Which oxygen saturation level should we use for very premature infants? A randomised controlled trial

Death or serious neurosensory disability at 2 years corrected for prematurity

Added 30/11/2015:
Serious neurosensory disability was defined before recruitment commenced as: a cognitive score of <70 (i.e. more than 2 standard deviations below the mean) using the Bayley Scale of Infant Development (BSID-3); severe visual loss (certifiable as legally blind or partially sighted); severe cerebral palsy (unable to walk without help at 2 years); deafness requiring (or too severe to benefit from) a hearing aid.

Before the neurodevelopmental assessments even began, a cut-off score <85 on the cognitive or language component of the Bayley-III was decided because this matched a cognitive score <70 on the BSID-II that was used from the outset in some of the NeOProM trials. Bayley-III assessments could not always be arranged. To minimize the risk of bias from post-randomization exclusions, alternative corroborative measures of cognition and language were therefore pre-specified in the Statistical Analysis Plan (SAP) on 27/11/2013 and in the final version of the SAP that was signed off on 31/03/2014 prior to unblinding or analysis of the results. Serious neurosensory disability was assessed blind to trial group assessment. Paediatricians completed a follow-up assessment including information about visual function, hearing, gross motor function, the results of the Bayley-III test or any other test of cognitive function, an assessment of language skills, an assessment of the degree of any developmental delay and information about general health. Parents were asked to fill in a parental questionnaire including information about general health, strengths and difficulties and Parent Report of Children’s Abilities–Revised (PARCA-R). If a Paediatrician report could not be obtained or was incomplete the missing information was sought from the family General Practitioner (GP). Tests reported on the Paediatrician form included the Wechsler Preschool and Primary Scales of Intelligence (WPPSI-III), the Denver Developmental Screening Test, the Griffiths Mental Development Scales, and the Schedule of Growing Skills. The primary outcome of death or serious neurosensory disability was first defined by death, severe visual loss, deafness or cerebral palsy. In the remaining infants the cognitive measure of serious neurosensory disability was first defined as a cognitive or language score <85 on the Bayley-III. If this was not available the outcome was classified using the Paediatrician’s assessment of developmental delay or language delay and then by the GP assessment. Free text on the forms returned by health professionals and parents was assessed independently by two assessors masked to group assignment to adjudicate cognitive outcome in a small number of cases. A secondary analysis of the results was pre-specified, excluding the alternative measures of disability.

1. Respiratory outcomes:
1.1. Days of endotracheal intubation
1.2. Days of nasal continuous positive airway pressure
1.3. Supplemental oxygen at a postmenstrual age of 36 weeks
1.4. Days of oxygen prior to home discharge
1.5. Days in oxygen after home discharge
2. Retinopathy of prematurity (ROP), plus disease, stage 3 and 4 disease
3. Patent ductus arteriosus requiring medical or surgical treatment
4. Necrotising enterocolitis, Bell stage 3 or 4
5. Changes in weight and head circumference from birth to 36 weeks postmenstrual age, and 2 years after delivery was due
6. Retinal structure when last seen for ophthalmic review; outcomes at age 2 years
7. Re-admissions to hospital until 2 years after delivery was due (and their cause)
8. Cerebral palsy (and its severity)
9. Visual disability
10. Deafness
11. Developmental delay using the Bayley Test Mental Developmental Index (MDI)
12. Other disability not classifiable as neurosensory in origin
13. All postneonatal (>27 days) deaths, together with their immediate and underlying cause