Foronda 2011.

Methods	Multicentre, randomized controlled trial
Participants	Setting: two hospitals in Sao Paulo, Brazil. Two paediatric ICUs Inclusion criteria: age between 28 days and 15 years; receiving mechanical ventilation for > 24 hours Exclusion criteria: intubation due to upper airway obstruction; diaphragmatic hernia or paralysis; long‐term ventilator use (dependent on invasive or non‐invasive ventilation before ICU admission); cyanotic congenital heart disease; primary pulmonary hypertension; neuromuscular disease; tracheostomy Participant numbers: 312 randomly assigned; 52 withdrawn (29 from protocol group, 23 from standard care group); 260 analysed
Interventions	Weaning protocol combining daily screening of breathing parameters and a two‐hour spontaneous breathing test Standard care procedures that did not include daily screening or a spontaneous breathing trial. The ventilator mode and settings were selected at the discretion of the attending physician. Pressure‐based ventilatory modes were used more frequently, although other modes were available. The most frequently used modes were pressure support, synchronized intermittent mandatory ventilation and pressure‐controlled ventilation, the latter of which was used more often in severe cases. Respiratory frequency and peak inspiratory pressure reductions were performed according to the presence of ventilatory parameters, including increased chest cage expansion, increased exhaled tidal volume and reduced partial pressure of carbon dioxide in arterial blood or end‐tidal carbon dioxide. Fraction of inspired oxygen and positive end‐expiratory pressure were reduced according to participant oxygenation to maintain arterial oxygen saturation between 92% and 98%
Outcomes	Mechanical ventilation duration (initiation of mechanical ventilation to first extubation) Extubation failure rate Need for postextubation noninvasive ventilation Ventilator‐associated pneumonia Accidental extubation
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The research fellow conducting the daily evaluation randomly assigned eligible patients by randomly selecting a sealed envelope from an opaque plastic bag containing a 1:1 ratio of test/control group numbers
Allocation concealment (selection bias)	Low risk	Sealed envelopes were used
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Medical staff were unaware of participant assignment until the participant passed the daily evaluation and a spontaneous breathing trial was indicated
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Research fellow was responsible for screening, randomisation and daily evaluations but was not involved in the decision to extubate
Incomplete outcome data (attrition bias) All outcomes	Low risk	Recruitment and attrition were reported
Selective reporting (reporting bias)	Low risk	Protocol was registered as ISRCTN37806223 and outcomes were reported
Other bias	Low risk	None was apparent