Very low-dose dexamethasone to facilitate extubation of preterm babies at risk of bronchopulmonary dysplasia: the MINIDEX feasibility RCT

BACKGROUND:Postnatal corticosteroids are used to improve lung function and reduce the incidence of bronchopulmonary dysplasia (BPD) in preterm babies. However, corticosteroids may be associated with adverse neurodevelopment. Despite a lack of evidence, some clinicians in the UK use very low-dose regimens of dexamethasone hoping for positive pulmonary effects and optimal neurodevelopment. OBJECTIVES:To assess the efficacy and safety of very low-dose dexamethasone at facilitating the extubation of ventilator-dependent preterm babies born at < 30 weeks’ gestation and who are at high risk of developing BPD. DESIGN:A multicentre, randomised, masked, parallel-group, placebo-controlled Phase 2b trial. The trial was designed as a feasibility study for a subsequent trial of clinical effectiveness. SETTING:The study was set in 11 tertiary neonatal units in the UK. PARTICIPANTS:Ventilator-dependent preterm babies born at < 30 weeks’ gestation aged 10–21 days, receiving at least 30% inspired oxygen and at high risk of developing BPD. Exclusions were babies who had received previous courses of postnatal steroids for respiratory disease; had a severe congenital anomaly affecting the lungs, heart or central nervous system, or had a surgical abdominal procedure or patent ductus arteriosus ligation; and had an illness or medication for which postnatal corticosteroid would be contraindicated (e.g. confirmed or suspected acute sepsis, acute necrotising enterocolitis/focal intestinal perforation or cyclo-oxygenase therapy). INTERVENTIONS:Babies were randomised to very low-dose dexamethasone (50 µg/kg/day for 13 doses) or a matched placebo. Samples of blood and bronchoalveolar lavage fluid from a subset of babies randomised at three participating sites were sent for cytokine analysis at randomisation and at days 5, 7, 10 and 14 of treatment. PRIMARY OUTCOME:Time to extubation. SECONDARY OUTCOMES:Secondary outcomes included rates of extubation by day 7 of the intervention; survival to 36 weeks’ postmenstrual age (PMA) or discharge home; respiratory morbidity to 36 weeks’ PMA or discharge home; cytokine profile; safety outcomes; and parent/family experience. RESULTS:The main metric of feasibility, namely recruitment, proved difficult. There was a tendency for open-label medication and a higher than predicted rate of suspected/confirmed sepsis among babies. Recruitment was halted after 22 babies had been enrolled. It was found that, compared with the placebo group, a higher proportion of babies were extubated at day 7 of life [5/8 (62.5%) in the very low-dose dexamethasone group vs. 2/6 (33.3%) in the placebo group] and duration of invasive ventilation was lower (a median of 23 days for the very low-dose dexamethasone group vs. a median of 31 days for the placebo group) in the very low-dose dexamethasone group. This is supported by a trend for an increased requirement for open-label rescue steroids in control group babies (41.7% in the very low-dose dexamethasone group vs. 80% in the placebo group). Given the limited sample size, only descriptive statistics can be given; firm conclusions cannot be drawn. LIMITATIONS:Small sample size and high rates of open-label treatment use. CONCLUSIONS:It is not feasible to conduct the required pragmatic trial of clinical effectiveness. FUTURE WORK:Assessment of very low-dose dexamethasone in this patient group requires careful consideration. STUDY REGISTRATION:Clinical Controlled Trials ISRCTN81191607. FUNDING:This project was funded by the Efficacy and Mechanism Evaluation (EME) programme, a Medical Research Council and National Institute for Health Research (NIHR) partnership. The report will be published in full in Efficacy and Mechanism; Vol. 6, No. 8. See the NIHR Journals Library website for further project information. The funding for the cytokine analysis is provided by the Children’s Charity Cerebra and is being carried out beyond the lifespan of the NIHR funding