Background The management of a hemodynamically significant patent ductus arteriosus (hsPDA) associated with cardiomegaly and at least more than 1.5 mm in preterm infants is still a controversial topic. Previously, prophylaxis with ibuprofen was practiced in some units Nowadays, a conservative strategy with the use of diuretics and fluid restriction is considered a treatment option, alongside medical treatment with cyclo-oxygenase inhibitors (eg. Ibuprofen,indomethacin), paracetamol and surgery. The impact of the these practices have not been well studied.

Objectives We aim to determine the impact of hsPDA on the short term growth and cognitive outcomes of very low birth weight (VLBW) infants managed at a tertiary unit in Singapore from 2014 to 2018.

Methods A retrospective study on growth and cognitive outcomes was conducted on patients recruited from the outpatient follow-up clinic. Baseline neonatal demographics, treatment of hsPDA and short term neonatal morbidities were collected. Information on postnatal growth, health outcomes and cognitive testing scores were collected as well. Categorical data were compared with Chi squared test between those with medical or surgery treatment and those with conservative treatment.Continuous data with normal distribution was reported with means and Student t test was used for analysis. Those with skewed distribution was reported with medians and Mann-Whitney test was performed.

Results 89 children were recruited at a median age of 34.5 (IQR 18–52) months, with equal gender distribution. Interestingly, infants (n=61) who had hsPDA requiring treatment were more mature (29.9 vs 27.5 weeks) and heavier 1305 g vs 985 g (P < 0.01) at birth compared to VLBWs with conservative treatment (n=21). Despite their maturity and also larger weight, those who required treatment were more likely to have been mechanically ventilated 66.7% vs 33.3% (p < 0.01). As such, those who required PDA treatment had increased incidence of chronic lung disease (CLD) at 36 weeks (83.8% vs 60.8%%,p=0.04), increased length of stay (79.5 vs 59.0 days, P <0.01) and requirement of inhaled medications on follow-up. Growth failure of VLBWs with conservative treatment was evident until 6 months corrected age. Median cognitive scores were not statistically different, 97(IQR 85–105) vs 92 (IQR 81–101) when measured at ages 2–5 years. No differences were seen in the hearing loss or cerebral palsy rates. No association between PDA treatment and CLD were identified with a low cognitive score of less than 85.

Conclusions In our small cohort, the presence of a hemodynamically significant PDA requiring treatment was associated with a higher incidence of CLD needing long term medication but no impairment of cognition and growth at follow-up.