Randomization, intervention, and data collection
Assignment of treatment [low dose (LD), 0.25mg/Kg/dose dexamethasone, max 4mg/Kg/dose; OR high dose (HD), 0.5mg/Kg/dose dexamethasone, max 8mg/Kg/dose] in the eligible patients was done by stratified block randomization using variable block sizes of four to eight. Random allocation sequence was generated by web-based program by an individual not involved in study. Patients were stratified according to age (<1 years, 1 to <5 years, and > 5 years) and duration of intubation (≤72 hours, >72 hours but< 7 days, and >7 days). Opaque sealed envelopes containing interventional doses of dexamethasone for two assignments were serially numbered on the outside. We could not achieve the blinding as we used two different doses of a commercially available drug preparation. However, none of the investigators were directly involved in the monitoring and clinical decisions during post-extubation period.
Control patients (HD patients) were planned to receive a total of six doses of intravenous dexamethasone (0.5mg/kg/dose, max 8mg/dose), first dose 24 hours before anticipated extubation and then every 6 hours for a total of six doses as per the current practice in the unit. Study patients (LD patients) were planned to receive six doses of intravenous dexamethasone (0.25 mg/Kg/dose, max 4mg/dose) similarly. The extubation was planned during daytime immediately after the 5thdose of dexamethasone. Study protocol was approved by Institute Ethics Committee, Institute Thesis Committee and Departmental Review Board.