Chronic Lung Disease (CLD) of Prematurity is a common yet challenging co-morbidity affecting extremely premature newborns. Multifactorial influences leading to this co-morbidity is known and targeted in various research studies. Gastroesophageal reflux (GER) is common among the same cohort of patients. The investigators hypothesize that recurrent milk reflux into the airways of the premature babies worsen the inflammation of premature lungs and is a major contributor of CLD.

The investigators hypothesize that Continuous feeding (CF) minimises GER and micro-aspiration, thereby reducing the incidence and severity of CLD in high-risk infants.

Our aim is to compare the effect of intermittent bolus versus continuous intra-gastric feeding on the incidence and severity of CLD in very low birth weight infants ≤ 1250 grams.

The pathogenesis of bronchopulmonary dysplasia (BPD) is complex and multifactorial. As a result of premature birth, developmental arrest during a critical period of fetal lung development compounded by mechanical, oxidative and other injuries sustained during neonatal respiratory care forms the basis of pathogenesis. BPD affects up to 50% of infants with birth weight less than 1000 g. Between 2000 and 2009, despite advancement of neonatal care, annual BPD rates reported by Vermont Oxford Network among very low birth weight infants varied from 26.2% to 30.4% without any decline. Severely affected infants often require prolonged ventilation, high oxygen use, alternative airway and several potent medications over the first few months to years of their lives. High mortality rates, neurodevelopmental delay, respiratory morbidity and growth failure are associated with BPD.

Treatment of severe BPD with or without pulmonary hypertension is challenging. Prolonging the pregnancy in the face of premature labour, treating perinatal infections, augmenting pulmonary maturity with corticosteroids, judicious oxygen use, lung protective ventilation and optimizing nutrition to promote growth are important and well established measures to prevent or modify the progress of the chronic lung disease.

It is common to find infants with BPD also having significant symptoms of reflux. Gastroesophageal reflux (GER) is a well-known co-morbidity among preterms and ex-preterms on chronic ventilation, many of whom go on to require surgical fundoplication to stop the reflux thus preventing further lung damage. Some have reported dramatic respiratory improvement after resolution of GER. In the early days of a preterm baby with respiratory distress, GER is common and silent. Among infants, diagnosis of pathologic GER from a benign one is difficult. Many neonatal intensive care units (NICUs) would investigate for GER only when faced with moderate to severe BPD to achieve better respiratory symptom control. However GER has not been studied well as a factor precipitating the development of BPD among VLBW neonates. This is the focus of the study.

Aspiration of gastric contents into the lung is a widespread phenomenon in mechanically ventilated preterm infants. In animal models of gastric aspiration, gastric particulates altered the pulmonary mechanics, increased pulmonary inflammatory cells, released pro-inflammatory mediators, and inactivated surfactant. Development of bacterial pneumonia is a well-recognized complication following aspiration of gastric contents. The investigators hypothesize that repeated aspirations would aggravate and accelerate an inflammatory response in the lung finally leading on to BPD. In addition oxygen mediated damage and mechanical ventilation potentiate lung injury due to aspiration. Logically, if GER and aspiration could be minimized, it could decrease the incidence and severity of BPD.

Certain positioning of the baby, small volume of feed increment, keeping a close watch on feed tolerance are practical ways of improving feeding tolerance and reducing GER. The intermittent bolus intra-gastric feeding method is commonly used to feed premature babies. Other alternatives are continuous intra-gastric (feed volume is slowly infused in the stomach over couple of hours through the nasogastric tube) and continuous transpyloric feeding (feeding tube passes beyond the stomach to the duodenum and feed volume is slowly infused over hours). Transpyloric continuous feeding as compared to intermittent gastric bolus feeding, has been found to significantly reduce ventilatory support requirements in extremely low birth weight (ELBW) infants, possibly via its effect of minimising GER. In this study, none of the babies who received transpyloric feeding developed significant BPD and in addition babies with significant BPD improved after switching to transpyloric method. Transpyloric feeding tubes however are challenging to insert, and intestinal perforation is an uncommon but significant adverse effect. This feeding method is also not physiological as it bypasses the stomach. It remains to be seen if continuous gastric feeds, which is easily administered and safer, would yield some of the advantages of continuous transpyloric feeds over intermittent gastric feeding.

A Cochrane review in 2011 of continuous intra-gastric versus intermittent bolus intra-gastric feeding for premature infants found conflicting results, and was unable to make recommendations regarding the benefits and risks of these feeding methods. Clinical outcomes of interest from these trials were related to growth, feeding tolerance and gastrointestinal complications. The Cochrane review importantly found no significant difference in somatic growth and incidence of necrotising enterocolitis (NEC) between either feeding methods. Another Cochrane review in 2014 did not identify any randomised trial that evaluated the effects of continuous versus intermittent bolus intragastric tube feeding on gastro-oesophageal reflux disease in preterm and low birth weight infants and opined that well-designed and adequately powered trials are needed in this field. There were no studies comparing the effect of the above feeding methods on respiratory outcomes either.

Trial objectives

Aim: To compare the effect of intermittent bolus versus continuous intra-gastric feeding on the incidence and severity of BPD in very low birth weight infants (≤ 1250 grams).

Hypothesis: Continuous feeding (CF) minimises silent GER and micro-aspiration, thereby reducing the incidence and severity of bronchopulmonary dysplasia (BPD) in high-risk infants when compared to intermittent bolus feeding (BF).

Statistical considerations

Sample size calculation: based on 2015 data from the Singapore National Very-Low-Birth-Weight (VLBW) Infant Network for infants ≤ 1250 grams, mortality rate was 12.9% and BPD rate (defined as any oxygen supplementation or any respiratory support at 36 weeks post-conceptional age) was 29.4%. Thus the composite primary outcome rate was 42.3%. For a primary outcome rate reduction from 45% to 22.5%, with a type 1 error rate of 5% and a power of 80%, a sample size of 68 infants in each arm is required, giving a total sample size of 136 infants.

Inclusion Criteria:

• Infants with a birth weight <1250g and a gestational age of between 24+0 - 33+6 weeks

Exclusion Criteria:

1. Major congenital malformation
2. Chromosomal abnormality
3. 10-minute Apgar score of =3
4. Not expected to survive beyond 72 hours of age
5. Bilateral grade 4 intraventricular haemorrhage (IVH)
6. Did not consent / Consent not available

na. Open label