A composite outcome that measures the occurrence of death or neurodevelomental impairment between birth and 22-26 months corrected age.

The analysis population includes all randomized infants with available data for death and/or neurodevelopmental impairment.

This is measured as Yes if an infant died between birth and 22-26 months corrected age; Otherwise, No.

The analysis population includes all randomized infants with available vital status.

This is measured as Yes if any hearing impairment or visual impairment is noted, if severe or moderate cerebral palsy is noted, or if the cognitive score of the Bayley III score is more than 1 standard deviation below the average; Otherwise, No.

The analysis population includes all randomized infants with available data for neurodevelopmental impairment.

This is measured as Yes if the Bayley Scale of Infant and Toddler Development (BSID)-III cognitive score is more than 1 standard deviation below the average; Otherwise, No.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if the Gross Motor Function Classification System (GMFCS) score is level II or higher; Otherwise, No. Higher values of the GMFCS are worse than lower values; a level of "I" denotes mild cerebral palsy (CP); level "II" or "III" moderate CP; level "IV" or "V" severe CP.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if the corrected visual acuity in the better eye of less than 20/200; Otherwise, No.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if bilateral hearing loss occurred for which hearing aids or cochlear implants were warranted; Otherwise, No.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if survived to discharge without severe morbidity, defined as bronchopulmonary dysplasia, retinopathy of prematurity (stage 3 or higher or requiring treatment), or serious brain abnormality; Otherwise, No.

The analysis population includes all randomized infants with available data up to hospital discharge or death.

This is measured as Yes if experienced bronchopulmonary dysplasia, diagnosed on the basis of the need for supplemental oxygen after a standardized oxygen reduction test at 36 weeks of postmenstrual age; Otherwise, No.

The analysis population includes all randomized infants with available data at 36 weeks postmenstrual age.

This is measured as Yes if experienced Retinopathy of Prematurity (ROP) Stage >=3 or received treatment for that condition; Otherwise, No. Higher stages of ROP indicate a worse outcome; the stages range from 1 for "mild" disease, to 5 for "severe" disease.

The analysis population includes all randomized infants with available data up to hospital discharge or death.

This is measured as Yes if experienced Grade 3 or 4 intraventricularhemorrhage, cystic periventricular leukomalacia, or ventriculomegaly diagnosed on ultrasonographic examination; Otherwise, No.

The analysis population includes all randomized infants with available data up to hospital discharge or death.

This is measured as Yes if experienced necrotizing enterocolitis (NEC), Bell's stage >=2; Otherwise, No. Higher scores of Bell's staging criteria denote a worse outcome, where "1" denotes suspect, "2" definite and "3" advanced NEC.

The analysis population includes all randomized infants with available data up to hospital discharge or death.

This is measured as the number of protocol compliant transfusions, clinically justified non-protocol transfusions and unjustified non-protocol transfusions (violations)

The analysis population includes all randomized infants with available data up to death, hospital discharge, or 36 weeks PMA, whichever occurs first

This is measured as the weight-for-age Z-score at 36 weeks postmenstrual age or at initial hospital discharge, whichever occurs first. The Z-score is determined using Olsen percentile curves, and is derived from a normal distribution, where 0 designates average weight-for-age, and negative scores denote less than average weight-for-age.

The analysis population includes all randomized infants with available data at 36 weeks postmenstrual age or at initial hospital discharge, whichever occurs first.

This is measured as the length-for-age Z-score at 36 weeks postmenstrual age or at initial hospital discharge, whichever occurs first. The Z-score is determined using Olsen percentile curves, and is derived from a normal distribution, where 0 designates average length-for-age, and negative scores denote less than average length-for-age.

The analysis population includes all randomized infants with available data at 36 weeks postmenstrual age or initial hospital discharge, whichever occurs first.

This is measured as the head circumference-for-age Z-score at 36 weeks postmenstrual age or at initial hospital discharge, whichever occurs first. The Z-score is determined using Olsen percentile curves, and is derived from a normal distribution, where 0 designates average head circumference-for-age, and negative scores denote less than average head circumference-for-age.

The analysis population includes all randomized infants with available data at 36 weeks postmenstrual age or at initial hospital discharge, whichever occurs first.

This is measured as the average postmenstrual age (in weeks) at final tracheal extubation in infants who were intubated.

The analysis population includes all randomized infants with available data on trachael extubation.

This is measured as the average postmenstrual age (in weeks) at final caffeine dose in infants who received caffeine treatment.

The analysis population includes all randomized infants with available data on caffeine treatment.

This is measured as the length of stay (in days) up to initial hospital discharge or death, whichever occurred first.

The analysis population includes all randomized infants with available data up to the initial hospital discharge or death, whichever occurs first.

This is measured as the amount of days it took for full enteral feeding to occur.

The analysis population includes all randomized infants with available data on enteral feeding.

This is measured as Yes if Gross Motor Function Classification System (GMFCS) is levels IV or V; Otherwise, No. Higher values of the GMFCS are worse than lower values; a level of "I" denotes mild cerebral palsy (CP); level "II" or "III" moderate CP; level "IV" or "V" severe CP.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if experienced Hydrocephalus shunt by follow-up; Otherwise, No.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as a head circumference-for-age Z-score of less than -2; Otherwise, No. The Z-score is determined using WHO percentile curves, and is derived from a normal distribution, where 0 designates average head circumference-for-age, and negative scores denote less than average head circumference-for-age.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if experienced one or more seizures since discharge or of regular use of anticonvulsants or seizure medications; Otherwise, No.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if obtained Respiratory disease necessitating readmission before follow-up; Otherwise, No.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if scored less than 85 on the Bayley Scale of Infant and Toddler Development (BSID)-III composite language score; Otherwise, No. Higher scores indicate better performance. Composite BSID-III scores of less than 85 are less than 1 standard deviation below the mean of 100.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if scored less than 85 on the Bayley Scale of Infant and Toddler Development (BSID)-III composite motor score; Otherwise, No. Higher scores indicate better performance. Composite BSID-III scores of less than 85 are less than 1 standard deviation below the mean of 100.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if scored less than 85 on the Bayley Scale of Infant and Toddler Development (BSID)-III composite cognitive score; Otherwise, No. Higher scores indicate better performance. Composite BSID-III scores of less than 70 are less than 2 standard deviations below the mean of 100.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if scored less than 85 on the Bayley Scale of Infant and Toddler Development (BSID)-III composite language score; Otherwise, No. Higher scores indicate better performance. Composite BSID-III scores of less than 70 are less than 2 standard deviations below the mean of 100.

The analysis population includes all randomized infants with available data at the two-year followup.

This is measured as Yes if scored less than 85 on the Bayley Scale of Infant and Toddler Development (BSID)-III composite motor score; Otherwise, No. Higher scores indicate better performance. Composite BSID-III scores of less than 70 are less than 2 standard deviations below the mean of 100.

The analysis population includes all randomized infants with available data at the two-year followup.