Davis 2003.

Study characteristics
Methods	Double blinded RCT
Participants	302 infants enrolled in the USA in 1997/1998 ≤ 24 hours of age, with evidence of RDS, between 600 g and 1200 g birth weight, ≥ 24 weeks of gestation receiving supplemental oxygen via mechanical ventilation and exogenous surfactant.
Interventions	Experimental (n = 154): intratracheal r‐h CuZnSOD (5 mg/kg in 2 mL/kg saline) Doses occurred 0.5 to 4.0 hours after exogenous surfactant. Further doses every 48 hours until 28 days of age or until extubation. Control (n = 148): 2 mL/kg of saline intratracheally was administered after exogenous surfactant treatment. Intervention was administered in 2 divided doses with the infant placed in the lateral decubitus position in 30° Trendelenburg to enhance distribution. If the infant was subsequently re‐intubated, the study medication was continued every 48 hours through 28 days of age.
Outcomes	Mortality during the 1st month, development of BPD at 28 days of life, Edwards Score at 28 days of life, oxygen requirement at 36 weeks’ PMA, number of days in oxygen, number of days of respiratory support, number of days in the hospital, and short‐term, NEC, pneumonia, IVH, ROP, PVL, number of episodes of significant pulmonary illness requiring treatment with asthma medications such as bronchodilators and corticosteroids, number and type of doctor’s visits, emergency department visits, and hospital admissions, any abnormalities in growth parameters or physical examination, economic evaluation (cost‐effectiveness, cost comparison).
Funding sources / declarations of interest	This study was supported by funds from Bio‐Technology General Corporation. The company also provided funds to some authors for preparation of materials for submission to the Food and Drug Administration.
Notes	Quote: "Study was cut short by DSMC due to little possibility of significant efficacy of r‐h CuZnSOD w/respect to death or incidence of death or BPD at 28 days of life."
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomization process not described. No baseline imbalances.
Allocation concealment (selection bias)	Unclear risk	Not enough details provided in the text.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Described as double‐blind by study authors.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Personnel was blinded to intervention. Additionally, (quote:) "all radiographs and sonograms were evaluated by a single pediatric radiologist who was blinded to treatment assignment".
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Quote: "Health assessment and physical examination were performed on 209 surviving infants (80%) with complete data available on 189 infants; 65 infants were lost to follow up". Relevant rate of lost to follow up.
Selective reporting (reporting bias)	Unclear risk	Protocol not available.
Other bias	Low risk	None.