children's Hospital of Philadelphia

Disclosure information not submitted.
RM

Ryan W. Morgan, MD, MTR

Associate Professor of Anesthesia and Critical Care
Children's Hospital of Philadelphia

Disclosure information not submitted.

Introduction: In adults, abrupt and sustained increases in end-tidal carbon dioxide (ETCO2) during cardiopulmonary resuscitation (CPR) predict return of spontaneous circulation (ROSC), but this has not been described in children. We aimed to determine the frequency with which ROSC is preceded by a rise in ETCO2 and the temporal association of increases in ETCO2 during CPR with ROSC in children. We hypothesized that ROSC would be preceded by an increase in ETCO2.

Methods: Retrospective cohort study of children in the ICU-RESUS trial who received in-hospital CPR with a tracheal tube in place. Prospectively collected ETCO2 waveforms were quantified. Among all events >/=2 minutes in duration with >/=1 minute of evaluable ETCO2 data, a regression model identified all instances of increases in ETCO2 >/=5, 10, and 20 mmHg during any 1-minute period of CPR. The outcome was sustained ROSC within 2 minutes of such an increase. Among events with sustained ROSC and evaluable ETCO2 data in the 120 seconds pre-ROSC, a simple regression model with fixed categorical patient and continuous time-relative-to-ROSC effects estimated the change in ETCO2 in the 120 seconds pre-ROSC. Patients were classified by pre-ROSC ETCO2 increases of >/=5, 10, and 20 mmHg.

Results: Of 167 included events, the median event-level ETCO2 was 21.7 [13.7, 29.7] mmHg. An increase in ETCO2 >/=5 mmHg occurred 291 times in 139 events; ROSC followed 39/139 (28%) first instances and 56/291 (28%) total instances. An increase in ETCO2 of >/=10 mmHg occurred 178 times in 107 events; ROSC followed 33/107 (31%) first instances and 43/178 (24%) total instances. An increase in ETCO2 of >/=20 mmHg occurred 65 times in 50 events; ROSC followed 17/50 (34%) first instances and 20/65 (31%) total instances. Among 110 events with evaluable pre-ROSC data, the median pre-ROSC rise in ETCO2 was +5.5 [-0.7, +15.7] mmHg; 57 (51.8%), 38 (34.5%) and 22 (20%) events had pre-ROSC increases of >/=5, 10, and 20 mmHg, respectively.

Conclusions: The temporal relationship between increases in ETCO2 and ROSC is variable in children. While a rise in ETCO2 often precedes ROSC, most increases in ETCO2 are often not followed by ROSC within 2 minutes. Further investigation is needed to better guide clinical decision-making.