Effect of high-flow oxygen versus T-piece ventilation strategies during spontaneous breathing trials on weaning failure among patients receiving mechanical ventilation: a randomized controlled trial

Abstract

Background A spontaneous breathing trial (SBT) is used to determine whether patients are ready for extubation, but the best method for choosing the SBT strategy remains controversial. We investigated the effect of high-flow oxygen versus T-piece ventilation strategies during SBT on rates of weaning failure among patients receiving mechanical ventilation. Methods This randomized clinical trial was conducted from June 2019 through January 2022 among patients receiving mechanical ventilation for ≥ 12 h who fulfilled the weaning readiness criteria at a single-center medical intensive care unit. Patients were randomized to undergo either T-piece SBT or high-flow oxygen SBT. The primary outcome was weaning failure on day 2, and the secondary outcomes were weaning failure on day 7, ICU and hospital length of stay, and ICU and in-hospital morality. Results Of 108 patients (mean age, 67.0 ± 11.1 years; 64.8% men), 54 received T-piece SBT and 54 received high-flow oxygen SBT. Weaning failure on day 2 occurred in 5 patients (9.3%) in the T-piece group and 3 patients (5.6%) in the high-flow group (difference, 3.7% [95% CI, − 6.1–13.6]; p = 0.713). Weaning failure on day 7 occurred in 13 patients (24.1%) in the T-piece group and 7 patients (13.0%) in the high-flow group (difference, 11.1% [95% CI, − 3.4–25.6]; p = 0.215). A post hoc subgroup analysis showed that high-flow oxygen SBT was significantly associated with a lower rate of weaning failure on day 7 (OR, 0.17 [95% CI, 0.04–0.78]) among those patients intubated because of respiratory failure (p for interaction = 0.020). The ICU and hospital length of stay and mortality rates did not differ significantly between the two groups. During the study, no serious adverse events were recorded. Conclusions Among patients receiving mechanical ventilation, high-flow oxygen SBT did not significantly reduce the risk of weaning failure compared with T-piece SBT. However, the study may have been underpowered to detect a clinically important treatment effect for the comparison of high-flow oxygen SBT versus T-piece SBT, and a higher percentage of patients with simple weaning and a lower weaning failure rate than expected should be considered when interpreting the findings. Clinical trial registration This trial was registered with ClinicalTrials.gov (number NCT03929328) on April 26, 2019.

Spontaneous Breathing Trial Techniques for Extubating Adults and Children Who Are Critically Ill

“…Of the 19 studies further assessed for eligibility, 9 were excluded (Figure 1). In addition to 31 previously included trials, we identified 10 new full publications, including 1 full publication that replaced a previously included abstract publication. Of the 40 trials included (n = 6797 randomized and n = 6716 patients with outcomes), 9 were new trials (n = 3130 patients analyzed), 6 trials compared 3 SBT techniques (but for 1 of the 3-arm trials, the 5-cm H 2 O and 10-cm H 2 O CPAP groups were combined for a single comparison with the third T-piece group), and 1 trial compared 4 SBT techniques (Table 1).…”

“…Twenty-one trials (52.5%) (5 new and 16 previously included) and 23 trials (57.5%) (6 new and 17 previously included) were judged to be at low risk of bias for randomization and allocation concealment, respectively. Four trials (10.0%) were judged to be at high risk of bias with regard to allocation concealment, including 1 quasi-randomized trial allocating patients based on even or odd days.…”

“…Only 2 trials (5.0%) assessed outcomes in a blinded manner. Twenty-one trials (52.5%) were judged to have complete outcomes reporting, including 6 new trials . Finally, 27 trials (67.5%) conducted an intention-to-treat analysis, and 33 trials (82.5%) did not stop early for benefit.…”

“…Twenty-one trials (52.5%) were judged to have complete outcomes reporting, including 6 new trials . Finally, 27 trials (67.5%) conducted an intention-to-treat analysis, and 33 trials (82.5%) did not stop early for benefit. Overall trial quality was moderate (eFigure in Supplement 1).…”

“…Twenty-one trials (52.5%) 30, 31, 40-44, 46-49, 52, 53, 59, 60, 62-65, 67, 69 were judged to have complete outcomes reporting, including 6 new trials. [62][63][64][65]67,69 Finally, 27 trials (67.5%)…”