Skip to main content
. 2014 Sep 9;2014(9):CD008638. doi: 10.1002/14651858.CD008638.pub2

Ma 2010.

Methods  
Participants Age ≥ 18 years
Not ventilated at time of ICU admission
Ventilation time > 48 hours
Improvement in condition after treatment
Stable vital signs
Participants meeting following criteria for weaning:

  1. Causes for respiratory failure and mechanical ventilation have been resolved or significantly improved

  2. P/F ratio > 200 ; PEEP ≤ 5 cm H2O; FiO2 ≤ 0.4; pH ≥ 7.25 (or for COPD, pH ≥ 7.30, PaO2 > 50 mmHg, FiO2 < 0.35)

  3. Haemodynamic stability, mean arterial pressure ≥ 65 mmHg without use of vasoactive drugs and no sedatives within 24 hours

  4. Ability to breathe independently

  5. Significantly improved pulmonary symptoms and chest x‐ray with no new infections


Did not include participants who passed an SBT
Interventions SmartCare™ (SC) versus synchronized intermittent mandatory ventilation, pressure ventilation (SP) group with T‐piece trials
Outcomes Weaning time (time from randomization to extubation)
Length of ICU stay
Clinician workload (ventilator adjustments per participant)
Ventilator‐associated pneumonia
Adverse event: reintubation
Adverse event: tracheostomy
Adverse event: pneumothorax
Adverse event: other—subcutaneous emphysema
Notes  
Risk of bias
Bias Authors' judgement Support for judgement
Random sequence generation (selection bias) Low risk Minimum balance index based on ICU admission sequence
Allocation concealment (selection bias) Unclear risk Strategy for allocation was based on gender, age, APACHE score, with points assigned to each category (gender: M vs F, age 18 to 44, 45 to 64, ≥ 65 and APACHE < 10, 11 to 15, > 15). Theoretical permutations were run (if assigned to SC or SP weaning), and permutation with lowest cumulative number of points determined treatment assignment for the next participant. Allocation was the responsibility of the researcher implementing the study. Participant assignment was not changed or reconsidered following randomization (i.e. cross‐overs)
Blinding of outcome assessment (detection bias) 
 All outcomes Unclear risk One investigator assessed and recorded study outcomes
Incomplete outcome data (attrition bias) 
 All outcomes Unclear risk Study withdrawals were due to consent withdrawal and VAP. We are uncertain of the numbers, but study authors confirmed that they were equally distributed between treatment groups. This study reported on 30 participants in the SC arm and 32 in the SP (control) arm
Selective reporting (reporting bias) Low risk Study author did not report summary continuous outcomes but provided subgroup data to enable computation of weaning time and length of ICU stay. This trial followed participants in the ICU but did not report ICU mortality
Did the trial stop early for benefit? Low risk Stopped early for futility. Study authors intended to enrol 100 participants. However, because of time constraints and graduate degree requirements, the trial was stopped early
Participants analysed according to the group allocated to? Low risk Study author confirmed that participants were analysed according to treatment assignment at admission