O2057: Terminal Extubation and Terminal Weaning: Withdrawal of Life Sustaining Therapy in Children. Secondary Analysis of the DONATE Study

Background: Terminal extubation (TE) and terminal weaning (TW) during withdrawal of life sustaining therapies (WLST) have been described and defined in adults. The recent Death One Hour After Terminal Extubation (DONATE) study aimed to validate a model developed to predict whether a child would die within one hour after discontinuation of mechanical ventilation for WLST. In this pre-planned secondary analysis, we aim to describe and define these practices in children, as well as compare characteristics of patients who did have ventilator support decreased prior to WLST with those who did not.

Methods: Secondary analysis of multi-center retrospective cohort study of 913 patients who died after WLST in ten pediatric ICUs in the United States. TE was defined as no wean of any ventilator support parameter in the 6 hours prior to WLST. TW was defined as any decreasing of ventilator support parameters in the six hours preceding WLST. Clinically relevant decrease of support (TWR; >=25% wean from maximum support within 6 hours prior to WLST for any individual ventilator parameter) was compared with likely incidental decrease of support (TWI; < 25% wean for all individual ventilator parameters).

Results: 71.4% (n=652) had TE without decrease in ventilator support in the 6 hours prior. 11% (n=100) had clinically relevant decrease in ventilator support prior to WLST; 17.6% (n=161) had likely incidental decrease in ventilator support prior to WLST. TWR subjects were noted to have clinically relevant decreases of their FiO2 and/or ventilator set rates prior to TE. There were single subjects each with >25% decrease of MAP, PEEP or PS. There was notable consistency across nine of the participating institutions with 67.9-83.6% (median of overall sample 71.4%) TE, however one of the institutions demonstrated a differing practice pattern with only 35.4% TE. Patients with respiratory illness were not more likely to have TW than patients who were comatose. There were no significant differences in any of the other evaluated patient characteristics between groups (weight, BMI, unit type, primary diagnostic category, presence of coma, time to death after WLST, analgosedative requirements, post-extubation respiratory support modality).

Conclusion: Decreasing of ventilator support prior to WLST with extubation in children does occur. This practice does not impact palliative analgosedation requirements or time to death after extubation.

We present these findings on behalf of the DONATE investigators.