CRITICAL CARE: ORIGINAL RESEARCH| VOLUME 163, ISSUE 1, P115-127, JANUARY 2023Download Full Issue

A 30-Minute Spontaneous Breathing Trial Misses Many Children Who Go On to Fail a 120-Minute Spontaneous Breathing Trial

Kelby E. Knox, Justin C. Hotz, Christopher J.L. Newth, et al

Chest 2023; 163: 115-127 Published:August 26, 2022

DOI:https://doi.org/10.1016/j.chest.2022.08.2212

Background

The optimal length of spontaneous breathing trials (SBTs) in children is unknown.

Research Questions

What are the most common reasons for SBT failure in children, and when do they occur? Can clinical parameters at the 30-min mark of a 120-min SBT predict outcome?

Study Design and Methods

We performed a secondary analysis of a clinical trial in pediatric ARDS, in which 2-h SBTs are conducted daily. SBT failure is based on objective criteria, including esophageal manometry for effort of breathing, categorized as passage, early failure (≤ 30 min), or late failure (30-120 min). Spirometry was used to calculate respiratory rate (RR), tidal volume (Vt), and rapid shallow breathing index (RSBI), in addition to pulse oximetry and capnography. Predictive models evaluated parameters at 30 min against SBT outcome, using receiver operating characteristic plots and area under the curve.

Results

We included 100 children and 305 SBTs, with 42% of SBTs being successful, 32% failing within 30 min, and 25% failing between 30 and 120 min. Of the patients passing SBTs at 30 min, 40% went on to fail by 120 min. High respiratory effort (esophageal manometry) was present in > 80% of failed SBTs. At the 30-min mark, there were no clear thresholds for RR, Vt, RSBI, Fio₂, oxygen saturation, or capnography that could reliably predict SBT outcome. Multivariable modeling identified RR (P < .001) and RSBI > 7 (P = .034) at 30 min, pre-SBT inspiratory pressure level (P = .009), and pre-SBT retractions (P = .042) as predictors for SBT failure, but this model performed poorly in an independent validation set with the receiver operating characteristic plot crossing the reference line (area under the curve, 0.67).

Interpretation

A 30-min SBT may be too short in children recovering from pediatric ARDS because many go on to fail between 30 and 120 min. Reassuring values of Vt, RR, and gas exchange at 30 min do not reliably predict SBT passage at 2 h, likely because they do not capture the effort of breathing.

Clinical Trial Registration

ClinicalTrials.gov; No.: NCT03266016; URL: www.clinicaltrials.gov