NIHR Signal Preterm babies do not benefit from a new IV feeding regimen

Published on 14 July 2016

Intravenous feeding solutions with a higher dose of amino acids or solutions containing fish oils do not provide better outcomes for preterm infants.

Many preterm babies can’t feed normally for the first few weeks after birth so they are given a mixture of protein, carbohydrate and lipid (oil or fat) intravenously. There is wide variation across the UK in terms of the type, composition and regimen of intravenous feeds.

This NIHR-funded NEON trial looked at infants born before 31 weeks. It tested the theory that increasing the amount of amino acids provided in IV feeding fluids may improve lean body mass and that giving more fish oils rather than plant oils may reduce liver problems.  The trial findings provide no evidence that higher intakes of amino acids or emulsions containing fish oils are beneficial compared to a standard approach, where the dose of amino acids is slowly increased and soya oil is used.

Preterm babies do not benefit from a new IV feeding regimen

Share your views on the research.

Why was this study needed?

Each year about 8000 extremely preterm infants, born before 31 weeks of gestation, are born in the UK.

Until they are able to tolerate milk feeds most of these babies receive intravenous feeding which starts at a variable time after birth and is usually increased slowly, though this differs across the country. Experts have called for standardisation of formulas and regimens but there is limited available evidence to help choose which is best.

This trial assessed whether the immediate introduction of the recommended daily intake of amino acids or a new lipid emulsion improved neonatal outcomes. The lipid emulsion had a lower ratio of n-6 to n-3 fatty acids and was composed of fish oil, olive oil, soya bean oil and medium-chain triglycerides.

What did this study do?

The randomised double-blind controlled trial compared different compositions of parenteral feeding of preterm infants, born before 31 weeks who were free from disease or abnormality.

Within 24 hours of birth 168 infants were randomly allocated to either the immediate introduction of the recommended daily intake of amino acids or the commonly used approach of low dose amino acids, increasing incrementally. The effect on lean body mass was measured in 133 infants at term-equivalent age. The researchers also compared adding fish oils, against the usual soya oil based emulsion, on lipid stored in the liver.

This was a well-designed trial, however as it was part of the Efficacy and Mechanism Evaluation Programme, it was focussed on a narrow range of relatively short term outcomes and looked at a relatively small number of infants from one region, which may reduce the reliability of these findings.

What did it find?

The trial evaluated the outcomes of 133 infants born before 31 weeks.

  • There was no difference between the groups in the median time it took until they were able to drink 150ml/kg of milk per day for 24 hours which was around 12 days.
  • No significant differences in lean mass were seen between the ‘recommended’ and incremental amino acid groups (adjusted mean difference 1.0g, 95% confidence interval (CI) -108 to 111g).
  • Liver lipid levels (assessed by magnetic resonance spectroscopy) were similar between the two groups comparing lipid composition (adjusted mean ratio 1.1, 95% CI 0.8 to 1.6).
  • Infants receiving the recommended daily intake of amino acids were more likely to have elevated blood urea nitrogen levels:
    • 75% of the ‘recommended’ group vs. 49% of the incremental group had blood urea nitrogen levels of more than 7mmol/l.
    • 49% of the ‘recommended’ group vs. 18% of the incremental group had blood urea nitrogen levels of more than 10mmol/l.
  • Head circumference at term was slightly smaller in the ‘recommended’ group compared to the incremental amino acid group (mean difference ‑0.8cm, 95% CI ‑1.5 to ‑0.1cm).
  • There were no significant differences in any secondary measure, including body fat, liver function tests, weight, length and death rate over the period of the study.

What does current guidance say on this issue?

There is no current UK national guideline on this topic and there is considerable variation in practice. Some hospitals use commercially prepared formulas, others use solutions made up locally by the pharmacy. Surveys commissioned by the Department of Health found that most neonatal units start parenteral nutrition in the hours or days after birth using low levels of macronutrients. These are increased over three to four days, but often do not reach recommended levels.

What are the implications?

This trial does not provide any evidence that the current practice of an incremental approach to nutrient administration to preterm infants should be changed.  Additionally it does not support any alteration in the composition of the intravenous lipid component. There remains a need for long-term follow-up to determine if either approach is associated with any difference in body composition and neurodevelopment.

Citation and Funding

S Uthaya, X Liu, D Babalis et al. Nutritional Evaluation and Optimisation in Neonates (NEON) trial of amino acid regimen and intravenous lipid composition in preterm parenteral nutrition: a randomised double-blind controlled trial. Efficacy Mech Eval. 2016; 3(2). 

This project was funded by the National Institute for Health Research Efficacy and Mechanism Evaluation programme, a Medical Research Council and National Institute for Health Research partnership (project number 08/99/04).

Why was this study needed?

Each year about 8000 extremely preterm infants, born before 31 weeks of gestation, are born in the UK.

Until they are able to tolerate milk feeds most of these babies receive intravenous feeding which starts at a variable time after birth and is usually increased slowly, though this differs across the country. Experts have called for standardisation of formulas and regimens but there is limited available evidence to help choose which is best.

This trial assessed whether the immediate introduction of the recommended daily intake of amino acids or a new lipid emulsion improved neonatal outcomes. The lipid emulsion had a lower ratio of n-6 to n-3 fatty acids and was composed of fish oil, olive oil, soya bean oil and medium-chain triglycerides.

What did this study do?

The randomised double-blind controlled trial compared different compositions of parenteral feeding of preterm infants, born before 31 weeks who were free from disease or abnormality.

Within 24 hours of birth 168 infants were randomly allocated to either the immediate introduction of the recommended daily intake of amino acids or the commonly used approach of low dose amino acids, increasing incrementally. The effect on lean body mass was measured in 133 infants at term-equivalent age. The researchers also compared adding fish oils, against the usual soya oil based emulsion, on lipid stored in the liver.

This was a well-designed trial, however as it was part of the Efficacy and Mechanism Evaluation Programme, it was focussed on a narrow range of relatively short term outcomes and looked at a relatively small number of infants from one region, which may reduce the reliability of these findings.

What did it find?

The trial evaluated the outcomes of 133 infants born before 31 weeks.

  • There was no difference between the groups in the median time it took until they were able to drink 150ml/kg of milk per day for 24 hours which was around 12 days.
  • No significant differences in lean mass were seen between the ‘recommended’ and incremental amino acid groups (adjusted mean difference 1.0g, 95% confidence interval (CI) -108 to 111g).
  • Liver lipid levels (assessed by magnetic resonance spectroscopy) were similar between the two groups comparing lipid composition (adjusted mean ratio 1.1, 95% CI 0.8 to 1.6).
  • Infants receiving the recommended daily intake of amino acids were more likely to have elevated blood urea nitrogen levels:
    • 75% of the ‘recommended’ group vs. 49% of the incremental group had blood urea nitrogen levels of more than 7mmol/l.
    • 49% of the ‘recommended’ group vs. 18% of the incremental group had blood urea nitrogen levels of more than 10mmol/l.
  • Head circumference at term was slightly smaller in the ‘recommended’ group compared to the incremental amino acid group (mean difference ‑0.8cm, 95% CI ‑1.5 to ‑0.1cm).
  • There were no significant differences in any secondary measure, including body fat, liver function tests, weight, length and death rate over the period of the study.

What does current guidance say on this issue?

There is no current UK national guideline on this topic and there is considerable variation in practice. Some hospitals use commercially prepared formulas, others use solutions made up locally by the pharmacy. Surveys commissioned by the Department of Health found that most neonatal units start parenteral nutrition in the hours or days after birth using low levels of macronutrients. These are increased over three to four days, but often do not reach recommended levels.

What are the implications?

This trial does not provide any evidence that the current practice of an incremental approach to nutrient administration to preterm infants should be changed.  Additionally it does not support any alteration in the composition of the intravenous lipid component. There remains a need for long-term follow-up to determine if either approach is associated with any difference in body composition and neurodevelopment.

Citation and Funding

S Uthaya, X Liu, D Babalis et al. Nutritional Evaluation and Optimisation in Neonates (NEON) trial of amino acid regimen and intravenous lipid composition in preterm parenteral nutrition: a randomised double-blind controlled trial. Efficacy Mech Eval. 2016; 3(2). 

This project was funded by the National Institute for Health Research Efficacy and Mechanism Evaluation programme, a Medical Research Council and National Institute for Health Research partnership (project number 08/99/04).

Nutritional Evaluation and Optimisation in Neonates (NEON) trial of amino acid regimen and intravenous lipid composition in preterm parenteral nutrition: a randomised double-blind controlled trial

Published on 1 March 2016

Uthaya S, Liu X, Babalis D, Dore C, Warwick J, Bell J, Thomas L, Ashby D, Durighel G, Ederies A, Yanez-Lopez M, Modi N.

Efficacy and Mechanism Evaluation Volume 3 Issue 2 , 2016

Background Parenteral nutrition (PN) is central to the care of very immature infants. Early intakes of higher amounts of amino acids and the use of lipid emulsions containing fish oils are recommended by current international recommendations. Objective To confirm the safety and demonstrate efficacy of the immediate introduction of the recommended daily intake of amino acids (Imm-RDI) and soya bean oil, medium-chain triglycerides, olive oil and fish oil lipid in PN to increase non-adipose (lean) body mass and decrease intrahepatocellular lipid (IHCL) content. Design Multicentre, double-blind, 2 × 2 factorial and randomised controlled trial (RCT). Setting Neonatal units in London and south-east England, UK. Participants Extremely preterm infants born before 31 weeks of gestation without major congenital or life-threatening abnormalities who could to be randomised to receive PN within 24 hours of birth. Interventions Infants were randomised within 24 hours of birth to receive PN containing either high [RDI of amino acids (Imm-RDI)] or low [incremental amino acids (Inc-AA) control] levels of amino acids. In addition, infants were randomised to receive either 20% SMOFlipid® (Fresenius Kabi AG, Richmond Hill, ON, Canada) or 20% Intralipid® (Fresenius Kabi AG, Richmond Hill, ON, Canada) (control). This resulted in four groups: (1) Inc-AA/Intralipid, (2) Inc-AA/SMOFlipid, (3) Imm-RDI/Intralipid and (4) Imm-RDI/SMOFlipid. The intervention was continued until infants were receiving 150 ml/kg/day of enteral feeds for 24 hours. Primary outcome measure For the amino acid intervention, this was non-adipose or lean body mass measured by magnetic resonance imaging. For the lipid composition intervention, this was IHCL content as measured by hepatic magnetic resonance spectroscopy. Primary outcomes were measured at term age equivalent, between 37 and 44 weeks postmenstrual age. Results We randomised 168 infants born before 31 weeks of gestation. We evaluated outcomes, at term, in 133 infants. There were no significant differences in non-adipose mass between the Imm-RDI and Inc-AA groups [adjusted mean difference 1.0 g, 95% confidence interval (CI) –108 to 111 g] or in levels of IHCLs between the SMOFlipid and Intralipid groups (adjusted mean SMOFlipid to Intralipid ratio 1.1, 95% CI 0.8 to 1.6). Infants receiving the Imm-RDI were more likely than Inc-AA infants to have blood urea nitrogen levels > 7 mmol/l [75% vs. 49% (p < 0.01)] and > 10 mmol/l [49% vs. 18% (p < 0.01)]. Furthermore, head circumference at term was smaller in the Imm-RDI group (mean difference –0.8 cm, 95% CI –1.5 to –0.1 cm; p = 0.02). There were no significant differences in any prespecified secondary outcomes, including adiposity, liver function tests, weight, length and mortality. Limitations Not all eligible babies were available for recruitment, as pharmacy staff trained in clinical trial procedures were unavailable at weekends in three of the four centres. We were able to assess brain volumes in only one-third of participants, as imaging was carried out while the participants were sleeping naturally and we measured primary outcomes first and continued to brain imaging only if the infant remained asleep. Conclusions Immediate delivery of the recommended daily intake of parenteral amino acids does not benefit body composition or growth to term and may be harmful; SMOFlipid does not affect IHCL content. Future work The long-term functional outcomes of early administration of RDI of amino acids and the use of SMOFlipid, including neurodevelopment, body composition and metabolic health, should be evaluated. Funding This project was funded by the Efficacy and Mechanism Evaluation programme, a Medical Research Council and National Institute for Health Research partnership.

Expert commentary

This is a well-conducted study about two key components of parenteral nutrition (intravenous feeds) administered to extremely preterm neonates. Neither a higher amount of amino acid nor the choice of fat affected important outcomes. These results will help professionals decide what to put into neonatal parenteral nutrition. This study raises questions such as whether including more energy in parallel to the increase in amino acids would improve outcomes.

Dr Mark Turner, Senior Lecturer/Consultant in Neonatology, University of Liverpool/ Liverpool Women’s NHS FT; Director of Research, LWFT; Advisor on International Liaison, NIHR Children’s Theme and Speciality Research Group; Chair, European Network for Paediatric Research at the European Medicines Agency