Sarna 1991.

Methods	Quasi‐randomised controlled trial January 1990 to July 1990.
Participants	Inclusion criteria: neonates with gastric bleeding Exclusion criteria: all neonates who were septicaemic, preterm, had prolonged prothrombin time or developed disseminated intravascular coagulation. Further information regarding participants: 48 term neonates mean gestational age: ranitidine group 38.3 ± 2.2 weeks versus control group 38.6 ± 2.2 weeks (P > 0.05). Mean birth weight: ranitidine group 2.4 ± 0.6 versus control group 2.6 ± 0.8 kg.
Interventions	Intervention (n = 20): ranitidine: 0.6 mg/kg loading dose followed by 0.15 mg/kg/hour intravenous infusion until bleeding had ceased for 24 hours. Control (n = 18): supportive therapy.
Outcomes	Duration of gastric bleeding. Mortality.
Notes	Conflicts of interest: not reported. Authors did not believe there were any adverse reactions attributable to the intervention.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Reported: "The neonates with gastric bleeding were alternatively, divided into two groups, on the basis of odd and even admission numbers." Quasi‐random allocation.
Allocation concealment (selection bias)	High risk	The methods of allocation concealment were not specified and the study was quasi‐random.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Blinding not reported.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinding of assessors not reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	10 (4 treatment group, 6 control) of 48 infants (21%) not reported for other clinical outcomes who died. 3 infants died after intracranial haemorrhage; 3 from hypoxic‐ischaemic encephalopathy; and 4 from respiratory distress syndrome.
Selective reporting (reporting bias)	Unclear risk	Protocol not available.
Other bias	Low risk	Infant characteristics were similar between intervention and control groups.