Dash 2015

Methods	Open‐label randomised controlled trial conducted in a level III neonatal intensive care unit (NICU) of a private hospital in Mumbai, India. Study period: March 2012 to September 2013
Participants	Inclusion criteria: preterm infants with birth weight ≤1500 g and echocardiography performed within the first 48 h of life demonstrating PDA size ≥1.5 mm at the narrowest diameter, left to right shunt across the duct and ratio of the diameter of the left atrium to that of the aortic root (LA:AO) > 1.5:1. Exclusion criteria: inability to administer the study drug within 48 h of birth, structural duct‐dependent congenital heart disease, renal disease (such as multicystic dysplastic kidney and polycystic disease of kidney), dysmorphic features or congenital anomalies likely to affect life expectancy or neurologic development, maternal tocolytic therapy with indomethacin or another prostaglandin inhibitor within 72 h prior to giving birth, overt clinical bleeding at more than 1 site, platelet count < 50 × 10⁹/L, hydrops fetalis, and infant not considered viable.
Interventions	The paracetamol group (n = 38) received paracetamol drops through an infant feeding tube at a dose of 15 mg/kg/dose 4 times daily for 7 days (28 doses). The indomethacin group (n = 39) received IV indomethacin at a dose of 0.2 mg/kg/dose, diluted with normal saline to make 5 mL solution and infused over 20 mins by syringe pump once daily for 3 days. As per study protocol, 2 additional extra doses of indomethacin were allowed in the indomethacin group, if clinical evaluation after 3 doses showed persistence of PDA as demonstrated by clinical signs and symptoms such as tachycardia, wide pulse pressure and persistent murmur.
Outcomes	Primary outcome: PDA closure; (the PDA was considered to be closed if there was no evidence of any flow in the ductus arteriosus on echocardiographic and Doppler flow assessment). Secondary outcomes: death, renal impairment defined as presence of either oliguria (urine output of < 0.5 mL/kg/h) over a 6 h period or serum creatinine levels more than twice the age appropriate norms, gastro‐intestinal bleeding defined as the presence of blood‐stained or coffee ground brown gastric aspirates, necrotising enterocolitis (NEC) diagnosed as per modified Bell’s staging; pulmonary haemorrhage was diagnosed if a blood tinged tracheal aspirate was obtained; early‐ and late‐onset sepsis screen was defined as positive C‐reactive protein (CRP) before and after first 72 h of life (CRP > 6 mg/L), respectively, early‐onset sepsis defined as isolation of pathogenic organism from a blood culture collected in first 72 h of life, late‐onset sepsis defined as isolation of pathogenic organism from a blood culture collected after first 72 h of life, retinopathy of prematurity (ROP) classified as per the International classification of retinopathy — ROP needing either laser or anti‐VEGF (Avastin) therapy was labelled as severe ROP. Grading of intraventricular haemorrhage (IVH) was performed according to the Papile grading system, and features of periventricular leukomalacia (PVL) were assessed, as was requirement of supplemental oxygen at 28 days of postnatal age. Bronchopulmonary dysplasia (BPD)/chronic lung disease (CLD) was defined by the need for supplemental oxygen at 36 weeks' PMA.
Notes	We requested additional information from Dr. Kabra on 30 November 2017 and obtained information about IVH, PVL and NEC. Dr. Kabra confirmed that an abstract at a meeting in Vienna was a publication from the same study. We received the responses on 3 December 2017.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random allocation software was used by a statistician, who was not part of the study
Allocation concealment (selection bias)	Low risk	Sequentially numbered sealed opaque envelopes were used
Blinding of participants and personnel (performance bias) All outcomes	High risk	The study was open label and study drugs were given over different length of time. Paracetamol was given through a feeding tube and indomethacin was given IV.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcomes were not assessed blinded to treatment groups
Incomplete outcome data (attrition bias) All outcomes	Low risk	Outcomes were reported on all randomised infants. 2 deaths occurred in each group prior to follow‐up ECHO on day 7
Selective reporting (reporting bias)	Unclear risk	The study was registered — Trial Registration No: CTRI/2012/12/003163 — but this was a retrospective registration so we cannot ascertain if there were deviations from the original protocol or not.
Other bias	Low risk	Appears free of other bias