Pressure support versus continuous positive airway pressure for predicting successful liberation from invasive ventilation in children: an open label, randomized non-inferiority trial.

Background: Pressure support (PS) as a spontaneous breathing trial (SBT) was considered inferior to continuous positive airway pressure (CPAP) and T-piece because PS underestimated post-extubation work of breathing in physiologic studies. We aimed to compare PS and CPAP as SBT methods for assessing clinical outcomes in children. Methods: This was an open label randomized non-inferiority trial conducted between December 2019 and August 2021 among children aged 1 month to 12 years deemed ready for weaning after at least 48 h of invasive ventilation in PICU. Children were randomized to undergo a 2-h SBT with PS of 8 cm H Findings: Of the 247 enrolled children, 244 completed the trial (121 in PS and 123 in CPAP group). Median (IQR) age was 24 (9, 84) months. Median (IQR) duration of invasive ventilation before randomization was 4.5 (3, 6.5) days. Successful liberation from invasive ventilation after first SBT occurred in 97 (80.2%) children in PS and 93 (75.6%) children in CPAP group [difference 4.6; 95% CI (-5.8, 15); p = 0.39]. First SBT pass rate between PS and CPAP [111 (91.7%) versus 105 (85.4%); difference 6.3; 95% CI (-1.6, 14.3); p = 0.12] was similar. Need for post-extubation respiratory support [52 (43%) versus 49 (40%)], rate of reintubation within 72 h [14 (11.6%) versus 12 (9.8%)] and median (IQR) length of PICU stay [9 (6, 15) versus 8 (5.5, 13) days] were comparable. Four (1.6%) children, all in CPAP group had unfavourable outcome (1 died, 3 discontinued care). Interpretation: In invasively ventilated children, 2-h SBT with pressure support was non-inferior to CPAP in predicting successful liberation from invasive ventilation. Funding: None.