Very small babies appear not to be affected by the rate of increasing milk feeds

A large-scale trial has found that the speed of increasing milk feed volumes in low birth weight or very low gestational age babies who are on intravenous feeding does not influence outcomes. This NIHR-funded study randomised preterm (below 32 weeks) or very low birth weight (less than 1,500g) babies to receive either daily milk feed increases in increments of 30ml per kilogram of bodyweight or 18ml per kilogram of bodyweight.

After two years of follow up, there was no significant difference in survival without moderate or severe neurodevelopmental disability between the groups. The two groups also had similar rates of serious infection, necrotising enterocolitis (a bowel disorder), and death.

This offers reassurance that faster introduction of milk through the nasogastric tube does not cause harm from bowel problems and that slower rates do not cause more infection from a longer need of the intravenous feeding line.

Why was this study needed?

In the UK, one to two per cent of babies are born at a very low birth weight (less than 1,500g) or are very premature (less than 32 weeks).

These babies are fed increasing volumes of milk each day through a small tube inserted into the nose or mouth and into the stomach until they reach full feeding volumes. They receive intravenous nutrition until they can digest nutritional volumes of milk.

Milk-feeding strategies may affect health long term as well as infant survival, but we don’t currently know how quickly to increase milk volume, and clinical advice varies. Limited previous evidence suggested faster rates may increase the risk of the severe bowel disorder, necrotising enterocolitis, while slower rates might increase the risk of sepsis due to longer periods of intravenous feeding.

What did this study do?

The Speed of Increasing Milk Feeds Trial (SIFT) compared faster (30ml per kilogram of body weight) with slower (18ml per kilogram of body weight) daily increments in milk feeding volumes.

This was a parallel-group randomised controlled trial conducted at 55 hospitals in the UK. Between 2013 and 2015, 2,804 infants born before 32 weeks gestation, with a birth weight of less than 1,500g, or both, were enrolled. At age two years, the primary outcome was assessed in 1,224 (87.4%) assigned to the faster increment feed and 1,246 (88.7%) assigned to the slower increment.

Caregivers were able to stop or alter the increase in feeding volume where clinically indicated. The trial was unblinded for practical reasons, but outcomes were reviewed by a committee who were unaware of trial-group assignments.

What did it find?

  • At two years, survival without moderate or severe neurodevelopmental disability occurred in 802 of 1,224 infants (65%) assigned to the faster increment and 848 of 1,246 (68.1%) assigned to the slower increment (adjusted risk ratio [aRR] 0.96, 95% confidence interval [CI] 0.92 to 1.01).
  • Late onset sepsis occurred in 414 of 1,389 infants (29.8%) in the faster-increment group and 434 of 1,397 (31.1%) in the slower-increment group (aRR 0.96, 95% CI 0.86 to 1.07).
  • Necrotising enterocolitis occurred in 70 of 1,394 infants (5%) in the faster-increment group and 78 of 1,399 (5.6%) in the slower-increment group (aRR 0.88, 95% CI 0.68 to 1.16).
  • At two years, death had occurred in 68 of 1,224 infants (5.6%) in the faster-increment group and 77 of 1,246 (6.2%) in the slower-increment group.
  • Moderate or severe motor impairment occurred in 7.5% of the infants in the faster-increment group and 5.0% of those in the slower-increment group (aRR 1.48, 99% CI 1.02 to 2.14). There was no evidence of a significant between-group difference for other disabilities looked at.

What does current guidance say on this issue?

NICE does not provide specific guidelines on this, but NHS policy guidance for post-natal care of preterm babies states that enteral milk feeds should be increased by 30ml per kilogram of weight per day to a maximum of 150mls per kilogram per day on day four.

Decisions as to when to start, stop and alter the volume of milk feeding should be overseen by a paediatrician.

What are the implications?

These results provide evidence that either rate at which milk volume is increased can be used for this vulnerable group of babies.

The slightly increased motor impairment rate in the faster group was surprising and could possibly be down to chance, but this is a finding that could be incorporated into future trial design.

Citation and Funding

Dorling J, Abbott J, Berrington J et al. Controlled trial of two incremental milk-feeding rates in preterm infants. N Engl J Med. 2019;381:1434-43.


This project was funded by the NIHR Health Technology Assessment Programme (project number 11/01/25).



NICE. Postnatal care: breastfeeding and formula feeding. NICE Pathway. London. National Institute for Health and Care Excellence; 2019.

Senterre T. Practice of enteral nutrition in low birth weight and extremely low birth weight infants. World Rev Nutr Diet. 2014;110:201-14.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre



Expert commentary

Infants born very early or very small are at increased risk of developmental delay and many other health problems compared to full-term, normal-weight infants. Many activities including feeding schedule, which would not generally affect a term infant, could lead to problems in vulnerable infants.

The Speed of Increasing Milk Feeds Trial (SIFT) found no difference in important health outcomes when milk volume was increased slowly compared to quickly. Parents and clinicians can feel confident that feeding volume increases from four days of age can be progressed according to infant tolerance and need and without harm.

Dr Diane Farrar, NIHR Post-Doctoral Research Fellow and Maternal Health Research Programme Manager, Bradford Institute for Health Research, Bradford Teaching Hospitals

The commentator declares no conflicting interests


Expert commentary

Although the SIFT trial was not powered to look at the outcomes of severe necrotising enterocolitis and sepsis, the former a serious inflammatory condition of the bowel thought to be associated with rapid increments of enteral feeds (using a nasogastric tube) and the latter associated with slower increments of enteral feeding because of the need to provide nutrition intravenously as parenteral nutrition, it provides reassurance to clinicians of a more progressive enteral feeding regimen.

However, the median time of advancement of feeds or commencement of the intervention was four days, which some might consider late, and is perhaps reflective of the ongoing concern among clinicians of more rapid advancement of enteral feeds in the immediate few days after birth.

As the authors conclude, more research is required into the type of feed (breast milk versus formula milk) and the potential interaction between the type of feed and speed of advancement on long-term outcomes.

Dr Sabita Uthaya, Clinical Senior Lecturer and Consultant in Neonatal Medicine, Chelsea and Westminster Hospital, Imperial College London

The commentator declares no conflicting interests


Expert commentary

Advancement to full milk feeds is potentially advantageous to the preterm baby and reduces the need for intravenous feeding, with its risk of sepsis. However, increasing milk feed volumes too rapidly can cause injury to the gut (necrotising enterocolitis).

Overall, this study demonstrated that more rapid establishment of full milk feeds and reduced duration of dependence on intravenous nutrition was not associated with differences in 2-year survival without moderate or severe disability, or infection and necrotising enterocolitis.

Clinical practice does not need to change but more research in particular subgroups of babies or types of milk might be helpful.

Dr Jane Hawdon, Consultant Neonatologist, Royal Free London NHS Foundation Trust

The commentator declares no conflicting interests