Bookshelf NBK350515

EFFICACY AND MECHANISM EVALUATION
VOLUME 3 ISSUE 2 MARCH 2016

ISSN 2050-4365
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
Sabita Uthaya, Xinxue Liu, Daphne Babalis, Caroline Dore, Jane Warwick,
Jimmy Bell, Louise Thomas, Deborah Ashby, Giuliana Durighel,
Ash Ederies, Monica Yanez-Lopez and Neena Modi
DOI 10.3310/eme03020
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
Sabita Uthaya,1,2* Xinxue Liu,3 Daphne Babalis,3,4

Caroline Dore,5 Jane Warwick,3 Jimmy Bell,6
Louise Thomas,6 Deborah Ashby,2 Giuliana Durighel,6
Ash Ederies,7 Monica Yanez-Lopez4 and Neena Modi1,2
1Chelsea and Westminster NHS Foundation Trust, London, UK
2Department of Medicine, Section of Infectious Diseases, Imperial College
London, London, UK
3Imperial Clinical Trials Unit, School of Public Health, Imperial College London,
London, UK
4Clinical Trials and Evaluation Unit, Royal Brompton and Harefield NHS
Foundation Trust, London, UK

5University College London Comprehensive Clinical Trials Unit, University College
London, London, UK
6Metabolic and Molecular Imaging Research Group, Medical Research Council
Clinical Science Centre, Imperial College London, London, UK

7Institute of Clinical Sciences, Imperial College London and Medical Research
Council Clinical Sciences Centre, Hammersmith Hospital, London, UK
*Corresponding author
Declared competing interests of authors: Sabita Uthaya is currently in the process of applying for a
patent for the trial parenteral nutrition formulations. Jane Warwick has had personal fees paid for
consultancy work by Novo Nordisk (Bagsværd, Denmark).
Published March 2016

DOI: 10.3310/eme03020
This report should be referenced as follows:
Uthaya S, Liu X, Babalis D, Dore C, Warwick J, Bell J, 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).
Efficacy and Mechanism Evaluation
ISSN 2050-4365 (Print)
ISSN 2050-4373 (Online)
This journal is a member of and subscribes to the principles of the Committee on Publication Ethics (COPE) (www.publicationethics.org/).
Editorial contact: nihredit@southampton.ac.uk
The full EME archive is freely available to view online at www.journalslibrary.nihr.ac.uk/eme. Print-on-demand copies can be purchased from
the report pages of the NIHR Journals Library website: www.journalslibrary.nihr.ac.uk
Criteria for inclusion in the Efficacy and Mechanism Evaluation journal

Reports are published in Efficacy and Mechanism Evaluation (EME) if (1) they have resulted from work for the EME programme, and
(2) they are of a sufficiently high scientific quality as assessed by the reviewers and editors.
EME programme
The Efficacy and Mechanism Evaluation (EME) programme was set up in 2008 as part of the National Institute for Health Research (NIHR)
and the Medical Research Council (MRC) coordinated strategy for clinical trials. The EME programme is broadly aimed at supporting ‘science
driven’ studies with an expectation of substantial health gain and aims to support excellent clinical science with an ultimate view to improving
health or patient care.
Its remit includes evaluations of new treatments, including therapeutics (small molecule and biologic), psychological interventions, public
health, diagnostics and medical devices. Treatments or interventions intended to prevent disease are also included.
The EME programme supports laboratory based or similar studies that are embedded within the main study if relevant to the remit of the EME
programme. Studies that use validated surrogate markers as indicators of health outcome are also considered.
For more information about the EME programme please visit the website: http://www.nets.nihr.ac.uk/programmes/eme
This report
The research reported in this issue of the journal was funded by the EME programme as project number 08/99/04. The contractual start date
was in February 2010. The final report began editorial review in January 2015 and was accepted for publication in October 2015. The authors
have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The EME editors and production
house have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the
final report document. However, they do not accept liability for damages or losses arising from material published in this report.
This report presents independent research. The views and opinions expressed by authors in this publication are those of the authors and do
not necessarily reflect those of the NHS, the NIHR, the MRC, NETSCC, the EME programme or the Department of Health. If there are verbatim
quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not
necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the EME programme or the Department of Health.
© Queen’s Printer and Controller of HMSO 2016. This work was produced by Uthaya et al. under the terms of a commissioning
contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and
study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement
is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be
addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre,
Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.
Published by the NIHR Journals Library (www.journalslibrary.nihr.ac.uk), produced by Prepress Projects Ltd, Perth, Scotland
(www.prepress-projects.co.uk).
Efficacy and Mechanism Evaluation Editor-in-Chief
Professor Raj Thakker May Professor of Medicine, Nuffield Department of Medicine, University of Oxford, UK
NIHR Journals Library Editor-in-Chief
Professor Tom Walley Director, NIHR Evaluation, Trials and Studies and Director of the HTA Programme, UK
NIHR Journals Library Editors
Professor Ken Stein Chair of HTA Editorial Board and Professor of Public Health, University of Exeter Medical
School, UK
Professor Andree Le May Chair of NIHR Journals Library Editorial Group (EME, HS&DR, PGfAR, PHR journals)
Dr Martin Ashton-Key Consultant in Public Health Medicine/Consultant Advisor, NETSCC, UK
Professor Matthias Beck Chair in Public Sector Management and Subject Leader (Management Group),
Queen’s University Management School, Queen’s University Belfast, UK
Professor Aileen Clarke Professor of Public Health and Health Services Research, Warwick Medical School,
University of Warwick, UK
Dr Tessa Crilly Director, Crystal Blue Consulting Ltd, UK
Dr Peter Davidson Director of NETSCC, HTA, UK
Ms Tara Lamont Scientific Advisor, NETSCC, UK
Professor Elaine McColl Director, Newcastle Clinical Trials Unit, Institute of Health and Society,
Newcastle University, UK
Professor William McGuire Professor of Child Health, Hull York Medical School, University of York, UK
Professor Geoffrey Meads Professor of Health Sciences Research, Health and Wellbeing Research and
Development Group, University of Winchester, UK
Professor John Norrie Health Services Research Unit, University of Aberdeen, UK
Professor John Powell Consultant Clinical Adviser, National Institute for Health and Care Excellence (NICE), UK
Professor James Raftery Professor of Health Technology Assessment, Wessex Institute, Faculty of Medicine,
University of Southampton, UK
Dr Rob Riemsma Reviews Manager, Kleijnen Systematic Reviews Ltd, UK
Professor Helen Roberts Professor of Child Health Research, UCL Institute of Child Health, UK
Professor Jonathan Ross Professor of Sexual Health and HIV, University Hospital Birmingham, UK
Professor Helen Snooks Professor of Health Services Research, Institute of Life Science, College of Medicine,
Swansea University, UK
Professor Jim Thornton Professor of Obstetrics and Gynaecology, Faculty of Medicine and Health Sciences,
University of Nottingham, UK
Please visit the website for a list of members of the NIHR Journals Library Board:
www.journalslibrary.nihr.ac.uk/about/editors
Editorial contact: nihredit@southampton.ac.uk
NIHR Journals Library www.journalslibrary.nihr.ac.uk

DOI: 10.3310/eme03020 EFFICACY AND MECHANISM EVALUATION 2016 VOL. 3 NO. 2
Abstract
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
Sabita Uthaya,1,2* Xinxue Liu,3 Daphne Babalis,3,4 Caroline Dore,5

Jane Warwick,3 Jimmy Bell,6 Louise Thomas,6 Deborah Ashby,2
Giuliana Durighel,6 Ash Ederies,7 Monica Yanez-Lopez4
and Neena Modi1,2
1Chelsea and Westminster NHS Foundation Trust, London, UK
2Department of Medicine, Section of Infectious Diseases, Imperial College London, London, UK
3Imperial Clinical Trials Unit, School of Public Health, Imperial College London, London, UK
4Clinical Trials and Evaluation Unit, Royal Brompton and Harefield NHS Foundation Trust,
London, UK
5University College London Comprehensive Clinical Trials Unit, University College London,
London, UK
6Metabolic and Molecular Imaging Research Group, Medical Research Council Clinical Science
Centre, Imperial College London, London, UK

7Institute of Clinical Sciences, Imperial College London and Medical Research Council Clinical
Sciences Centre, Hammersmith Hospital, London, UK
*Corresponding author s.uthaya@imperial.ac.uk
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.
© Queen’s Printer and Controller of HMSO 2016. This work was produced by Uthaya et al. under the terms of a commissioning contract issued by the Secretary of State for
Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals
provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be
v
addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science
Park, Southampton SO16 7NS, UK.
ABSTRACT
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.
Trial registration: Current Controlled Trials ISRCTN29665319 and EudraCT 2009-016731-34.
Funding: This project was funded by the Efficacy and Mechanism Evaluation programme, a Medical
Research Council and National Institute for Health Research partnership.
vi
Contents
List of tables ix
List of figures xi
List of abbreviations xiii
Plain English summary xv
Scientific summary xvii
Chapter 1 Introduction 1
Background 1
Preterm infants 1
Rationale for trial 1
Nutritional requirements of preterm babies 2
Parenteral nutrition 3
Previous studies of parenteral nutrition 3
Risks and benefits of parenteral nutrition 4
Need for the Nutritional Evaluation and Optimisation in Neonates trial 5
Chapter 2 Research objectives 7

Amino acid intervention 7
Lipid intervention 7
Chapter 3 Methods 9
Trial design 9
Participants 9
Inclusion criteria 9
Exclusion criteria 9
Interventions 9
Outcomes 11
Primary outcomes 11
Secondary outcomes 11
Data collection 11
Electronic case record form 11
Timescale of trial evaluations 11
Schedule of investigations 12
Clinical investigations 13
Anthropometry 13
Blood pressure measurements 13
Magnetic resonance imaging 13
Quantitative insulin sensitivity check index 15
Pharmacovigilance definitions and procedures 15
Serious adverse events 15
Expectedness and causality of serious adverse events 16
Reporting of adverse events 16
Adverse events 16
Statistical considerations 17
vii
CONTENTS
Sample size 17
Randomisation 18
Blinding 18
Statistical methods 18
Missing data 19
Statistical analysis plan 19
Trial organisation 19
Trial management 19
Trial sponsor 19
Ethical considerations 19
Research governance 20
Regulatory requirements 20
Trial registration 20
National Institute for Health Research Clinical Research Network portfolio 20
Summary of protocol amendments 20
Trial committees 21
Data management 22
Risk assessment and monitoring plan 22
Monitoring visits 22
Investigational medicinal product manufacturer 22
Patient and public involvement 22
Chapter 4 Results 23
Participant flow 23
Screening 23
Recruitment and retention 24
Recruitment rate 24
Baseline data 24
Chapter 5 Discussion 59
Chapter 6 Conclusions and recommendations 63

Health-care recommendations 63
Research recommendations 63
Acknowledgements 65
References 67
Appendix 1 Distribution of primary and secondary outcomes after transformation 73
Appendix 2 Parent information sheet, magnetic resonance information sheet

and consent form 75
viii
List of tables
TABLE 1 Summary of interventions 10
TABLE 2 Summary of tests and investigations 12
TABLE 3 Definitions for assessment of causality 16
TABLE 4 Definitions of SpAEs including thresholds for reporting to the DMEC 17
TABLE 5 Baseline characteristics for all infants randomised 28
TABLE 6 Baseline characteristics for all infants completing MRI assessment 29
TABLE 7 Parenteral nutrition details and blood culture results for all
infants randomised 30
TABLE 8 Parenteral nutrition details and blood culture results for all infants
completing MRI assessment 31
TABLE 9 Trial PN intake during the first 7 days for all infants randomised 32
TABLE 10 Trial PN intake during the first 7 days for all infants completing
MRI assessment 33
TABLE 11 Total nutrition intake during the first 7 days, 3 weeks, 4 weeks and
by 34 weeks of gestational age for all infants randomised 35
TABLE 12 Nutritional intake over the first 2 weeks for all infants randomised 37
TABLE 13 Nutritional intake during the study period for all infants randomised 38
TABLE 14 Nutritional intake over first 2 weeks for all infants completing
MRI assessment 40
TABLE 15 Nutritional intake during the study period for all infants completing
MRI assessment 41
TABLE 16 Safety data: summary of laboratory AEs by treatment for all

infants randomised 44
TABLE 17 Safety data: summary of laboratory AEs by treatment for all infants
completing MRI assessment 45
TABLE 18 Safety data: summary of SAEs by treatment 47
TABLE 19 Baseline characteristics and trial outcomes (all infants completing

primary outcome assessments) 48
ix
LIST OF TABLES
TABLE 20 Adipose tissue compartments in litres for all infants completing

MRI assessment 51
TABLE 21 Summary of the covariates for secondary analysis 52
TABLE 22 Primary outcomes (pairwise comparisons) 53
x
List of figures
FIGURE 1 Classification of AT depots 14
FIGURE 2 The Consolidated Standards of Reporting Trials diagram 23
FIGURE 3 Summary of screening data for all trial sites 24
FIGURE 4 Cumulative recruitment vs. target recruitment for the duration

of the trial 25
FIGURE 5 Cumulative retention (number of MR scans) vs. target retention for the
duration of the trial 26
FIGURE 6 Total recruitment and retention per centre 27
FIGURE 7 Means (95% CIs) of Inc-AA and Imm-RDI in two lipid subgroups for
(a) non-adipose body mass; and (b) IHCL content on a log-scale 47
FIGURE 8 Time trend for babies’ weights across four groups 54
FIGURE 9 Daily protein intake from all sources in the first 2 weeks across four groups 54
FIGURE 10 Daily carbohydrate intake from all sources in the first 2 weeks across
all four groups 55
FIGURE 11 Daily fat intake from all sources in the first 2 weeks across all four groups 55
FIGURE 12 Daily energy intake from all sources in the first 2 postnatal weeks 56
FIGURE 13 Daily protein intake from all sources after first 2 weeks across all
four groups 56
FIGURE 14 Daily carbohydrate intake from all sources after first 2 weeks across
all four groups 57
FIGURE 15 Daily fat intake from all sources after first 2 weeks across four groups 57
FIGURE 16 Daily energy intake from all sources after the first 2 postnatal weeks 58
FIGURE 17 Distribution of primary and secondary outcomes after transformation 73
xi
List of abbreviations
AE adverse event LCT long-chain triglyceride
AT adipose tissue MCT medium-chain triglyceride
CI confidence interval MHRA Medicines and Healthcare products
Regulatory Agency
DMEC Data Monitoring and Ethics
Committee MR magnetic resonance
eCRF electronic case record form MRI magnetic resonance imaging
EME Efficacy and Mechanism Evaluation MRS magnetic resonance spectroscopy
FOV field of view NEON Nutritional Evaluation and
Optimisation in Neonates
ICTU Imperial Clinical Trials Unit
NICU neonatal intensive care unit
IHCL intrahepatocellular lipid
NIHR National Institute for Health
Imm-RDI immediate introduction of the
Research
recommended daily intake of
amino acids PIS parent information sheet
Imm-RDI/ immediate introduction of the PN parenteral nutrition
Intralipid recommended daily intake of
QUICKI quantitative insulin sensitivity
amino acids and 20% Intralipid
check index
Imm-RDI/ immediate introduction of the RCT randomised controlled trial
SMOFlipid recommended daily intake of
amino acids and 20% SMOFlipid RDI recommended daily intake
IMP investigational medicinal product SAE serious adverse event
Inc-AA incremental amino acids SD standard deviation
Inc-AA/ incremental amino acids and SMOFlipid a mixture of soya bean oil, MCTs,
Intralipid 20% Intralipid olive oil and fish oils, supplemented
with vitamin E
Inc-AA/ incremental amino acids and
SMOFlipid 20% SMOFlipid SpAE specific adverse event
SSCNAAT superficial subcutaneous
IQR interquartile range
non-abdominal adipose tissue
IVRS interactive voice recognition system
TSC Trial Steering Committee
LBM lean body mass
xiii
Plain English summary

I nfants born extremely preterm (defined as born before 31 weeks of gestation) spend several weeks and
months in intensive care and are subject to various complications relating to their prematurity. Outside
the womb, meeting the nutritional demands of these infants presents challenges. Experts have called
for changes in how we feed babies, but we do not know if giving more nutrition is better for babies.
We studied two aspects of parenteral nutrition, a fluid used to feed babies through their veins to
overcome gut immaturity. First, we compared a stepwise increase in protein intake with giving the baby
what is the recommended daily intake. Second, we compared the type of fat currently used in parenteral
nutrition with a newer combination of fat that has been shown to be less harmful to the liver. Babies were
allocated to one or the other group by chance. This is so that both groups are similar at the start of the
study so that any difference that is found at the end can be explained by the difference in the nutrition we
gave them. Using special magnetic scans (to measure body muscle mass and fat in the liver) we studied
the babies around the time of their due date. We found that, provided extremely preterm babies were fed
milk from the start, giving the recommended daily intake of protein from birth instead of gradually
increasing the intake did not result in any difference in muscle mass at term age. In addition, the new type
of fat did not show any benefit over the old type of fat.
xv
Scientific summary
Background
Delivering nutrition to very immature babies is challenging. Parenteral nutrition (PN) requires reliable
intravenous access, pharmacist support and clinical expertise in minimising and treating complications.
Gastrointestinal immaturity precludes early administration of milk volumes sufficient to support growth.
In practice, PN and milk feeds are commenced at variable intervals after birth, with nutrient delivery
increased incrementally. As a consequence, cumulative nutrient deficits are common and, by term,
the majority of very preterm infants are lighter and shorter than healthy term-born counterparts. Although
optimal postnatal growth velocity is uncertain, the association between slower growth and greater
likelihood of neurodevelopmental impairment and cerebral palsy has provided justification for early PN
provision. High amino acid intakes have been advocated, with the recommended daily intake (RDI) calculated
on the basis of redressing cumulative deficits as well as matching intrauterine growth velocity. Intravenous
lipid preparations containing fish oils have been recommended on the basis of clinical observations
suggesting that they may be protective against hepatic dysfunction, a frequent concomitant of PN.
A diet with a low protein-to-energy ratio results in lower lean body mass and greater adiposity. Thus,
in the short term, weight gain, though a widely used outcome measure, may not be as revealing as body
composition. Monitoring lipid tolerance is problematic, as normative ranges for circulating lipids remain
inadequately defined in very preterm babies and relationships to long-term outcomes are unclear.
Whole-body magnetic resonance imaging (MRI) can be employed to assess body composition directly and
in vivo magnetic resonance spectroscopy (MRS) to non-invasively assess hepatic lipid; the latter compares
favourably with the gold standard, liver biopsy, for the quantitative assessment of hepatic steatosis.
We designed a clinical trial to test the hypotheses that the immediate delivery of the RDI of parenteral
amino acids compared with incremental provision is more efficacious in increasing lean (non-adipose) body
mass at term, and a mechanism of action of 20% soya bean oil, medium-chain triglycerides, olive oil, fish
oil lipid (SMOFlipid®; Fresenius Kabi AG, Richmond Hill, ON, Canada) compared with 20% Intralipid®
(Fresenius Kabi AG, Richmond Hill, ON, Canada) is to reduce intrahepatocellular lipid (IHCL) content.
Objectives
Amino acid intervention

To evaluate whether or not immediate rather than incremental introduction of the RDI of amino acids
(Imm-RDI) in extremely preterm infants results in:
l greater accrual of non-adipose (lean) body mass at term (primary objective)

l increased brain volume at term (secondary objective)
l reduced insulin resistance at term (secondary objective)
l reduced ratio of internal to subcutaneous adipose tissue (AT) at term (secondary objective)
l a lower drop in weight standard deviation (SD) score between birth and term equivalent
(secondary objective).
xvii
SCIENTIFIC SUMMARY
Lipid intervention
To evaluate whether or not 20% SMOFlipid (with a lower ratio of n-6 to n-3 fatty acids) compared with
20% Intralipid in extremely preterm infants results in:
l reduced IHCL content at term age equivalent (primary objective)

l reduced incidence of hypertriglyceridaemia and hyperbilirubinaemia (secondary objective).
Methods
Trial design
This was a multicentre, randomised, 2 × 2 factorial and double-blind controlled trial in four UK centres,
in London and south-east England. Eligible preterm infants were randomised, within 24 hours of birth, to
receive (1) either incremental amino acids (Inc-AA) in PN or the RDI of amino acids (Imm-RDI) from day 1;
and (2) either 20% Intralipid or 20% SMOFlipid.
There were four randomised groups:
1. Inc-AA and 20% Intralipid (Inc-AA/Intralipid)

2. Inc-AA and 20% SMOFlipid (Inc-AA/SMOFlipid)
3. Imm-AA and 20% Intralipid (Imm-RDI/Intralipid)
4. Imm-AA and 20% SMOFlipid (Imm-RDI/SMOFlipid).
Participants
Preterm infants (born before 31 weeks of gestation) requiring nutritional support in the form of PN.
Inclusion criteria
l Preterm infants born before 31 weeks of gestation (defined as ≤ 30 weeks and 6 days).
l Written informed consent from parents.
Exclusion criteria
l Major congenital or life-threatening abnormalities.

l Inability to randomise in time to allow administration of trial PN within 24 hours of birth.
Interventions
There were two interventions: (1) the amount of amino acids in PN and (2) the type of lipid formulation.
All other components of PN were consistent across the four treatment groups. The intervention was
commenced within 24 hours of birth. Nutritional intake, both parenteral and enteral, was guided by
prespecified protocols that were provided in an investigator’s manual. In the control arm of amino acid intake,
infants received 1.7 g/kg/day amino acids on day 1 of postnatal life. This increased to 2.1 g/kg/day on day 2
and a maximum of 2.7 g/kg/day from day 3. In the intervention group, infants received 3.6 g/kg/day from
day 1. On days 1 and 2, PN was provided in an aqueous form at a concentration of 90 ml/kg/day increasing
to 120 ml/kg/day from day 3 onwards. Carbohydrate intake was 8.6 g/kg/day from day 1. Lipid intake was
2 g/kg/day on day 1 increasing to 3 g/kg/day from day 2 onwards. Infants were also randomised to receive
lipid as either 20% Intralipid or 20% SMOFlipid. Day 1 was defined as the duration between birth and when
the first bag of PN was changed. Bag changes occurred daily at 17.00. PN was dispensed only between 09.00
and 17.00. The duration of day 1 was variable and dependent on infant time of birth. Subsequently, all
infants received the intended volumes as described above.
xviii
The interventions ceased once the infant was established on milk feeds of 150 ml/kg/day for at least
24 hours. If the infant was subsequently placed nil by mouth after this point, PN was prescribed in
accordance with local practice as determined by the supervising clinician.
Outcomes
Primary outcomes
Efficacy of the early introduction of the RDI of amino acids was assessed by whole-body MRI to measure
non-adipose or lean mass. The efficacy of lipid composition was assessed by MRS to measure IHCL
content. These assessments were done at term age equivalent, between 37 and 44 weeks
postmenstrual age.
Measurement of lean body mass

Lean body mass was calculated by subtracting AT mass from the weight of the baby on the day of the scan.
Measurement of intrahepatocellular lipid content

Efficacy of SMOFlipid was assessed by liver MRS to measure IHCL content. This was done at term age
equivalent, between 37 and 44 weeks postmenstrual age.
Secondary outcomes
l Quantity and distribution of AT.

l Total and regional brain volumes.
l Metabolic index of insulin sensitivity [as measured by the quantitative insulin sensitivity check
index (QUICKI)].
l Serum lipids and bilirubin.
l Incidence of death.
l Anthropometry.
Sample size and statistical analysis

The sample size was based on the estimate that 64 infants in each pairwise group (Imm-RDI vs. Inc-AA)
would provide 80% power (two-sided; 5% significance) to detect a 200-g difference in non-adipose mass
assuming a SD of 400 g. This represents half the difference in non-adipose mass identified between very
preterm and term-born infants in a prior experimental cohort. We have previously reported IHCL values for
very preterm babies at term [mean lipid-to-water ratio 1.75 (SD 1.85), range 0.14–7.72]. As the distribution
is positively skewed, a loge-transformation was used to provide IHCL mean lipid-to-water ratio [0.121
(SD 1.052); range –1.97 to 2.04]. It was calculated that 64 infants in each pairwise group would provide
80% power (5% significance) to detect a difference in mean IHCL values of 0.53 on the logarithmic scale.
Back-transforming to the original scale of measurement, this is equivalent to a 40% decrease in IHCL
content in the intervention group. It was assumed there would be no interaction between the interventions.
Allowing for a 10% mortality and up to 10% dropout (including babies still in hospital at 44 weeks
postmenstrual age), the aim was to recruit 160 infants or until 64 infants in each pairwise group completed
primary outcome evaluations.
A modified intention-to-treat analysis was used, as it was anticipated that it would not be possible to
obtain primary outcome measures in all infants. For the amino acid and lipid interventions, a multiple
regression was used with non-adipose mass (g) or IHCL content (natural logarithmic scale) as the dependent
variable and amino acid group (Inc-AA or Imm-RDI), lipid group (Intralipid or SMOFlipid), stratifying variables
(gestational age, birthweight and centre), sex and age at assessment as the independent variables.
An interaction term was added to assess if the effect of amino acid regimen is influenced by lipid type.
In a planned secondary analysis, illness severity and nutritional intake was incorporated in the regression
models to investigate their role as potential effect modifiers. All analyses were performed using Stata 13
(StataCorp LP, College Station, TX, USA).
xix
SCIENTIFIC SUMMARY
Results
Of the 437 infants born before 31 weeks of gestation, 168 infants were randomised. A total of 133
infants were available for assessment of the primary outcome measures. Baseline characteristics of sex,
gestational age at birth, anthropometry, maternal demographics, mode of delivery, antenatal steroid use,
blood pressure on admission and time to commencing PN were similar across the four groups.
The median time to achieve a milk intake of 150 ml/kg/day for 24 hours for all infants randomised was
similar across the four groups {Inc-AA/Intralipid: 12 days [interquartile range (IQR) 9–17.5 days];
Inc-RDI/SMOFlipid: 11.5 days [IQR 9–16 days]; Imm-RDI: 11 days [IQR 10–14 days]; and Imm-RDI/SMOFlipid:
13 days [IQR 9.5–18 days]}. The median length of hospital stay for all infants randomised was similar across
the four groups [Inc-AA/Intralipid: 69.5 days (IQR 52–95 days); Inc-RDI/SMOFlipid: 61 days (IQR 45–88 days);
Imm-RDI: 63 days (IQR 45–95 days); and Imm-RDI/SMOFlipid: 66.5 days (IQR 44–98 days)].
Nutritional intake from trial PN during the first week was similar across the four groups, except in the intake
of protein. For ease of comparison between enteral and parenteral intakes, we express parenteral amino
acid intake as protein (1 g of amino acids ≡ 0.89 g of protein). Trial PN protein intake was higher in the
Imm-RDI arms, and carbohydrate and lipid intakes were similar across the four groups.
In relation to primary outcome measures, there were no significant differences in the quantity of non-AT mass
between the groups randomised to Inc-AA and those randomised to the Imm-RDI {adjusted mean difference
Imm-RDI, 1 g [95% confidence interval (CI) –108 to 111 g]; p = 0.98}. For the lipid composition intervention,
there was no significant difference in IHCL content between the groups randomised to receive 20% Intralipid
than for 20% SMOFlipid (adjusted mean ratio of lipid to water 1.1, 95% CI 0.8 to 1.6; p = 0.58).
There were no significant differences between the groups in the proportion of infants with abnormal
biochemical indices namely serum glucose, worst base deficit in the previous 24 hours, total serum
bilirubin, conjugated bilirubin, serum cholesterol, serum triglycerides, serum sodium, serum potassium,
serum phosphate, serum calcium, serum creatinine, and alanine transaminase. However, Imm-RDI infants
were more likely than Inc-AA infants to have blood urea nitrogen levels > 7 mmol/l [75% vs. 49%
(p < 0.01)] and and > 10 mmol/l [49% vs. 18% (p < 0.01)]. 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, at term age equivalent, in secondary outcome measures of the
quantity and distribution of AT, measure of insulin sensitivity (QUICKI), total cerebral volume, whole-brain
volume, weight and length.
Conclusions
We conclude that commencement within 24 hours of birth of an Inc-AA regimen providing a maximum of
2.7 g/kg/day together with the early introduction of milk feeds, compared with the immediate provision
of an amino acid intake of 3.6 g/kg/day, does not appear to be detrimental to body composition and may
be safer. In addition, SMOFlipid does not reduce IHCL accumulation.
Extremely preterm infants at term age equivalent, with the early provision of PN according to a
standardised regimen, can achieve the body composition nearer that of healthy term-born infants.
Before either of the interventions studied in this trial can be recommended as routine practice, long-term
follow-up of functional outcomes of neurodevelopment as well as long-term body composition and
metabolic health of both the trial interventions is essential.
xx
The results do not support the calls for more aggressive nutrition in the extremely preterm infant nor the
routine use of SMOFlipid as reflected in international consensus statements (higher amounts of amino
acids) or as is increasingly seen in current practice.
A key ancillary observation of this trial is that the use of standard PN regimens is feasible, is acceptable
to clinicians, even when blinded, can deliver desired nutritional intake without manipulation and is safe.
In our opinion, standardised regimens that have been tested in the context of a randomised controlled trial
should be adopted in routine clinical practice to reduce the clinical risk to infants from variation in practice.
We suggest that high amounts of amino acids be used only in the context of randomised clinical trials.
Optimal amino acid intakes and intravenous lipid formulations for extremely preterm infants remain to
be established.
Trial registration
This trial is registered as ISRCTN29665319 and EudraCT 2009–016731–34.
Funding
This project was funded by the Efficacy and Mechanism Evaluation programme, a Medical Research
Council and National Institute for Health Research partnership.
xxi
Chapter 1 Introduction
Background
Preterm infants
Extremely preterm infants, born before 31 weeks of gestation, account for 1–1.5% of deliveries in the UK.
Of around 70,000 infants born preterm in the UK each year, about 8000 are born before 31 weeks of
gestation. The UK has one of the highest rates of preterm birth in Europe, as well as one of the highest
rates of neonatal mortality. These infants spend a prolonged period in hospital and are subject to long
periods of poor nutrition. By the time preterm infants reach term age, the overwhelming majority exhibit
‘growth failure’ when compared with healthy term-born infants.1 Long-term follow-up studies show that
there appears to be catch-up growth in infancy and through adolescence.2 Although this may be reassuring,
catch-up growth is associated with adverse metabolic health and renal impairment.3,4 However, growth
failure is associated with neurodevelopmental impairment and cerebral palsy.5,6
Rationale for trial
Nutrition is a major factor influencing growth and possibly long-term metabolic health. Protein deficiency
and a high-fat, high-carbohydrate diet characterises preterm nutrition during this period regardless of
whether it is provided intravenously or enterally. A low-protein diet and low protein-to-energy ratio in
preterm infants results in a decrease in lean body mass (LBM) and increased deposition of adipose tissue
(AT).7 Thus, weight gain per se may not be as important as weight gain composition. In preterm infants,
a low-protein, high-carbohydrate diet has also been shown to be associated with insulin resistance in
adolescence.8 Preterm infants, at present, do not receive routine metabolic follow-up assessments and so
the exact burden of subsequent metabolic ill health cannot be quantified.9
There is good evidence that there are critical periods in development where nutrition has long-term effects
on later health. It has been shown that by the end of the first week of life, cumulative energy and protein
deficits in infants born before 30 weeks of gestation are 400 kcal/kg and 14 g/kg, respectively.10,11 Preterm
formulae and fortified maternal milk meet the recommended daily intake (RDI) of macronutrients, but
deficits accumulated in the period after birth combined with factors that increase requirements result in a
progressive deficit that is not made up or increases the magnitude of later catch-up growth.
Preterm infants have increased prepubertal insulin resistance compared with term-born infants.12
Compared with term-born infants, as adults they have higher blood pressure13,14 and are more likely to
have glucose intolerance,15 insulin resistance and dyslipidaemia.16 Insulin resistance in prepubertal children
born extremely preterm has been associated with neonatal nutrition. Preterm infants were found to be
insulin resistant compared with term-born infants. The diet of preterm infants was characterised as being
low in protein in the first month and high in fat subsequently. Those who gained most weight in infancy
were most insulin resistant and found to have a high carbohydrate intake in the first month of life.8
Another group has demonstrated that a period of nutritional deprivation (though not specifically of any
one macronutrient) in the early postnatal period may have beneficial effects on insulin resistance in
preterm infants in adolescence.17 We have previously shown aberrant AT partitioning, increased
intrahepatocellular lipid (IHCL) content and increased insulin resistance in preterm infants at term age
equivalent compared with healthy term infants.18,19 Our data suggest that, even as early as at term
equivalent age, preterm infants demonstrate the manifestations of cardiovascular risk factors.
1
INTRODUCTION
Improving the quality and quantity of nutrition in this period has the potential to improve not just
short-term outcomes but also the long-term neurodevelopmental and metabolic health of this vulnerable
group of infants. Preterm infants constitute a group that continues to utilise NHS resources throughout life
because of the long-term sequelae of prematurity. On average, health and societal costs for preterm
children at 6 years of age exceed that of a child born at term by approximately threefold.20
Nutritional requirements of preterm babies
Traditionally, RDIs have been based on the composition of fetal and newborn weight gain. Source data
are derived from the studies of Fomon and Nelson21 and Ziegler et al.22 on fetal cadavers of different
gestational ages. Based on the weight-gain composition at different periods of gestation and hence the
accretion rate of lean mass and fat mass, the dietary intake of energy necessary for preterm newborns to
achieve an intrauterine growth rate has been estimated as:
E intake = Eexcreted + E stored + Eexpended , (1)
where excreted energy comprises faeces and urine, stored energy is energy stored as protein and fat
(based on fetal accretion rate) and expended energy = resting metabolic rate + energy of activity +
thermoregulation (based on studies in growing preterm infants).
Using the above data, the American Academy of Pediatrics and the European Society for Paediatric
Gastroenterology Hepatology and Nutrition have published RDIs for preterm infants.23–25 These RDIs have
been used to inform this study. Putet et al.7 have pointed out that knowledge of growth rate is insufficient
to derive the optimum nutritional intake of preterm infants. The authors suggests that knowledge of
weight gain composition (lean and fat mass) is essential to estimate the ideal ratio of protein to energy in
order to avoid the deposition of excess energy as fat. Our previous work lends strength to this concept
as we have shown that preterm infants receiving current conventional intakes have a carbohydrate
and fat-rich diet, with a deficiency of protein and that they have a higher proportion of AT than
term-born infants.11
In a non-randomised study, Roggero et al.26, used whole-body plethysmography to measure weight gain
and LBM accural at 1 month post-term age in preterm infants fed either a high-protein diet (> 3 g/kg/day)
(n = 26) or a low-protein diet (< 3 g/kg/day) (n = 22). Weight gain was significantly lower in the high-protein
group than in the low-protein group {mean [standard deviation (SD)]: 946.7 g [375.2 g] vs. 1238 g [407 g];
p < 0.05}, but LBM accrual was asignificantly higher (approximately 4% higher as a percentage of
body weight).
Recent reviews have concluded that current nutritional practices contribute to long-term impairment and
recommend early introduction of the RDI of macronutrients.27,28 However, the evidence for this is based on
tolerability and growth outcomes, and not on body composition.
2
Parenteral nutrition
Early nutritional intake in extremely preterm infants is wholly or in part delivered intravenously as
parenteral nutrition (PN) because of immaturity of the gastrointestinal system. The median duration of
PN after birth in infants born before 31 weeks of gestation is 12 days. Often PN is recommenced later in
an infant’s neonatal course if the clinical condition precludes enteral feeding. Each day of PN costs the
NHS £80–100 per infant. A typical tertiary neonatal unit spends up to £150,000 per year on PN. There are
currently various PN preparations in routine use that vary in both composition and usage, but none has
previously been tested in this country in the setting of a large randomised controlled trial (RCT).
Some solutions are commercially prepared, whereas others are made up in local hospital pharmacies.
This has been the focus of a scoping exercise that was commissioned by the Department of Health
because of serious concern of clinical risk to patients.29 The survey carried out as part of the exercise
confirmed that current practice among neonatologists with respect to PN varies widely and is based on
limited evidence. There was also considerable variation in the preparation of PN and guidelines for use.
One hundred and sixteen hospitals reported providing PN to neonates and completed the survey relating
to neonates. The principal investigator was a member of the clinical group that developed and analysed
the survey and prepared the report. The report, which was published in November 2011, called for urgent
measures to standardise practice in both the technical and clinical aspects of use of PN in neonates and
children, and for the development of evidence-based guidelines for the use of PN.29 A further report from
the National Confidential Enquiry into Patient Outcome and Death (NCEPOD), to which the principal
investigator contributed, came to similar conclusions.30
Current widespread practice is to institute PN several hours to days after birth and to introduce
macronutrients in PN at a dose below that of the RDI and increase the quantity slowly over a period of
3–4 days, sometimes longer, often not achieving the RDI. This practice is non-evidence based and results in
cumulative deficits in protein and energy over the first 2 weeks of life. This practice is more prevalent
with respect to amino acids than carbohydrates and fat. Long-term use of PN results in liver impairment
and even failure. This is a particular problem in neonatal units caring for infants with bowel problems that
preclude or limit enteral feeding. There are now newer preparations of fat (SMOFlipid®; Fresenius Kabi AG,
Richmond Hill, ON, Canada) that have been found to be liver protective and are currently used in infants
on long-term PN.31 There is a need for studies to investigate the efficacy of these newer lipid solutions in
reducing liver impairment.
Previous studies of parenteral nutrition
Recent reviews have concluded that current nutritional practices contribute to growth failure and
recommend early introduction of the RDI of macronutrients in PN.27,28 However, the quality of the evidence
on which this is based is grade B (RCT with minor limitations, overwhelming consistent evidence from
observational studies) and only based on outcomes such as tolerability and growth, despite recognition
that the ideal postnatal growth rate of a preterm infant is unknown. No data exist on the effect on
body composition.
We have shown that the body composition of preterm infants is different from that of healthy term-born
infants. Preterm infants had a significantly reduced LBM and pattern of AT distribution associated with
metabolic complications.19 Tan et al.32,33 studied the effect of hyperalimentation on head growth. No
differences between the two groups were found, but non-randomised analyses showed protein and
energy deficits to be correlated with poor head growth. Eighty per cent of babies in the intervention group
had significant protein/energy deficits at the end of the first 4 weeks. A major drawback of this study was
that participants in this study were recruited up to 7 days after birth, by which time significant deficits
are known to have developed. The study was also underpowered to detect a significant effect on the
primary outcome.
3
INTRODUCTION
A systematic review of the effect of early administration of PN on growth outcomes in preterm infants
included eight RCTs and 13 observational studies.34 The review was limited by the disparate growth
outcome measures. Early PN reduced the time to regain birthweight by (a mean) 2.2 days [95% confidence
interval (CI) 1.1 to 3.2 days] in RCTs and 3.2 days (95% CI 2.0 to 4.4 days) in observational studies.
The maximum percentage weight loss with early PN was lower by (a mean) 3.1% (95% CI 1.7% to 4.5%)
for RCTs and by 3.5% (95% CI 2.6% to 4.3%) for observational studies. Early PN also improved weight at
discharge or 36 weeks postmenstrual age by (a mean) 14.9 g (95% CI 5.3 to 24.5 g) in observational
studies, but no benefit was shown for length or head circumference.34
A trial comparing two different amounts of amino acids (2.4 g/kg/day vs. 3.6 g/kg/day, with a lipid intake
of 2–3 g/kg/day; and an additional third arm of 2.4 g/kg/day of amino acids, with a delayed introduction of
lipids) from birth demonstrated an improved nitrogen balance on day 2 in the arms with early initiation
of lipids. There was no improvement in nitrogen balance with greater amounts of amino acids.35
A systematic review of the early introduction of lipids (defined as introduction within the first 2 days after
birth) and the use of new lipid emulsions included 14 RCTs.36 Early initiation of lipids had no impact on
any of the outcome measures, including death, bronchopulmonary dysplasia, necrotising enterocolitis,
patent ductus arteriosus, sepsis, intraventricular haemorrhage, significant jaundice and hypertriglyceridaemia.
The meta-analysis of the effects of lipid emulsions that are not purely soya bean based showed no difference
in outcomes of death, duration of respiratory support or rate of weight gain. There was a lower rate of sepsis
with the lipid emulsions that were not purely soya bean based, but the difference was not statistically
significant. However, the authors concluded that large-scale RCTs are needed to determine the efficacy of
newer lipids.36
We recently published a systematic review of preterm PN summarising the evidence to date.37 The review
concludes that the evidence base for current recommendations is based on historical evidence and there
are no long-term studies of the impact of PN on health and neurodevelopment.
Risks and benefits of parenteral nutrition
Parenteral nutrition is an independent risk factor for sepsis in neonates, associated with a 40-fold greater
risk, which makes its judicious use a priority. The risks associated with any form of PN are metabolic
disturbances (hyperglycaemia, hyperlipidaemia, electrolyte imbalances), infection38 and catheter-related
complications. However, these risks are unavoidable as PN is the only option for feeding extremely preterm
infants until they are established on enteral nutrition.
Parenteral nutrition is also associated with cholestasis and liver impairment.39 Instituting PN containing
the RDI of amino acids on the day of birth, as in the intervention arm, may result in a higher incidence of
metabolic acidosis and high concentrations of urea nitrogen in the blood. Until now, only one study has
investigated the efficacy of the early introduction of amino acids (3.5 g/kg/day) combined with a lipid
emulsion (3 g/kg/day), in high concentrations, within the first 2 hours of life. Early lipid introduction resulted
in an increased positive nitrogen balance without an increased incidence of metabolic or respiratory
complications.40 However, there was a small, but statistically significant, increase in serum bilirubin, without
clinical implications. Other studies in preterm infants using this approach have not found an increased
incidence of this problem.40–42
The lipid solution currently used, Intralipid 20% (Fresenius Kabi AG, Richmond Hill, ON, Canada), is a
first-generation lipid emulsion based on soya bean oil, which is very rich in n-6 polyunsaturated fatty acids.
However, an excess intake of n-6 polyunsaturated fatty acids in PN is associated with an unbalanced fatty
acid pattern in cell membranes, with possible modified function, and with increased lipid peroxidation.43
Second-generation emulsions are represented by medium-chain triglyceride (MCT) and long-chain
triglyceride (LCT) mixtures, and emulsions containing olive oils. MCT–LCT mixtures are cleared from the
4
bloodstream more quickly and generate more immediate energy. Emulsions containing olive oils provide a
more physiological mixture of fatty acids with less lipid peroxidation. An example of a third-generation
emulsion is SMOFlipid (a mixture of soya bean oil, MCTs, olive oil and fish oils, supplemented with vitamin E).
This emulsion is designed to increase the amount of n-3 fatty acids, thereby reducing the ratio of n-6 to n-3
fatty acids (in accordance with current recommended levels).43 SMOFlipid 20% is well tolerated in infants
without changing lipid peroxidation parameters,31,44 and beneficial effects on liver function and serum
triglyceride concentrations have been described.31
Need for the Nutritional Evaluation and Optimisation in

Neonates trial
In spite of evidence demonstrating that introducing the RDI of macronutrients early appears to be safe
and results in improved protein retention and better growth in the short term, clinical practice has
remained variable because of the absence of evidence from RCTs with clinically meaningful outcomes.
If early introduction of the RDI of macronutrients was shown, in the setting of a RCT, to improve not just
growth measured by anthropometry, but a better measure of growth (i.e. increase in LBM and better brain
growth, with the long-term benefits that in turn result from these) it has the potential to impact the vast
majority of neonatal unit graduates. There is an urgent need for therapy with PN to be evidence based.
5
Chapter 2 Research objectives

T he two main research objectives were to study the effects of two parenteral nutrition interventions
(amino acid quantity and lipid composition) in extremely preterm infants.
Amino acid intervention
To evaluate whether or not immediate rather than incremental introduction of the RDI of amino acids
(Imm-RDI) in extremely preterm infants results in:
l higher non-adipose (lean) body mass at term (primary objective)

l increased brain volume at term (secondary objective)
l reduced insulin resistance at term (secondary objective)
l lower ratio of internal to subcutaneous AT at term (secondary objective)
l the standard deviation (SD) score for weight undergoing a smaller drop between birth and term
equivalent (secondary objective).
Lipid intervention
To evaluate whether or not 20% SMOFlipid (with a lower ratio of n-6 to n-3 fatty acids) compared with
20% Intralipid in extremely preterm infants results in:
l reduced IHCL content at a term age equivalent (primary objective)

l a reduced incidence of hypertriglyceridaemia and hyperbilirubinaemia (secondary objective).
7
Chapter 3 Methods
Trial design
This was a multicentre, randomised, 2 × 2 factorial and double-blind controlled trial in four London and
south-east England centres in the UK. Eligible preterm infants were randomised within 24 hours of birth to
receive (1) either incremental amino acids (Inc-AA) in PN or the Imm-RDI from day 1; and (2) either 20%
Intralipid or 20% SMOFlipid.
There were four randomised groups:
1. Inc-AA and 20% Intralipid (Inc-AA/Intralipid)

2. Inc-AA and 20% SMOFlipid (Inc-AA/SMOFlipid)
3. Imm-RDI and 20% Intralipid (Imm-RDI/Intralipid)
4. Imm-RDI and 20% SMOFlipid (Imm-RDI/SMOFlipid).
Participants
Preterm infants (born before 31 weeks of gestation) requiring nutritional support in the form of PN.
Inclusion criteria
l Preterm infants born before 31 weeks of gestation (defined as ≤ 30 weeks and 6 days).
l Written informed consent from parents.
Exclusion criteria
l Major congenital or life-threatening abnormalities.

l Inability to randomise in time to allow administration of trial PN within 24 hours of birth.
Interventions
There were two main interventions, namely the amount of amino acids in PN and the type of lipid
formulation. All other components of PN were consistent across the four treatment groups. The
intervention was commenced within 24 hours after birth. Nutritional intake, both parenteral and enteral,
was guided by prespecified protocols that were provided in an investigator’s manual.
The interventions ceased once the infant was established, for at least 24 hours, on enteral feeds of
150 ml/kg/day. If the infant was subsequently nil by mouth after this point, PN was prescribed in
accordance with local practice as determined by the supervising clinician.
A summary of the interventions is provided in Table 1.
9
METHODS
TABLE 1 Summary of interventions
Intervention component Day 1 Day 2 Day 3 onwards
Inc-AA/Intralipid
Volume (excluding lipid volume) (ml/kg/day) 90 90 120
Protein (g/kg/day) 1.5 1.9 2.4
Amino acid equivalent (g/kg/day) 1.7 2.1 2.7
Carbohydrate (glucose) (g/kg/day) 8.6 8.6 8.6
20% Intralipid (g/kg/day) 2 3 3

Inc-AA/SMOFlipid
20% SMOFlipid (g/kg/day) 2 3 3
Imm-RDI/Intralipid
20% Intralipid (g/kg/day) 2 3 3
Imm-RDI/SMOFlipid

20% SMOFlipid (g/kg/day) 2 3 3
10
Outcomes
Primary outcomes
The efficacy of the early introduction of the RDI of amino acids was assessed by whole-body magnetic
resonance imaging (MRI) to measure lean mass, and by the quantity and distribution of AT. This
assessment was done at term age equivalent. The infants were scanned between 37 and 44 weeks
postmenstrual age.
Measurement of lean body mass

Lean body mass was calculated by subtracting AT mass from the weight of the infant on the day of the scan.
Measurement of intrahepatocellular lipid content

The efficacy of SMOFlipid was assessed by liver magnetic resonance spectroscopy (MRS) to measure IHCL
content. This was done at term age equivalent, between 37 and 44 weeks postmenstrual age.
Secondary outcomes
l Quantity and distribution of AT.

l Total and regional brain volumes.
l Metabolic index of insulin sensitivity [qualitative insulin sensitivity check index (QUICKI)].
l Serum lipids and bilirubin.
l Incidence of death.
l Anthropometry.
Data collection
Electronic case record form

Data management was through the InForm 4.6 (SP0c, build 1088; Oracle Corporation, Redwood, CA, USA)
integrated trial management system, a web-based data entry system that builds an Oracle Database 10g
(Enterprise Edition release 10.2.0.4.0 – 64bit; Oracle Corporation, Redwood, CA, USA) for each individual
clinical trial. Trial data were captured on a bespoke web-based electronic case record form (eCRF) with built-in
validation rules to identify data entry errors in real time and a full audit trail of data entry and changes.
All persons entering data were trained prior to start-up and given personal login details, with access to forms
restricted according to site and role. The eCRF was designed in accordance with the requirements of the trial
protocol and access to the eCRF was password protected and included a controlled level of access.
Timescale of trial evaluations
Daily evaluations
The first daily evaluation started at the time of birth and was completed when the first bag of trial PN was
changed and on the first day of postnatal life. Subsequent evaluations occurred 24 hours from this time
point (± 2 hours), every day from birth and until 37 weeks postmenstrual age or discharge from the
neonatal intensive care unit (NICU) (where days were calculated from the date PN was initiated).
Weekly evaluations
The first weekly evaluation occurred 7 ± 2 days from randomisation and each 7 days (± 2 days) thereafter
until 37 weeks corrected age or discharge from the NICU.
Monthly evaluation
The first monthly evaluation occurred 30 days (± 5 days) from randomisation and each 30 days (± 5 days)
thereafter until 37 weeks corrected age or discharge from the NICU.
For infants who received long-term PN, which is for at least 28 continuous days, serum trace elements
were measured.
11
METHODS
The 37-week evaluation

This evaluation took place when the infant reached 37 weeks postmenstrual age (± 1 week) or when the
infant was discharged from the NICU, whichever occurred sooner.
The end-of-study evaluation

The end-of-study evaluation took place as soon as possible after the infant was discharged from the NICU
at 37–44 weeks postmenstrual age. In the case of one hospital (Chelsea and Westminster NHS Foundation
Trust) with onsite access to a magnetic resonance (MR) scanner, infants aged between 37 and 44 weeks
postmenstrual age who were otherwise well but not ready for discharge, were scanned prior to discharge.
Schedule of investigations
A summary of tests and investigations performed is provided in Table 2.
TABLE 2 Summary of tests and investigations
End of study
(37–44 weeks
37 weeks and discharge
Evaluation Baseline Daily Weekly Monthly corrected age from the NICU)
Informed consent ✓
Eligibility ✓
Randomisation ✓
Weight a a ,b a
✓
Length a a
✓
Head circumference a a
✓
Blood pressure a
✓ ✓
Nutritional intake ✓ ✓
Safety
a ,b
Blood glucose (highest and
lowest in previous 24 hours)
a ,b
Worst base deficit on blood gas
(in previous 24 hours)
a ,b a
Serum bilirubin, LFTs, serum
urea, creatinine and electrolytes
a,b
Serum lipid and cholesterol
a ,b
Trace elements (zinc, copper,
manganese, aluminium and
selenium)
AE tracking ✓ ✓ ✓ ✓
Efficacy
QUICKI ✓
Whole-body and brain MRI, MRS ✓
Blood spot ✓ ✓
Urine sample and stool sample ✓

✓, for research purposes; AE, adverse event; LFT, liver function test.
a Routine care.
b While on PN.
12
Clinical investigations
Anthropometry
Weight, length and head circumference measurements are routinely used to monitor infant growth.
Weight was recorded on a daily basis until discharge and at the end of study visit while the infant received
PN, and weekly when the infant did not receive PN. Length and head circumference were recorded on a
weekly basis until discharge and at the end of study visit.
Blood pressure measurements

Systolic and diastolic blood pressure were measured in the right upper limb using a non-invasive blood
pressure monitor and a cuff that covered at least two-thirds of the right upper limb and encompassed the
entire arm in the resting state.
Magnetic resonance imaging

The MRI measurements were carried out during normal sleep without the need for sedation. All the MRI
measurements (body composition, hepatic MRS and brain MRI) took a total of 45–60 minutes. The infants
were monitored with pulse oximetry and a trained neonatal doctor was present throughout the scan.
Parents were invited to be present in the console room.
Magnetic resonance imaging body composition
Acquisition of images
Scans were undertaken after discharge from hospital at the Robert Steiner MR Unit, Imperial College
Healthcare NHS Trust at a dedicated research scanning facility on a 1.5-T Phillips Achieva scanner
(Philips, Best, the Netherlands). Babies born at the lead site (Chelsea and Westminster NHS Foundation
Hospital) who were still inpatients between 37 and 44 weeks postmenstrual age and unlikely to be
discharged home in time to be scanned in the research scanner were scanned while inpatients at Chelsea
and Westminster NHS Foundation Hospital scanner on a 1.5-T Siemens Avanto scanner (Siemens,
Erlangen, Germany).
For images that were acquired on the Phillips 1.5-T system, a T1-weighted rapid-spin-echo sequence
(repetition time of 500 milliseconds, echo time of 17 milliseconds, echo train length of 3) using a Q body
coil was used. The slice thickness was 5 mm and the interslice difference was 5 mm. Voxel size was
0.31 × 0.31 × 0.31 cm. Acquisition time was approximately 12 minutes. For images acquired on the Siemens
scanner a T1 turbo-spin-echo sequence was used (with a repetition time of 514 milliseconds and an echo
time of 11 milliseconds).
Analysis of images
Analysis of all MR images was undertaken independently of the investigators, blind to subject identity and
treatment, by Vardis Group (London, UK; www.vardisgroup.com). Images were analysed by a single observer,
using a commercially available software program (SliceOMatic, Version 4.2; Tomovision, Montreal, QC,
Canada). A filter was used to distinguish between different grey-level regions on each slice. This was then
verified and, where necessary, edited using the interactive slice editor program. AT area (cm2) for each slice
was calculated as the sum of the voxels multiplied by the voxel area. AT volume (cm3) for each slice was
calculated by multiplying the tissue area by the sum of the slice thickness and the interslice distance. The
coefficient of variation for these measurements was < 3%.45 AT volume in litres was converted to AT mass in
kg, assuming a value for the density of AT of 0.90 kg/l.46,47
AT mass, kg = (AT volume, l) × 0:90: (2)
Adipose tissue mass was used to determine percentage AT as:
Percentage AT mass = (AT mass, kg)=(body mass, kg). (3)
13
METHODS
Total AT volume was calculated as the sum of six individually quantified AT compartments – superficial
subcutaneous abdominal AT, superficial subcutaneous non-abdominal AT, deep subcutaneous abdominal AT,
deep subcutaneous non-abdominal AT, internal abdominal AT and internal non-abdominal AT – as previously
described45 (Figure 1). Total subcutaneous AT was calculated as the sum of abdominal superficial
subcutaneous, abdominal deep subcutaneous, non-abdominal superficial subcutaneous and non-abdominal
deep subcutaneous AT. Total internal AT was calculated as the sum of internal abdominal and internal
non-abdominal AT.
Hepatic magnetic resonance spectroscopy
Acquisition of spectra
Hydrogen-1 (1H) MR spectra were acquired at 1.5 T from the right lobe of the liver using a point-resolved
spectroscopy sequence (repetition time 1500 milliseconds/repetition time 135 milliseconds) without water
saturation and with 128 signal averages. Transverse images of the liver were used to ensure accurate
positioning of the (20 × 20 × 20 mm) voxel in the liver, avoiding blood vessels, the gall bladder and fatty
tissue. For spectra acquired on the 1.5-T Siemens Avanto scanner a voxel size of 15 × 15 × 15 mm
was used.
Analysis of spectra
Spectra were analysed in the time domain using the advanced method for accurate robust and efficient
spectral fittings algorithm included in the Java-based MR user interface software package (version 1.3; MRUI
consortium; www.jmrui.eu/) by a single investigator (LT) who was blind to the treatment category.48–50
Peak areas for all resonances were obtained and lipid resonances were quantified with reference to water
resonance, after correcting for T1 and T2. Hepatic water, known to be relatively constant, was used as an
internal standard and the results are presented as the percentage ratio of fat CH2 to water.
Brain magnetic resonance imaging
Acquisition of images
Brain imaging was performed on infants using a dedicated eight-channel paediatric coil. Three-dimensional
T1-weighted fast-gradient echo images were acquired in a sagittal plane with using the following
parameters: field of view (FOV) 220 × 158 mm; 192 slices; slice thickness 1 mm; an acquired voxel size
0.82 × 0.97 mm; matrix 256; echo time 4.6 milliseconds; repetition time 17 milliseconds; flip angle 13°;
and acquisition time 6 minutes.
Total AT
Superficial Deep
Internal
subcutaneous subcutaneous
(IAT)
(SSCAT) (DSCAT)
Abdominal Non-abdominal Abdominal Non-abdominal Abdominal Non-abdominal

(SSCAAT) (SSCNAAT) (DSCAAT) (DSCNAAT) (IAAT) (INAAT)
FIGURE 1 Classification of AT depots. DSCAAT, deep subcutaneous abdominal adipose tissue; DSCAT, deep
subcutaneous adipose tissue; DSCNAAT, deep subcutaneous non-abdominal adipose tissue; IAAT, internal
non-abdominal adipose tissue; IAT, internal adipose tissue; INAAT, internal non-abdominal adipose;
SSCAAT, superficial subcutaneous abdominal adipose tissue; SSCAT, superficial subcutaneous adipose tissue;
SSCNAAT, superficial subcutaneous non-abdominal adipose tissue. Adapted with permission from Modi N et al.,
Pediatric Research 2009;65:584–7.45
14
Whenever possible, and with time permitting, the following brain scans were also undertaken:
l T2-weighted turbo-spin-echo sequence acquired in an axial plane with FOV 220 × 220 mm; 94 slices;
slice thickness 2 mm; acquired voxel size 1.15 × 1.42 mm; slice gap 1 mm; matrix 256; echo time
160 milliseconds; repetition time 15,077 milliseconds; flip angle 90°; and acquisition time 2 minutes.
l A three-dimensional time-of-flight MR angiography sequence to assess the anterior cerebral artery,
middle cerebral artery and posterior cerebral artery. The imaging parameters used were FOV
175 × 144 mm; 75 slices; one stack; slice thickness 0.8 mm; slice gap 0 mm, voxel size 0.61 × 0.61 mm;
echo time 12 milliseconds; repetition time 23 milliseconds; flip angle 16°; matrix 512; and acquisition
time 5 minutes.
l Fifteen direction diffusion tensor imaging for assessment of white matter integrity also formed part of
the protocol, with the following imaging parameters: FOV 224 × 224 mm; 49 slices; slices thickness
2.5 mm; slice gap 0 mm; acquired voxel size 2 × 2 mm; matrix 128; echo time 49 milliseconds;
repetition time 49,709 milliseconds; maximum b factor 750; number of b factors 2; and acquisition
time 6 minutes.
Analysis of images
A specialist in neonatal neurology reported all brain MRI images for clinical purposes. A note was made of
any congenital or acquired lesions. The type and severity of these was recorded for all cases. Scans with
parenchymal brain lesions were excluded from subsequent quantitative analysis.
A quantitative whole-brain segmentation program was used to segment the brain and its constituent
structures using the T2-weighted image data.51 These volumetric data could be obtained only from images
that were of adequate quality with good signal-to-noise ratio and absence of motion artefact.
The following outcomes were measured:
l total cerebral volume: sum of the volumes of basal ganglia, thalami (deep grey matter), cerebrospinal
fluid, grey matter, white matter and lateral ventricles
l whole-brain volume: sum of the volumes of basal ganglia, thalami (deep grey matter), grey matter and
white matter
l posterior fossa volume: sum of the volumes of cerebellum and brainstem.
Quantitative insulin sensitivity check index

The QUICKI is a marker of insulin resistance calculated from pre-feed serum insulin and blood glucose.
Homeostatic model assessment is the gold standard for measuring insulin resistance, but is invasive and
cannot be justified ethically in this patient group. Furthermore, the QUICKI has been validated against
homeostatic model assessment52 and has been used in neonates before.53 Measurement of the QUICKI
was carried out at term age and samples taken at the time of routine (pre-feed) blood tests.
Pharmacovigilance definitions and procedures
Serious adverse events

An adverse event (AE) was defined as any untoward medical occurrence in a patient administered an
investigational medicinal product (IMP), in accordance with clinical trial regulations. An AE was considered
serious and reportable via the eCRF if any of the following criteria occurred:
l it resulted in death
l it was life-threatening
l it resulted in prolongation of existing inpatient hospitalisation
l it resulted in persistent or significant disability or incapacity.
15
METHODS
Expectedness and causality of serious adverse events

The trial protocol specified that a range of serious adverse events (SAEs) would be expected, either as a
consequence of preterm birth or if they were listed in any of the summaries of product characteristics, and
this expectedness was recorded on the eCRF for each SAE report.
Causal relationship to the IMP was defined according to Table 3.
Reporting of adverse events

The trial eCRF included dedicated forms for reporting SAEs. Investigators were advised to report SAEs via
the eCRF within 24 hours of becoming aware of the event and to include an assessment of expectedness
and causality in the SAE report. The clinical trials unit and chief investigator reviewed each SAE report
within 2 working days.
Adverse events
The only non-serious AEs that were reportable were values of triglycerides, bilirubin and other safety
parameters above or below prespecified levels, and these are summarised in Table 4. These were labelled
as ‘specific adverse events’ (SpAEs) reportable via the eCRF. The eCRF incorporated in-built checks to flag
any occurrence of a SpAE during the data entry process to the local teams. Guidance for the management
of these events was provided to the participating centres in a trial-specific investigator’s manual. SpAEs
related to safety parameters were collected daily during the period of trial PN administration.
As the levels selected for SpAEs were consistent with normal ranges used in standard neonatal clinical care
and, in accordance with the new Medicines and Healthcare products Regulatory Agency (MHRA) guidance
on risk-adapted approach to managing clinical trials, the Nutritional Evaluation and Optimisation in
Neonates (NEON) trial was equivalent to standard care, additional reporting and review of SpAEs were
not required. The trial Data Monitoring and Ethics Committee (DMEC) reviewed a selection of SpAEs
throughout the duration of the trial.
The thresholds for SpAEs as well as those requiring reporting to the DMEC are summarised in Table 4.
TABLE 3 Definitions for assessment of causality
Relationship Description
Unrelated There is no evidence of any causal relationship
Unlikely There is little evidence to suggest there is a causal relationship (e.g. the event did not occur within a
reasonable time after administration of the trial medication). There is another reasonable explanation for
the event (e.g. the patient’s clinical condition, other concomitant treatment)
Possiblea There is some evidence to suggest a causal relationship (e.g. because the event occurs within a
reasonable time after administration of the trial medication). However, the influence of other factors may
have contributed to the event (e.g. the patient’s clinical condition, other concomitant treatments)
Probablea There is evidence to suggest a causal relationship and the influence of other factors is unlikely
Definitelya There is clear evidence to suggest a causal relationship and other possible contributing factors can be
ruled out
Not assessable There is insufficient or incomplete evidence to make a clinical judgement of the causal relationship
SUSAR, suspected unexpected serious adverse reaction.
a SUSAR: if an AE was considered serious, unexpected and related to the IMP (possible, probable or definitely related) this
would have met the definition of SUSAR requiring expedited reporting to the MHRA, Research Ethics Committee and
sponsor. There were no SUSARs for the NEON trial.
16
TABLE 4 Definitions of SpAEs including thresholds for reporting to the DMEC
Assessment (blood test) Level requiring SpAE report Level requiring reporting to the DMEC
Glucose < 2.6 mmol/l or > 15 mmol/l Not reported to the DMEC
Worst base deficit in previous > 15 mmol/l > 15 mmol/l

24 hours
Total serum bilirubin > 150 µmol/l > 150 µmol/l, only after 3 weeks on PNa
Conjugated bilirubin > 40 µmol/l > 40 µmol/l
Cholesterol > 6 mmol/l > 10 mmol/l
Triglycerides > 2.5 mmol/l > 5 mmol/l
Sodium < 131 mmol/l or > 150 mmol/l Not reported to the DMEC
Potassium < 3.2 mmol/l or > 9 mmol/l Not reported to the DMEC
Phosphate < 1.5 mmol/l or > 3 mmol/l Not reported to the DMEC
Calcium < 1 mmol/l or > 3 mmol/l Not reported to the DMEC
Urea < 1.5 mmol/l or > 7 mmol/l > 10 mmol/l
Creatinine > 170 µmol/l Not reported to the DMEC
Alanine transaminase > 60 IU/l Not reported to the DMEC
Zinc < 8 µmol/l Not reported to the DMEC
Copper < 2 µmol/l Not reported to the DMEC
Manganese > 30 nmol/l Not reported to the DMEC

Aluminium > 0.4 µmol/l Not reported to the DMEC
Selenium < 20 µg/l Not reported to the DMEC

a The infant must have been on PN for at least 3 weeks for this to meet the requirements for reporting to the DMEC.
Annual safety reports

Annual safety reports were provided to the Research Ethics Committee and MHRA, in accordance with
clinical trial regulations, on the anniversary of the clinical trial authorisation each year. A total of three
annual safety reports were submitted over the course of the trial.
Statistical considerations
Sample size
The mean directly measured LBM of preterm infants when studied in 2003 was 2.1 kg (SD 0.4 kg).17 The
mean in healthy term-born infants was 2.6 kg (SD 0.21 kg; mean difference 450 g, 95% CI 300 to 610 g).
A sample size of 64 infants in each group was therefore chosen, as this would allow detection of a 200 g
difference between the groups with 80% power and at 5% significance. This was considered a clinically
important increase in lean mass.
Since the publication of our paper on IHCL,18 measurements were available for a total of 15 infants with
gestational ages ranging from 24 weeks to 32.6 weeks. IHCL had a mean lipid-to-water ratio of 1.75
(SD 1.85, range 0.14–7.72); the distribution is clearly positively skewed. A loge-transformation was therefore
used to achieve approximate normality. On the natural logarithmic scale the mean IHCL lipid-to-water ratio
was 0.121 (SD 1.052, range –1.97 to 2.04). A sample size of 64 infants in each group would therefore have
80% power to detect a difference in means of 0.526 on the logarithmic scale as significant at the 5%
significance level (with a t-test). Transforming back to the original scale of measurement, this is equivalent
to a 40% decrease in IHCL content in the intervention group.
17
METHODS
Assuming a 10% mortality prior to term and a 10% dropout rate, the aim was to recruit 80 infants to
each group or until 64 infants in each group had undergone MRI and MRS, a total of 128 scans.
Randomisation
Randomisation was performed using an interactive voice recognition system (IVRS) telephone
randomisation system. Sealed Envelope Ltd (London, UK) provided the IVRS and randomisation list.
Randomisation was performed using minimisation, with a 25% chance of simple random allocation (based
on the procedure outlined in Pocock54). Randomisation was stratified by gestational age at birth (23–26 or
27–31 completed weeks of gestation), birthweight (< 500 g, 500–1000 g, > 1000 g) and centre. Multiple
births were randomised individually.
Blinding
Unblinded trial PN was delivered to the pharmacy department at each participating centre. Trained
pharmacy staff were responsible for blinding the trial PN prior to dispensing the supply for administration
to each infant.
Secure copies of the randomisation list were held by each pharmacy team in case of the need for
emergency unblinding. There was no requirement for unblinding at any point over the course of the study.
Statistical methods
The analysis of this 2 × 2 factorial randomised trial was performed ‘at the margins’ of the 2 × 2 table,
assuming that the two factors are operating independently. In addition, summary measures were
presented for each cell of the 2 × 2 table and an interaction ratio/difference was calculated.55 A ‘modified’
intention-to-treat method was used to analyse the results as it was accepted that a proportion of infants
would not be able to attend for MRI. With the exception of infants in whom MRI assessment was not
completed, all infants were analysed according to their allocation.
The primary outcome measures for this trial were non-adipose (lean) body mass and IHCL content; the
secondary outcomes were growth (weight, length and head circumference), brain growth and development
(assessed by MRI) and measure of insulin sensitivity (by the QUICKI). Growth parameters are the only
outcomes that were measured sequentially; all other outcomes, including the two primary outcomes,
were measured on a single occasion at term age equivalent.
For outcomes measured on a single occasion, a regression model containing the stratifying variables
(gestational age, birthweight and centre), nutritional interventions (amino acid and lipid), sex and age at
time of measurement were used to estimate the effects of each intervention.
For the amino acid intervention primary outcome, a multiple regression was used with non-adipose body
mass (g) as the dependent variable and amino acids (incremental vs. RDI), lipids (20% SMOFlipid vs.
Intralipid), gestational age, birthweight, centre, sex and age at MRI as the independent variables to assess
the effect of amino acids on non-adipose body mass. An interaction term was also included to assess
whether or not the effect of amino acids regimen on non-adipose body mass is influenced by choice
of lipids.
Similarly, for the lipid intervention primary outcome, a multiple regression was used with IHCLs at natural
logarithmic scale as the dependent variable and amino acid (incremental vs. RDI), lipids (20% SMOFlipid vs.
Intralipid), gestational age, birthweight, centre, sex and age at MRI as the independent variables to assess
the effect of lipids on IHCL content. Again, an interaction term was included to assess whether or not the
effect of lipids on IHCL content is affected by amino acid quantity.
18
A planned secondary analysis was used to investigate the role of illness severity and nutritional intake as
potential modifiers of the effects of each intervention by adding these variables to the regression models.
The secondary analysis investigated the role of illness severity, maternal breast milk and post-PN intake,
including PN period and post-PN period, as potential modifiers of the effects of each intervention by
adding these variables to the regression model. All analyses were performed using Stata 13 (StataCorp LP,
College Station, TX, USA).
All analyses were performed on an intention-to-treat basis, but as the primary outcomes can be
ascertained in only those infants attending the end of study evaluation, up to 20% of primary outcomes
are expected to be missing. We have assumed that these outcomes are missing at random.
Missing data
Owing to the nature of this study, it was expected that a number of infants would not undergo the
end-of-study MRI (primarily because of death, ill-health or withdrawal of the subject). This was taken into
account when calculating the sample size. The statistical analysis plan prespecified that we would analyse
only those infants who could be scanned. The reasons for non-attendance were recorded in the
withdrawal form. We aimed to comment on the implications that the missing data patterns had on the
results from the analysis.
No missing data imputation was carried out except for infant weight over study period. Infant weight was
recorded every day during the trial study period and weekly once infants were off the trial. As the daily
infant weight was used in the descriptive analysis only, we did not carry out multiple imputations. Instead,
we used simple imputation by using the nearest measured weight, either before or after the day of missing
weight, to impute the missing data.
Statistical analysis plan

A statistical analysis plan was prepared by the trial investigators and trial statistician and reviewed and
agreed by the Trial Steering Committee (TSC) and DMEC prior to the end of the recruitment period.
Trial organisation
Trial management
The UK Clinical Research Collaboration-registered Imperial Clinical Trials Unit (ICTU) was responsible for
trial management, quality assurance, trial statistics, and development and maintenance of the trial
database. The Clinical Trials and Evaluation Unit at the Royal Brompton and Harefield NHS Foundation Trust
carried out trial and data management, which was one of the ICTU groups at the time of the trial.
The ICTU core staff and the InForm team are supported by the National Institute for Health Research (NIHR)
Biomedical Research Centre based at Imperial College Healthcare NHS Trust and Imperial College London.
Trial sponsor
The sponsor of the trial was Imperial College London. The sponsor’s role is clearly set out in the European
Clinical Trials Directive (http://ec.europa.eu/health/human-use/clinical-trials/directive/index_en.htm) and NHS
Research Governance documents (www.gov.uk/government/uploads/system/uploads/attachment_data/file/
139565/dh_4122427.pdf). Imperial College London signed a clinical trial agreement with each of the
participating centres prior to the start of the trial.
Ethical considerations
The trial was conducted in accordance with the Declaration of Helsinki on research involving human subjects.
The study protocol, parent information sheet (PIS) and consent form were submitted to the Research Ethics
Committee prior to the start of the study and a favourable opinion was obtained on 8 December 2009.
19
METHODS
Consent
Where possible, parents were approached prior to their infant’s birth to give them the PIS and discuss
the trial. Full written informed consent was taken after birth using the ethically approved PIS and
consent form.
Research governance
The trial was carried out in accordance with the NHS Research Governance Framework, and local NHS
permission was granted by the research and development departments at each participating site prior to
recruitment commencing.
Regulatory requirements
As a randomised trial of an IMP, the NEON trial was conducted in accordance with the European Clinical
Trials Directive and the Medicines for Human Use (Clinical Trials) Regulations 2004.56 The trial received
clinical trials authorisation from the MHRA on 8 January 2010 and was registered in the European
Community with the European Clinical Trials Database (EudraCT) number 2009-016731-34.
Trial registration
The trial was registered on the International Standard Randomised Controlled Trial Number (ISRCTN)
clinical trial database with reference ISRCTN29665319.
National Institute for Health Research Clinical Research Network portfolio

The NEON trial was adopted on the NIHR Clinical Research Network and Medicines for Children Research
Network portfolios. Accrual data were uploaded onto the NIHR Clinical Research Network database on a
monthly basis.
Summary of protocol amendments

The ethics committee and MHRA made the following amendments to the trial protocol following approval
of the first version of the document:
l Protocol version 2:
¢ Clarifications implemented following review by TSC:
¢ clarification that randomisation would be performed by minimisation with 25% chance of

random allocation
¢ randomisation to be stratified by birthweight in addition to existing factors (centre and
gestational age at birth)
¢ addition of a monthly evaluation to assess trace elements for infants on PN for > 28 days.
¢ Administrative corrections.
¢ Addition of a metabonomic substudy (funded separately and not reported in this article).
¢ Additional blood samples on days 1 and 5 of life to assess inflammatory markers and lipid profile.
The intention was to conduct a substudy to collect these samples at the lead site but it was
never implemented.
¢ Clarification of randomisation time window. The protocol previously stated that infants must be
randomised within 12 hours of birth. The purpose of this time window was to allow adequate time
for preparation and dispensing of trial PN. The time window was revised for this version of the
protocol so that infants needed to be randomised in enough time to allow administration of PN
within 24 hours.
¢ Administrative corrections.
20
¢ The protocol was amended to include a follow-up visit for neurodevelopmental outcomes at
2 years corrected age using the Bayley Scale of Infant Development, the Hammersmith Optimality
Score as well as parental questionnaires (Social-Emotional scale of the Bayley Scales and the
Quantitative Checklist for Autism in Toddlers). A funding application for this additional visit was not
successful, so the additional visit was not implemented.
Trial committees
Trial steering committee

A TSC was established to oversee the conduct of the study. The TSC met three times over the course
of the trial: on 26 February 2010, 14 December 2011 and 22 November 2012. Copies of the minutes from
each meeting were sent to the funder, the Efficacy and Mechanism Evaluation (EME) programme of the
NIHR. The TSC approved the trial protocol prior to the start of the study and received regular recruitment
reports throughout the duration of the trial.
The TSC membership is listed below.
l Independent members:
¢ Professor Richard Cooke (chairperson).

¢ Mrs Lorraine Dob (parent representative).
¢ Dr Paul Clarke.
¢ Professor Robert Hume.
l Investigators:
¢ Dr Sabita Uthaya (chief investigator).

¢ Professor Neena Modi.
¢ Caroline Doré.
¢ Professor Ian Wong.
¢ Professor Jimmy Bell.
¢ Professor Deborah Ashby.
Data Monitoring and Ethics Committee

An independent DMEC was established to review SAE reports and the results of interim analyses.
The DMEC meetings took place on 2 August 2010, 13 October 2011 and 27 September 2012.
The first DMEC meeting, to agree the charter outlining operational details and responsibilities, took place
early in the trial, on 2 August 2010. The second meeting to review interim data for the first 32 infants
was on 13 October 2011 and the final interim analysis for 64 infants took place on 27 September 2012.
The DMEC provided feedback reports for each meeting to the chairperson of the TSC and this was
reviewed, as applicable, at subsequent TSC meetings.
Data Monitoring and Ethics Committee membership:
l Professor Peter Brocklehurst (chairperson).

l Professor Tim Cole (independent statistician).
l Professor Tony Nunn.
l Dr Helen Mactier.
21
METHODS
Data management
Predefined data ranges were included in the eCRF, which raised automated queries if data outside of the
expected range were entered. In addition to the automated queries, the trial data were reviewed on a
regular basis by the data manager to look for discrepancies and errors. Furthermore, the trial statistician
also performed a series of checks on snapshots of data to look for inconsistencies. The checks performed
by the data manager and statistician were documented in a prespecified data management plan, which
was updated over the course of the study as required.
Risk assessment and monitoring plan

A risk assessment was performed by the ICTU quality assurance manager prior to the start of the trial.
The result of the risk assessment indicated that the study was low risk and that 20% of trial data, 100%
consent forms and 100% SAEs should be source verified. A monitoring plan was prepared in accordance
with the risk assessment to specify the frequency of monitoring visits and number of source data
verification required.
Monitoring visits
A site initiation visit was performed at all participating centres. Interim monitoring visits were carried out
approximately annually, depending on the recruitment rate, and closeout visits were carried out at all
centres following the final follow-up visit for the last patient recruited. The monitoring visits were
conducted by the trial manager.
Investigational medicinal product manufacturer

The IMP for the NEON trial was manufactured by Bath-ASU (Wiltshire, UK), a MHRA-licensed
manufacturing unit, with expertise in producing aseptic products.
Patient and public involvement

The TSC membership included a parent representative who was invited to attend all TSC meetings and
included in all relevant correspondence. The long-term follow-up of the infants involved in the trial
was something parents of babies who took part in the study were keen to see.
Parents were consulted during preparation of the PIS and the charity Bliss was also approached during the
design phase of the study. Parent representatives contributed by suggesting changes to the PIS, including
reducing the length and complexity of information.
22
Chapter 4 Results
Participant flow
The flow of patients is summarised in Figure 2, including the number of patients screened, randomised
and completing the trial.
Screening
Four hundred and sixty infants below 31 weeks of gestational age were admitted to the participating
hospitals over the duration of the trial. Of the 382 infants meeting the eligibility criteria, 168 were
randomised to the trial. Figure 3 summarises the percentage of eligible patients recruited to the trial and
reasons for non-recruitment.
Infants
Infants transferred to non-trial site
< 31 weeks of
immediately after birth; not eligible
gestational age
for inclusion in the trial (n = 78)
(N = 460)
Reasons not recruited

Number of • Declined, n = 73
infants eligible • Recruited to another trial, n = 3
(N = 382) • Missed (pharmacy unavailable), n = 46
• Missed/not approached, n = 33
• Unknown, n = 59
Infants
randomised
(N = 168)
Inc-AA/Intralipid Inc-AA/SMOFlipid Imm-RDI/Intralipid Imm-RDI/SMOFlipid

(N = 42) (N = 42) (N = 41) (N = 43)

(N = 34) (N = 28) (N = 34) (N = 37)
• Deaths, n = 3 • Deaths, n = 7 • Deaths, n = 3 • Deaths, n = 3
• Withdrawals, n = 5 • Withdrawals, n = 7 • Withdrawals, n = 4 • Withdrawals, n = 3
(four lost to (two lost to follow-up; (two lost to (two lost to follow-up;
follow-up; one and two still inpatients follow-up; one still and one withdrew
still inpatient at at 44 weeks; one inpatient at 44 weeks; consent)
44 weeks) investigator and one investigator
withdrawal;a one set withdrawala)
of parents consented
to a conflicting trial;
and one withdrew
consent)
FIGURE 2 The Consolidated Standards of Reporting Trials diagram. a, Investigator withdrawal: in both cases, this
occurred when the infant was transferred to a non-trial site very soon after randomisation and was therefore
unable to receive the trial intervention.
23
RESULTS
19%
1%
Parents refused
44% Recruited to another study
Missed: pharmacy unavailable
12% Missed: not approached
Unknown
Recruited
9%
15%
FIGURE 3 Summary of screening data for all trial sites.
Recruitment and retention
Recruitment lasted for 3 years; the first infant was recruited on 6 July 2010 and the last on 31 July 2013.
The actual recruitment period was longer than the original target of 2.5 years because of delays starting
the trial at all sites. The delays in starting the trial were associated with the following:
(a) Identifying a suitable manufacturer for the trial with an IMP licence to produce PN and the capacity to
support the trial.
(b) Agreement from each centre to support excess treatment costs because of the cost difference
between standard hospital PN and trial PN supply, including signing a procurement contract for each
participating pharmacy.
(c) Obtaining NHS permission at each site was lengthy, the procurement process was a factor for this.
(d) Inability to recruit during weekends and holidays. Pharmacy departments at three out of four sites
could not support recruitment at weekends or during Christmas and Easter, which reduced the
recruitment rate.
Recruitment rate
The target recruitment rate for the study was six patients per month, based on all four centres recruiting.
The average monthly recruitment rate once all centres were activated (January 2012) was consistent
with the target, that is, six patients per month.
Figures 4–6 summarise cumulative recruitment and retention over the course of the study, and recruitment
and retention per centre.
Baseline data
The baseline characteristics of the infants recruited to the study and those who completed the MR
assessment of primary outcome measures are shown in Tables 5 and 6, respectively. Of the 437 infants
born before 31 weeks of gestation, 168 infants were randomised. A total of 133 infants were available for
assessment of the primary outcome measures. Baseline characteristics of sex, gestational age at birth,
anthropometry, maternal demographics, mode of delivery, antenatal steroid use and time to commencing
PN were similar across the four groups.
24
Target recruitment
Actual recruitment
July
June
May
2013
April
March
February
January
December
November
October
September
August
2012
July
June
May
April
Duration of the trial
March FIGURE 4 Cumulative recruitment vs. target recruitment for the duration of the trial.
February
January
December
November
October
September
August
2011
July
June
May
April
March
February
January
December
November
2010
October
September
August
July
180
160
140
120
100
80
60
40
20
Number of patients recruited
25
26
Number of MR scans
0
20
40
60
80
100
120
140
July
August
September
October
2010
November
December
January
February
March
April
May
June
July
2011
August
September
October
November
December
January
February
March
Duration of the trial
April
May
June
July
2012 August
September
October
November
December
January
FIGURE 5 Cumulative retention (number of MR scans) vs. target retention for the duration of the trial.
February
March
April
May
June
2013
July
August
September
October
Target MR scans
Actual MR scans performed
RESULTS
Number of patients recruited

MR scans completed
Medway Maritime
13
Hospital
20
Park Hospital
Northwick
25
28
Centre
West Middlesex
25
Hospital
29
FIGURE 6 Total recruitment and retention per centre.

Westminster
Chelsea and
70
Hospital
91
100
90
80
70
60
50
40
30
20
10
Number of patients/MR scans completed
27
RESULTS
TABLE 5 Baseline characteristics for all infants randomiseda
Imm-RDI/ Imm-RDI/
Inc-AA/Intralipid Inc-AA/SMOFlipid Intralipid SMOFlipid
Characteristic (N = 42) (N = 42) (N = 41) (N = 43)
Infant sex, n (%)
Male 28 (66.7) 26 (61.9) 21 (51.2) 22 (51.2)
Gestational age (weeks), mean (SD) 27.8 (1.9) 27.5 (2.4) 28.1 (2.1) 27.8 (2.1)
Multiple births, n (%)

Yes 6 (14.3) 6 (14.3) 9 (22.0) 15 (34.9)
Birthweight (kg), mean (SD) 1.03 (0.29) 1.05 (0.34) 1.04 (0.28) 1.06 (0.29)
Birth length (cm), mean (SD) 35.1 (3.5); n = 31 34.6 (4.2); n = 32 35.1 (3.9); n = 26 35.2 (5.2); n = 32
Head circumference (cm), 25.3 (2.0); n = 41 25.0 (3.0); n = 40 25.3 (1.9); n = 37 25.6 (2.9); n = 39
mean (SD)
Birthweight (z-score), mean (SD) –0.2 (1.0); n = 42 0.1 (1.0); n = 41 –0.2 (1.0); n = 41 0 (0.9); n = 43
Birth length (z-score), mean (SD) –1.0 (1.0); n = 30 –0.9 (1.2); n = 24 –1.1 (1.0); n = 25 –0.8 (1.5); n = 29
Head circumference (z-score), –0.5 (0.9); n = 41 –0.3 (1.0); n = 39 –0.7 (0.9); n = 37 –0.2 (1.6); n = 41
mean (SD)
Mother’s age (years), mean (SD) 32.9 (5.3); n = 42 31.3 (7.7); n = 42 32.9 (6.3); n = 40 32.5 (6.6); n = 43
Mother’s weight (kg),b mean (SD) 66.4 (13.3); n = 34 65.9 (11.4); n = 25 64.9 (13.0); n = 30 68.5 (15.2); n = 33
b
Mother’s height (cm), mean (SD) 161.9 (7.8); n = 33 164.9 (7.7); n = 27 161.3 (9.2); n = 27 164.5 (8.6); n = 32
b
Father’s weight (kg), mean (SD) 80.8 (10.7); n = 27 82.3 (13.2); n = 22 85.3 (16.1); n = 24 86.3 (14.9); n = 31
Father’s height (cm),b mean (SD) 178.4 (6.5); n = 28 179.6 (6.8); n = 22 175.7 (10.0); n = 22 182.0 (9.7); n = 30
Mother’s ethnicity, n (%)

White 16 (38.1) 19 (45.2) 21 (51.2) 21 (48.8)
Asian 14 (33.3) 7 (16.7) 12 (29.3) 12 (27.9)
Black 6 (14.3) 13 (31.0) 6 (14.6) 6 (14.0)
Mixed 2 (4.8) 2 (4.8) 1 (2.4) 2 (4.7)
Other 3 (7.1) 0 (0) 1 (2.4) 2 (4.7)
Missing 1 (2.4) 1 (2.4) 0 (0) 0 (0)
Mode of delivery, n (%)

Vaginal 8 (19.1) 18 (42.9) 16 (39.0) 17 (39.5)
Elective caesarean 7 (16.7) 3 (7.1) 4 (9.8) 2 (4.7)
Emergency caesarean 27 (64.3) 21 (50.0) 21 (51.2) 24 (55.8)
Antenatal steroids, n (%)
Yes 30 (71.4) 34 (81.0) 32 (78.1) 35 (81.4)
No 7 (16.7) 6 (14.3) 7 (17.1) 4 (9.3)
Unknown 5 (11.9) 2 (4.8) 2 (4.9) 4 (9.3)

Time from birth to starting PN, 18.4 (12.3–22.7); 19.5 (13.6–22.8); 20.4 (12.6–23.6); 17.7 (13.0–22.4);
(hours),c median (IQR) n = 42 n = 41 n = 40 n = 43
IQR, interquartile range.
a Data presented are means (SD) for continuous variables and frequency (percentage) for categorical variables.
b Anthropometries measured at booking.
c Data presented are medians (IQR, lower quartile, upper quartile).
28
TABLE 6 Baseline characteristics for all infants completing MRI assessmenta
Imm-RDI/ Imm-RDI/
Infant sex, n (%)
Male 20 (58.8) 18 (64.3) 17 (50.0) 19 (51.4)
Gestational age (weeks), mean (SD) 28.0 (1.8) 28.0 (2.1) 28.4 (2.1) 27.7 (2.0)
Multiple births, n (%)

Yes 4 (11.8) 3 (10.7) 8 (23.5) 13 (35.1)
Birthweight (kg), mean (SD) 1.06 (0.29) 1.10 (0.32) 1.09 (0.28) 1.06 (0.29)
Birth length (cm), mean (SD) 35.5 (3.5); n = 28 35.1 (4.0); n = 24 35.6 (3.5); n = 24 34.9 (4.9); n = 27
Head circumference (cm), 25.3 (2.0); n = 34 25.6 (2.6); n = 26 25.5 (1.9); n = 32 25.7 (2.9); n = 34
mean (SD)
Birthweight (z-score), mean (SD) –0.1 (0.9) 0 (1.0) –0.2 (1.0) 0.1 (0.9)
Birth length (z-score), mean (SD) –0.9 (1.1); n = 28 –1.0 (1.3); n = 21 –1.0 (1.0); n = 23 –1.1 (1.4); n = 25
Head circumference (z-score), –0.5 (0.9); n = 34 –0.4 (1.0); n = 26 –0.7 (0.9); n = 32 –0.2 (1.7); n = 34
mean (SD)
Mother’s age (years), mean (SD) 32.6 (5.4); n = 34 30.3 (7.8); n = 26 32.2 (6.4); n = 33 32.7 (6.7); n = 34
Mother’s weight (kg),b mean (SD) 67.6 (14.5); n = 27 63.8 (11.5); n = 17 64.7 (13.3); n = 26 68.5 (16.1); n = 29
Mother’s height (cm), mean (SD) b

162.4 (7.2); n = 26 165.1 (7.3); n = 19 162.1 (9.1); n = 23 164.8 (9.1); n = 28
b
Father’s weight (kg), mean (SD) 82.3 (11.5); n = 21 81.5 (14.0); n = 15 84.1 (15.9); n = 22 87.8 (14.7); n = 28
Father’s height (cm),b mean (SD) 177.8 (6.1); n = 22 179.3 (7.4); n = 15 175.6 (10.2); n = 20 182.8 (9.6); n = 27
Mother’s ethnicity, n (%)

White 13 (38.2) 11 (39.3) 17 (50.0) 19 (51.4)
Asian 10 (29.4) 6 (21.4) 11 (32.4) 9 (24.3)
Black 5 (14.7) 10 (35.7) 5 (14.7) 6 (16.2)
Mixed 2 (5.9) 1 (3.6) 0 (0) 2 (5.4)
Other 3 (8.8) 0 (0) 1 (2.9) 1 (2.7)
Missing 1 (2.94) 0 (0) 0 (0) 0 (0)
Mode of delivery, n (%)

Vaginal 6 (17.6) 9 (32.1) 13 (38.2) 15 (40.5)
Elective caesarean 5 (14.7) 2 (7.1) 4 (11.8) 1 (2.7)
Emergency caesarean 23 (67.7) 17 (60.7) 17 (50.0) 21 (56.8)
Antenatal steroids, n (%)
Yes 24 (70.6) 21 (75.0) 26 (76.5) 30 (81.1)
No 5 (14.7) 5 (17.9) 6 (17.7) 4 (10.8)
Unknown 5 (14.7) 2 (7.1) 2 (5.9) 3 (8.1)

Time from birth to starting PN, 16.9 (10.5–22.3); 19.4 (12.1–22.3); 20.0 (12.4–23.5); 17.7 (13.2–22.4);
(hours),c median (IQR) n = 34 n = 28 n = 34 n = 37
IQR, interquartile range.
a Data presented are means (SD) for continuous variables and frequency (percentage) for categorical variables.
b Anthropometries measured at booking.
c Data presented are medians (IQR, lower quartile, upper quartile).
29
RESULTS
The time to achieve a milk intake of 150 ml/kg/day for 24 hours for all infants randomised was similar
across the four groups [Inc-AA/Intralipid, median 12 days, interquartile range (IQR) 9–17.5 days;
Inc-AA/SMOFlipid, median 11.5 days, IQR 9–16 days; Imm-RDI/Intralipid, median 11 days, IQR 10–14 days;
Imm-RDI/SMOFlipid, median 13 days, IQR 9.5 –18 days]. The length of hospital stay for all infants
randomised was similar across the four groups (Inc-AA/Intralipid, median 69.5 days, IQR 52–95 days;
Inc-AA/SMOFlipid, median 61 days, IQR 5–88 days; Imm-RDI/Intralipid, median 63 days, IQR 45–95 days;
Imm-RDI/SMOFlipid, median 66.5 days, IQR 44–98 days) (Tables 7 and 8).
TABLE 7 Parenteral nutrition details and blood culture results for all infants randomised
Imm-RDI/ Imm-RDI/
Route of PN administration, median (IQR)
Peripheral (days) 0 (0–2); n = 42 1 (0–5); n = 41 0.5 (0–2.5); n = 40 1 (0–3); n = 43
Central (days) 11.5 (8–20); n = 42 13 (8–20); n = 41 11 (9–15.5); n = 40 12 (9–18); n = 43
Days from delivery to achieve milk 12 (9–17.5); n = 32 11.5 (9–16); n = 28 11 (10–14); n = 30 13 (9.5–18); n = 36
intake of 150 ml/kg/day for 24 hours
Reason for stopping PN,a frequency (%)

Investigator’s decision 5 (11.9) 3 (7.1) 3 (7.3) 9 (20.9)
Investigator’s decision and 0 (0) 1 (2.4) 0 (0) 0 (0)

investigator manual
Investigator’s manual 2 (4.8) 1 (2.4) 2 (4.9) 2 (4.7)
Operational 2 (4.8) 1 (2.4) 2 (4.9) 3 (6.7)
Withdrawal 1 (2.4) 0 (0) 0 (0) 0 (0)

SAE 0 (0) 2 (4.8) 0 (0) 0 (0)
b
Positive blood culture, frequency (%) 15 13 9 14
Fungus 2 (13.3) 1 (7.7) 0 (0) 1 (7.1)
Gram-negative bacilli 5 (33.3) 4 (30.8) 3 (33.3) 1 (7.1)
Gram-positive bacilli 1 (6.7) 0 (0) 0 (0) 0 (0)
Gram-positive cocci CoNS 2 (13.3) 3 (23.1) 4 (44.4) 7 (50.0)
Gram-positive cocci excluding 3 (20.0) 4 (30.8) 1 (11.1) 5 (35.7)

CoNS
Gram-positive cocci not 2 (13.3) 1 (7.7) 1 (11.1) 0 (0)

specified
Positive blood cultures 8 9 8 8

(while on PN)b
Length of stay in hospital (days), 69.5 (52–95); n = 38 61 (45–88); n = 33 63 (45–95); n = 38 66.5 (44–98); n = 38
median (IQR)
CoNS, coagulase-negative staphylococci.
a There can be more than one reason for each infant.
b Growth of a known pathogen on culture; data presented are the number of infants who had at least one positive result.
30
TABLE 8 Parenteral nutrition details and blood culture results for all infants completing MRI assessment
Imm-RDI/ Imm-RDI/
Route of PN administration, median (IQR)
Peripheral (days) 0 (0–2) 1 (0–5.5) 1 (0–3) 1 (0–3)
Central (days) 11 (8–17) 13.5 (8.5–19.5) 10.5 (9–15) 12 (9–18)
Days from delivery to achieve milk 11 (9–16); n = 28 11.5 (9–16); n = 22 11 (10–13.5); n = 28 13 (10–18); n = 33
intake of 150 ml/kg/day for
24 hours, median (IQR)
Reason for stopping PN,a frequency (%)

Investigator’s decision 3 (8.8) 2 (7.1) 3 (8.8) 6 (16.2)
Investigator’s manual 2 (4.8) 1 (2.4) 2 (4.9) 1 (2.7)
Operational 2 (4.8) 1 (2.4) 2 (4.9) 3 (8.1)
Positive blood culture,b frequency (%) 9 4 5 12
Fungus 2 (22.2) 0 (0) 0 (0) 1 (8.3)

Gram-negative bacilli 3 (33.3) 3 (75.0) 3 (60.0) 0 (0)
Gram-positive bacilli 0 (0) 0 (0) 0 (0) 0 (0)
Gram-positive cocci CoNS 1 (11.1) 0 (0) 1 (20.0) 6 (50.0)
Gram-positive cocci excluding 2 (22.2) 0 (0) 0 (0) 5 (41.7)

CoNS
Gram-positive cocci not 1 (11.1) 1 (25.0) 1 (20.0) 0 (0)

specified
Positive blood cultures (while on PN)b 5 3 4 7

Length of stay in hospital (days), 69.5 (55–96); n = 34 59 (44–85); n = 28 60.5 (44–88); n = 34 67 (47–98.5); n = 36
median (IQR)
CoNS, coagulase-negative staphylococci.
a There can be more than one reason for each infant.
b Growth of a known pathogen on culture; data presented are the number of infants who had at least one positive result.
Nutritional intake from trial PN during the first week was similar across the four groups, except for
the intake of protein. On day 4, for all infants randomised, when infants randomised to Inc-AA intake
achieved the maximum intake, the protein intake was 2.5 g/kg and 2.6 g/kg in the Inc-AA/Intralipid and
Inc-AA/SMOFlipid groups, respectively, compared with 3.3 g/kg and 3.1 g/kg in the Imm-RDI/Intralipid and
Imm-RDI/SMOFlipid groups, respectively for all infants randomised (Table 9). Table 10 shows data for
babies who completed the MR scan. The median cumulative protein intake from trial PN during the first
2 weeks after birth for all randomised infants in the incremental arm was 22.4 g (IQR 16.0–28.4 g) and
20.9 g (IQR 15.3–28.4 g) in the Inc-AA/Intralipid and Inc-AA/SMOFlipid groups, respectively, compared with
25.9 g (IQR 22.6–32.5 g) and 29.5 g (IQR 23.2–37.2 g) in the Imm-RDI/Intralipid and Imm-RDI/SMOFlipid
groups, respectively. The median cumulative protein intake from all sources between birth and 34 weeks
postmenstrual age for all babies randomised was 138.2 g (IQR 109.9–170.7 g) and 119.0 g (IQR
91.1–161.0 g) in the Inc-AA/Intralipid and Inc-AA/SMOFlipid groups, respectively, compared with 124.8 g
(IQR 103.1–175.3 g) and 148.3 g (IQR 122.1–170.7 g) in the Imm-RDI/Intralipid and Imm-RDI/SMOFlipid
groups, respectively. Tables 11–13 show data of nutritional intake for all babies randomised. Tables 14 and
15 show data of nutritional intake for babies who completed the MR scan.
31
RESULTS
TABLE 9 Trial PN intake during the first 7 days for all infants randomised
Imm-RDI/ Imm-RDI/
Trial PN intake by day (N = 42) (N = 42) (N = 41) (N = 43)
Day 1,a mean (SD) n = 39 n = 34 n = 37 n = 41
Aqueous volume (ml/kg) 71.1 (36.2) 69.5 (34.3) 69.2 (36.7) 68.1 (35.6)
Lipid volume (ml/kg) 8.8 (5.2) 8.6 (4.6) 8.5 (5.3) 7.9 (4.5)
Protein (g/kg) 1.2 (0.7) 1.2 (0.6) 2.5 (1.3) 2.4 (1.3)
Carbohydrate (g/kg) 6.8 (3.5) 6.7 (3.3) 6.6 (3.5) 6.4 (3.4)
Fat (g/kg) 1.8 (1.0) 1.7 (0.9) 1.7 (1.1) 1.6 (0.9)
Day 2, mean (SD) n = 39 n = 38 n = 39 n = 42
Protein (g/kg) 2.1 (0.5) 1.9 (0.7) 3.1 (0.5) 3.1 (0.7)
Fat (g/kg) 3.0 (2.0) 2.6 (1.0) 2.7 (0.5) 2.8 (0.8)
Day 3, mean (SD) n = 39 n = 37 n = 38 n = 42
Protein (g/kg) 2.5 (0.5) 2.5 (0.7) 3.1 (0.7) 3.1 (0.7)
Fat (g/kg) 2.8 (0.6) 2.9 (0.7) 2.6 (0.8) 2.6 (0.9)
Day 4, mean (SD) n = 40 n = 37 n = 38 n = 40

Protein (g/kg) 2.5 (0.6) 2.6 (0.5) 3.3 (0.4) 3.1 (0.5)
Fat (g/kg) 2.6 (1.0) 2.7 (0.8) 2.8 (0.8) 2.5 (0.7)
Day 5, mean (SD) n = 42 n = 38 n = 38 n = 41
Protein (g/kg) 2.4 (0.7) 2.5 (0.5) 3.0 (0.7) 3.0 (0.6)
Fat (g/kg) 2.5 (0.8) 2.6 (0.7) 2.6 (0.7) 2.4 (0.8)
Day 6, mean (SD) n = 42 n = 38 n = 38 n = 41
Protein (g/kg) 2.3 (0.8) 2.3 (0.7) 2.8 (0.8) 2.9 (0.7)
Fat (g/kg) 2.3 (1.0) 2.4 (0.9) 2.3 (1.0) 2.3 (1.0)
32
TABLE 9 Trial PN intake during the first 7 days for all infants randomised (continued )
Imm-RDI/ Imm-RDI/
Day 7, mean (SD) n = 41 n = 36 n = 36 n = 38
Protein (g/kg) 2.1 (0.8) 2.1 (0.7) 2.7 (0.8) 2.7 (0.7)
Fat (g/kg) 2.0 (1.1) 1.9 (1.0) 2.3 (1.3) 2.1 (1.0)
a Day 1 is defined from birth to first 17.00.
TABLE 10 Trial PN intake during the first 7 days for all infants completing MRI assessment
Imm-RDI/ Imm-RDI/
Day 1,a mean (SD) n = 32 n = 25 n = 31 n = 36
Protein (g/kg) 1.2 (0.7) 1.2 (0.6) 2.5 (1.3) 2.4 (1.3)
Fat (g/kg) 1.7 (1.0) 1.7 (0.9) 1.8 (1.1) 1.6 (0.9)
Day 2, mean (SD) n = 33 n = 27 n = 34 n = 36
Protein (g/kg) 2.0 (0.4) 1.9 (0.5) 3.2 (0.5) 3.1 (0.6)
Fat (g/kg) 2.9 (2.1) 2.7 (0.9) 2.7 (0.5) 2.9 (0.7)
Day 3, mean (SD) n = 32 n = 28 n = 34 n = 36
Protein (g/kg) 2.6 (0.5) 2.5 (0.7) 3.1 (0.7) 3.2 (0.6)
Fat (g/kg) 2.8 (0.6) 2.8 (0.8) 2.6 (0.8) 2.7 (0.9)
continued
33
RESULTS
TABLE 10 Trial PN intake during the first 7 days for all infants completing MRI assessment (continued )
Imm-RDI/ Imm-RDI/
Day 4, mean (SD) n = 33 n = 28 n = 34 n = 35
Protein (g/kg) 2.5 (0.6) 2.6 (0.5) 3.3 (0.4) 3.2 (0.3)
Fat (g/kg) 2.5 (1.0) 2.7 (0.8) 2.8 (0.7) 2.6 (0.6)
Day 5, mean (SD) n = 34 n = 28 n = 34 n = 36
Protein (g/kg) 2.4 (0.7) 2.5 (0.6) 3.0 (0.7) 3.0 (0.6)
Fat (g/kg) 2.5 (0.9) 2.7 (0.7) 2.6 (0.7) 2.4 (0.9)
Day 6, mean (SD) n = 34 n = 28 n = 34 n = 35
Protein (g/kg) 2.3 (0.8) 2.4 (0.8) 2.7 (0.9) 3.0 (0.6)
Fat (g/kg) 2.3 (1.0) 2.4 (0.9) 2.3 (1.1) 2.3 (1.0)
Day 7, mean (SD) n = 33 n = 26 n = 32 n = 33
Protein (g/kg) 2.1 (0.8) 2.1 (0.8) 2.7 (0.8) 2.8 (0.7)
Fat (g/kg) 2.0 (1.1) 2.0 (1.1) 2.4 (1.4) 2.2 (1.0)
34
TABLE 11 Total nutrition intake during the first 7 days, 3 weeks, 4 weeks and by 34 weeks of gestational age for
all infants randomised
Imm-RDI/ Imm-RDI/
(N = 42) (N = 42) (N = 41) (N = 43)
Day 1,a mean (SD) n = 40 n = 36 n = 39 n = 41

Protein (g/kg) 1.3 (0.7) 1.2 (0.7) 2.4 (1.4) 2.5 (1.3)
Fat (g/kg) 1.8 (1.1) 1.8 (1.1) 1.7 (1.2) 1.7 (1.0)
Total energy (kcal/kg) 50.1 (26.5) 47.5 (25.8) 51.5 (31.4) 52.7 (28.3)
Day 2, mean (SD) n = 41 n = 39 n = 39 n = 42
Protein (g/kg) 2.1 (0.6) 2.0 (0.8) 3.3 (0.6) 3.2 (0.7)
Fat (g/kg) 3.1 (2.0) 2.9 (1.3) 3.0 (0.7) 3.1 (1.0)
Day 3, mean (SD) n = 41 n = 38 n = 38 n = 42
Protein (g/kg) 2.7 (0.8) 2.8 (0.9) 3.4 (0.8) 3.4 (0.8)
Fat (g/kg) 3.2 (1.0) 3.5 (1.3) 3.3 (1.1) 3.3 (1.3)
Day 4, mean (SD) n = 42 n = 38 n = 38 n = 43

Protein (g/kg) 2.7 (0.8) 3.1 (0.8) 3.7 (0.5) 3.4 (1.1)
Fat (g/kg) 3.3 (1.3) 3.7 (1.2) 3.7 (1.2) 3.3 (1.4)
Day 5, mean (SD) n = 42 n = 38 n = 38 n = 42
Protein (g/kg) 2.9 (0.8) 3.2 (0.7) 3.7 (0.8) 3.6 (0.7)
Fat (g/kg) 3.7 (1.4) 4.1 (1.3) 4.1 (1.5) 3.7 (1.2)
Day 6, mean (SD) n = 41 n = 38 n = 38 n = 43
Protein (g/kg) 3.0 (0.7) 3.2 (0.7) 3.6 (0.7) 3.6 (0.9)
Fat (g/kg) 3.9 (1.6) 4.4 (1.3) 4.3 (1.6) 3.9 (1.7)
continued
35
RESULTS
TABLE 11 Total nutrition intake during the first 7 days, 3 weeks, 4 weeks and by 34 weeks of gestational age for
all infants randomised (continued )
Imm-RDI/ Imm-RDI/
(N = 42) (N = 42) (N = 41) (N = 43)
Day 7, mean (SD) n = 42 n = 38 n = 38 n = 41
Protein (g/kg) 3.0 (0.7) 3.1 (0.6) 3.7 (0.6) 3.6 (0.8)
Fat (g/kg) 3.9 (1.6) 4.3 (1.6) 4.6 (1.7) 4.1 (1.8)
First 3 weeks, mean (SD) n = 42 n = 41 n = 40 n = 43
Protein (g) 61.4 (23.0) 58.8 (29.5) 70.7 (26.0) 71.2 (29.7)
Carbohydrate (g) 238.7 (90.1) 223.7 (109.8) 248.4 (90.4) 243.0 (103.6)
Fat (g) 105.5 (47.8) 99.2 (56.2) 109.4 (48.8) 104.1 (52.8)
Total energy (kcal) 2150 (875) 2023 (1057) 2261 (893) 2193 (1003)
First 4 weeks, mean (SD) n = 42 n = 41 n = 40 n = 43
Protein (g) 85.5 (32.0) 80.4 (42.7) 96.3 (38.1) 98.1 (41.0)
Carbohydrate (g) 339.5 (120.8) 313.7 (158.2) 349.5 (131.0) 346.9 (144.7)
Fat (g) 156.1 (65.8) 143.9 (80.7) 159.0 (72.0) 155.2 (76.2)
Until 34 weeks postmenstrual n = 42 n = 41 n = 40 n = 43

age, mean (SD)
Protein (g) 143.7 (55.9) 117.6 (57.9) 133.5 (51.9) 144.6 (47.3)
Carbohydrate (g) 574.7 (200.8) 481.9 (228.1) 525.7 (196.9) 544.8 (185.9)
Fat (g) 272.0 (98.7) 225.0 (111.2) 237.2 (94.1) 249.3 (97.0)
36
TABLE 12 Nutritional intake over the first 2 weeks for all infants randomiseda
Imm-RDI/ Imm-RDI/
(N = 42) (N = 42) (N = 41) (N = 43)
Had maternal expressed breast 40 (95.2) 38 (90.5) 38 (92.7) 42 (97.7)

milk, n (%)
Cumulative maternal expressed 0.76 (0.42–1.56); 0.94 (0.21–1.33); 0.68 (0.34–1.44); 0.59 (0.17–1.11);
breast milk (l), median (IQR) n = 40 n = 38 n = 38 n = 42
Had formula milk, n (%) 12 (28.6) 11 (26.2) 16 (39.0) 13 (30.2)
Cumulative formula intake (l), 0.03 (0.01–0.29); 0.16 (0.04–1.10); 0.47 (0.03–1.00); 0.37 (0.02–1.56);
median (IQR) n = 12 n = 11 n = 16 n = 13
Had trial PN, n (%) 42 (100.0) 41 (97.6) 40 (97.6) 43 (100.0)
Time on trial PN (days), median 11.0 (8.0–14.0); 12.0 (9.0–14.0); 11.0 (9.0–13.0); 12.0 (9.0–14.0);
(IQR) n = 42 n = 41 n = 40 n = 43
Cumulative trial PN intake, median (IQR)
Aqueous volume (l) 1.02 (0.74–1.29); 0.93 (0.70–1.30); 0.93 (0.81–1.18); 1.04 (0.81–1.36);
n = 42 n = 41 n = 40 n = 43
Lipid volume (l) 0.13 (0.09–0.17); 0.11 (0.08–0.17); 0.12 (0.10–0.15); 0.13 (0.09–0.18);
n = 42 n = 41 n = 40 n = 43
Protein (g) 22.4 (16.0–28.4); 20.9 (15.3–28.4); 25.9 (22.6–32.5); 29.5 (23.2–37.2);
n = 42 n = 41 n = 40 n = 43
Carbohydrate (g) 76.6 (57.7–96.1); 69.3 (53.1–96.6); 69.8 (61.1–87.8); 79.5 (62.7–100.4);
n = 42 n = 41 n = 40 n = 43
Fat (g) 23.6 (16.7–31.6); 21.0 (15.1–31.4); 21.4 (17.5–27.7); 24.6 (16.7–31.8);
n = 42 n = 41 n = 40 n = 43
Received non-trial PN, n (%) 3 (7.1) 1 (2.4) 1 (2.4) 3 (7.0)
Time on non-trial PN (days), 4.0 (3.5–4.0); n = 3 1.0 (1.0–1.0); n = 1 1.0 (1.0–1.0); n = 1 2.0 (1.5–5.0); n = 3
median (IQR)
Cumulative non-trial PN intake, median (IQR)
Aqueous volume (l) 0.46 (0.40–0.47); 0.02 (0.02–0.02); 0.07 (0.07–0.07); 0.37 (0.20–0.44);
n=3 n=1 n=1 n=3
Lipid volume (l) 0.02 (0.01–0.04); 0.00 (0.00–0.00); 0.00 (0.00–0.00); 0.03 (0.01–0.03);
n=3 n=1 n=1 n=3
Protein (g) 1.34 (1.17–1.54); 0.07 (0.07–0.07); 0.15 (0.15–0.15); 1.04 (0.57–1.21);
n=3 n=1 n=1 n=3
Carbohydrate (g) 42.57 (40.82–45.99); 2.47 (2.47–2.47); 3.35 (3.35–3.35); 29.75 (17.04–35.48);
n=3 n=1 n=1 n=3
Fat (g) 4.40 (2.20–7.90); 0.30 (0.30–0.30); 0.53 (0.53–0.53); 4.44 (2.61–5.85);
n=3 n=1 n=1 n=3
Cumulative non-trial and trial PN intake, median (IQR)
Aqueous volume (l) 1.04 (0.81–1.29); 0.93 (0.70–10.30); 0.94 (0.81–1.18); 1.04 (0.81–1.36);
n = 42 n = 41 n = 40 n = 43
Lipid volume (l) 0.13 (0.09–0.17); 0.11 (0.08–0.17); 0.12 (0.10–0.15); 0.13 (0.09–0.18);
n = 42 n = 41 n = 40 n = 43
Protein (g) 22.4 (16.0–28.4); 20.9 (15.3–28.4); 25.9 (22.6–32.5); 29.5 (23.2–37.2);
n = 42 n = 41 n = 40 n = 43
Carbohydrate (g) 78.0 (61.8–96.1); 69.3 (53.1–96.6); 70.3 (61.1–87.8); 79.5 (62.7–100.4);
n = 42 n = 41 n = 40 n = 43
Fat (g) 23.6 (16.8–31.6); 21.0 (15.1–31.4); 21.4 (17.5–27.7); 24.6 (16.7–31.8);
n = 42 n = 41 n = 40 n = 43
a Data presented are medians (IQR, lower quartile, upper quartile).
37
RESULTS
TABLE 13 Nutritional intake during the study period for all infants randomiseda
Imm-RDI/
Inc-AA/Intralipid Inc-AA/SMOFlipid Imm-RDI/Intralipid SMOFlipid
(N = 42) (N = 42) (N = 41) (N = 43)
Glucose and insulin, median (IQR)

Cumulative intravenous 15.2 (5.39–50.2); 18.1 (7.20–54.0); 11.8 (5.23–45.7); 22.2 (7.99–55.5);
dextrose (g) n = 42 n = 41 n = 40 n = 43
Received insulin, n (%) 11 (26.2) 12 (28.6) 5 (12.2) 10 (23.3)
Electrolytes, median (IQR)
Additional sodium (mmol) 17.54 (8.96–24.4); 7.46 (5.12–18.4); 12.18 (9.01–42.7); 10.01 (5.44–22.5);
n = 14 n = 14 n=9 n = 14
Additional potassium 9.97 (6.79–11.73); 2.73 (1.47–4.88); 2.93 (1.73–8.24); 3.72 (1.64–7.14);
(mmol) n=8 n=8 n=8 n = 10
Received donor milk, n (%) 15 (35.7) 12 (28.6) 13 (31.7) 17 (39.5)
Cumulative donor milk 0.38 (0.08–1.07); 0.30 (0.02–0.47); 0.38 (0.05–1.10); 0.19 (0.03–0.36);
(l, total volume per baby), n = 15 n = 12 n = 13 n = 17
median (IQR)
Received maternal breast 41 (97.6) 38 (90.5) 38 (92.7) 43 (100.0)
milk, n (%)
Cumulative maternal expressed breast milk (l, total volume per baby), median (IQR)
During trial PN phase 0.72 (0.45–0.88); 0.61 (0.43–0.81); 0.57 (0.30–0.81); 0.63 (0.26–0.89);
n = 41 n = 38 n = 38 n = 43
During non-trial PN phase 1.92 (0.38–4.82); 4.01 (1.83–5.08); 1.79 (0.44–4.51); 2.78 (0.70–4.50);
n=8 n=8 n=8 n=8
Over study period 8.05 (4.50–13.2); 7.16 (1.94–12.1); 6.74 (1.27–12.1); 7.61 (1.73–15.7);
n = 41 n = 38 n = 38 n = 43
Number of days having 13.0 (10.0–24.0); 23.0 (14.5–41.8); 14.0 (1.0–29.0); 27.0 (13.0–44.0);
Fortifier, median (IQR) n = 23 n = 12 n = 17 n = 17
Received formula milk, n (%) 30 (71.4) 23 (54.8) 29 (70.7) 29 (67.4)
Cumulative formula intake (l, total volume per baby), median (IQR)
n = 10 n = 10 n = 15 n = 14
During non-trial PN phase 2.05 (0.44–3.60); 2.09 (0.00–4.35); 0.78 (0.00–2.12); 2.30 (0.95–5.06);
n=6 n=4 n=7 n=6
n = 30 n = 23 n = 29 n = 29
Received trial PN, n (%) 42 (100.0) 41 (97.6) 40 (97.6) 43 (100.0)
Time on trial PN (days), 11.0 (8.00–16.8); 12.0 (9.00–17.0); 11.0 (9.75–13.2); 12.0 (9.00–17.0);
median (IQR) n = 42 n = 41 n = 40 n = 43
Cumulative trial PN intake
Aqueous volume (l) 1.04 (0.77–1.56); 0.97 (0.71–1.39); 0.95 (0.81–1.33); 1.19 (0.86–1.56);
n = 42 n = 41 n = 40 n = 43
Lipid volume (l) 0.13 (0.09–0.19); 0.13 (0.08–0.18); 0.12 (0.10–0.16); 0.14 (0.10–0.19);
n = 42 n = 41 n = 40 n = 43
During trial PN phase, median (IQR)
Protein (g) 22.8 (16.9–34.4); 21.6 (15.3–30.5); 26.7 (23.2–37.5); 33.5 (24.0–42.5);
n = 42 n = 41 n = 40 n = 43
Carbohydrate (g) 78.0 (57.7–116.0); 72.7 (55.0–104.8); 71.9 (62.6–101.3); 90.4 (64.8–114.6);
n = 42 n = 41 n = 40 n = 43
Fat (g) 23.9 (16.7–35.7); 23.6 (15.1–33.0); 21.6 (18.0–29.5); 25.0 (18.6–35.0);
n = 42 n = 41 n = 40 n = 43
38
TABLE 13 Nutritional intake during the study period for all infants randomiseda (continued )
Imm-RDI/
(N = 42) (N = 42) (N = 41) (N = 43)
Time on non-trial PN (days), 5.5 (3.25–19.8); 22.0 (14.50–28.5); 28.5 (9.50–58.5); 12.0 (3.00–15.8);
median (IQR) n = 10 n=8 n=8 n = 10
Aqueous volume (l) 0.47 (0.39–2.44); 2.12 (1.89–2.37); 2.91 (0.87–9.50); 1.23 (0.47–2.44);
n = 10 n=8 n=8 n = 10
Lipid volume (l) 0.06 (0.02–0.26); 0.33 (0.32–0.39); 0.38 (0.13–1.40); 0.16 (0.05–0.26);
n = 10 n=8 n=8 n = 10
Protein (g) 0.0 (0.0–0.00); 0.0 (0.0–0.00); 0.0 (0.0–0.04); 0.0 (0.0–0.07);
n = 10 n=8 n=8 n = 10
Carbohydrate (g) 0.0 (0.0–0.00); 0.0 (0.0–0.00); 0.0 (0.0–0.84); 0.0 (0.0–3.25);
n = 10 n=8 n=8 n = 10
Fat (g) 0.0 (0.0–0.00); 0.0 (0.0–0.00); 0.0 (0.0–0.13); 0.0 (0.0–0.59);
n = 10 n=8 n=8 n = 10
Over study period, median (IQR)
Protein (g) 1.65 (1.34–6.54); 7.32 (5.60–7.66); 10.76 (4.60–31.95); 3.09 (0.39–4.28);
n = 10 n=8 n=8 n = 10
Carbohydrate (g) 49.4 (41.1–220.6); 222.7 (188.6–265.4); 338.7 (166.7–1136.4); 108.6 (11.3–146.3);
n = 10 n=8 n=8 n = 10
Fat (g) 10.5 (4.04–46.5); 58.8 (56.76–69.1); 66.0 (23.68–247.1); 28.3 (8.78–46.4);
n = 10 n=8 n=8 n = 10
Aqueous volume (l) 1.12 (0.85–1.85); 1.30 (0.83–1.64); 1.06 (0.87–1.48); 1.38 (0.95–2.04);
n = 42 n = 41 n = 40 n = 43
Lipid volume (l) 0.17 (0.10–0.24); 0.16 (0.10–0.21); 0.13 (0.11–0.20); 0.18 (0.12–0.24);
n = 42 n = 41 n = 40 n = 43
Protein (g) 22.8 (16.9–34.4); 21.6 (15.3–30.5); 26.7 (23.2–37.5); 33.5 (24.0–42.5);
n = 42 n = 41 n = 40 n = 43
Carbohydrate (g) 78.0 (60.9–116.0); 72.7 (55.0–104.8); 71.9 (62.6–101.3); 90.4 (66.3–114.6);
n = 42 n = 41 n = 40 n = 43
Fat (g) 23.9 (16.8–35.7); 23.6 (15.1–33.0); 21.6 (18.0–29.5); 25.0 (18.6–35.0);
n = 42 n = 41 n = 40 n = 43
Protein (g) 23.7 (18.0–36.4); 23.9 (17.8–31.6); 29.5 (24.1–38.4); 33.5 (25.2–42.5);
n = 42 n = 41 n = 40 n = 43
Carbohydrate (g) 87.8 (64.5–139.6); 94.8 (64.0–136.5); 81.0 (65.1–113.4); 101.4 (70.8–149.7);
n = 42 n = 41 n = 40 n = 43
Fat (g) 31.3 (18.8–43.6); 28.2 (19.0–37.4); 23.2 (19.3–36.4); 32.0 (22.6–44.1);
n = 42 n = 41 n = 40 n = 43
Cumulative nutritional intake from birth until 34 weeks postmenstrual age (includes PN and milk intake), median (IQR)
Protein (g) 138.2 (109.9–170.7); 119.0 (91.1–161.0); 124.8 (103.1–175.3); 148.3 (122.1–170.7);
n = 42 n = 41 n = 40 n = 43
Carbohydrate (g) 565.6 (446.6–756.7); 557.9 (338.4–673.2); 508.0 (399.7–682.4); 590.4 (440.4–680.9);
n = 42 n = 41 n = 40 n = 43
Fat (g) 272.8 (217.2–346.5); 246.1 (157.6–306.1); 235.7 (182.7–295.7); 261.5 (201.9–313.0);
n = 42 n = 41 n = 40 n = 43
a Data presented are medians (IQR, lower quartile, upper quartile) for continuous variables and frequency (percentage) for
categorical variables.
39
RESULTS
TABLE 14 Nutritional intake over first 2 weeks for all infants completing MRI assessmenta
Inc-AA/
Inc-AA/Intralipid SMOFlipid Imm-RDI/Intralipid Imm-RDI/SMOFlipid
(N = 34) (N = 28) (N = 34) (N = 37)
Had maternal expressed breast 32 (94.1) 27 (96.4) 33 (97.1) 36 (97.3)

milk, n (%)
Cumulative maternal expressed 0.79 (0.60–1.63); 0.97 (0.54–1.62); 0.77 (0.42–1.54); 0.76 (0.29–1.14);
breast milk (l), median (IQR) n = 32 n = 27 n = 33 n = 36
Had formula milk, n (%) 10 (29.4) 9 (32.1) 14 (41.2) 11 (29.7)
Cumulative formula intake (l), 0.03 (0.01–0.19); 0.26 (0.09–1.11); 0.35 (0.03–1.12); 0.37 (0.05–1.32);
median (IQR) n = 10 n=9 n = 14 n = 11
Had trial PN, n (%) 34 (100.0) 28 (100.0) 34 (100.0) 37 (100.0)
Time on trial PN (days), median 11.0 (8.00–14.0); 12.0 (9.00–14.0); 11.0 (9.25–13.0); 12.0 (9.00–14.0);
(IQR) n = 34 n = 28 n = 34 n = 37
Aqueous volume (l) 1.02 (0.74–1.30); 1.03 (0.81–1.33); 0.94 (0.86–1.21); 1.07 (0.81–1.39);
n = 34 n = 28 n = 34 n = 37
Lipid volume (l) 0.13 (0.09–0.18); 0.13 (0.10–0.18); 0.12 (0.10–0.16); 0.13 (0.09–0.18);
n = 34 n = 28 n = 34 n = 37
Protein (g) 22.4 (16.0–28.9); 22.3 (17.6–29.1); 26.2 (23.9–33.2); 31.0 (23.0–37.9);
n = 34 n = 28 n = 34 n = 37
Carbohydrate (g) 76.6 (57.7–97.2); 79.1 (59.7–100.1); 70.6 (64.4–89.6); 83.7 (62.0–102.1);
n = 34 n = 28 n = 34 n = 37
Fat (g) 24.3 (17.0–32.2); 24.5 (18.1–32.2); 21.7 (18.1–28.4); 25.0 (17.1–32.9);
n = 34 n = 28 n = 34 n = 37
Time on non-trial PN (days) 4.0 (4.0–4.0); 1.0 (1.0–1.0); 1.0 (1.0–1.0); 5.0 (3.5–6.5);
n=2 n=1 n=1 n=2
Aqueous volume (l) 0.47 (0.47–0.48); 0.02 (0.02–0.02); 0.07 (0.07–0.07); 0.44 (0.41–0.47);
n=2 n=1 n=1 n=2
Lipid volume (l) 0.01 (0.01–0.02); 0.00 (0.00–0.00); 0.00 (0.00–0.00); 0.03 (0.03–0.04);
n=2 n=1 n=1 n=2
Protein (g) 1.37 (1.18–1.55); 0.07 (0.07–0.07); 0.15 (0.15–0.15); 1.21 (1.12–1.29);
n=2 n=1 n=1 n=2
Carbohydrate (g) 45.99 (44.28–47.69); 2.47 (2.47–2.47); 3.35 (3.35–3.35); 35.48 (32.62–38.34);
n=2 n=1 n=1 n=2
Fat (g) 2.20 (1.10–3.30); 0.30 (0.30–0.30); 0.53 (0.53–0.53); 5.85 (5.14–6.56);
n=2 n=1 n=1 n=2
Aqueous volume (l) 1.04 (0.82–1.30); 1.03 (0.81–1.33); 0.95 (0.86–1.21); 1.07 (0.81–1.39);
n = 34 n = 28 n = 34 n = 37
Lipid volume (l) 0.13 (0.09–0.18); 0.13 (0.10–0.18); 0.12 (0.10–0.16); 0.13 (0.09–0.18);
n = 34 n = 28 n = 34 n = 37
Protein (g) 22.4 (16.0–28.9); 22.3 (17.6–29.1); 26.2 (23.9–33.2); 31.0 (23.0–37.9);
n = 34 n = 28 n = 34 n = 37
Carbohydrate (g) 78.0 (63.0–97.2); 79.1 (60.5–100.1); 71.7 (64.4–89.6); 83.7 (62.0–102.1);
n = 34 n = 28 n = 34 n = 37
Fat (g) 24.3 (17.0–32.2); 24.5 (18.4–32.2); 21.7 (18.1–28.4); 25.0 (17.1–32.9);
n = 34 n = 28 n = 34 n = 37
a Data presented are medians (IQR, lower quartile, upper quartile).
40
TABLE 15 Nutritional intake during the study period for all infants completing MRI assessmenta
Imm-RDI/
(N = 34) (N = 28) (N = 34) (N = 37)
Glucose and insulin, median (IQR)
Cumulative 9.27 (4.56–54.8); 23.64 (7.04–75.1); 11.81 (5.68–25.9); 22.20 (8.89–53.4);

intravenous dextrose n = 34 n = 28 n = 34 n = 37
(g), median (IQR)
Insulin (number of 7 (20.6) 6 (21.4) 4 (11.8) 7 (18.9)

babies who received
insulin), n (%)
Electrolytes, median (IQR)
Additional sodium 20.93 (11.81–34.2); 11.13 (5.00–22.6); 9.85 (8.87–13.5); 10.01 (5.44–22.5);
(mmol) n = 10 n = 11 n=7 n = 14
Additional potassium 10.38 (6.45–12.29); 3.71 (1.62–5.58); 2.18 (1.46–5.00); 4.19 (1.60–7.85);
(mmol) n=7 n=7 n=7 n=9
Received donor milk, 11 (32.4) 9 (32.1) 12 (35.3) 15 (40.5)

n (%)
Cumulative donor milk 0.38 (0.08–1.07); 0.27 (0.02–0.44); 0.24 (0.04–1.15); 0.19 (0.03–0.84);
(l, total volume per n = 11 n=9 n = 12 n = 15
baby), median (IQR)
Received maternal 33 (97.1) 27 (96.4) 33 (97.1) 37 (100.0)
expressed breast milk,
n (%)
Cumulative maternal expressed breast milk (l), median (IQR)
n = 33 n = 27 n = 33 n = 37
During non-trial PN 1.75 (0.33–4.46); 4.16 (2.86–5.13); 1.79 (0.50–3.16); 2.78 (0.70–4.50);
phase n=5 n=7 n=6 n=8
n = 33 n = 27 n = 33 n = 37
Number of days having 13.0 (10.0–21.0); 20.0 (16.0–36.0); 13.0 (1.0–28.2); 25.5 (13.0–45.0);
fortifier, median (IQR) n = 21 n=9 n = 16 n = 16
Received formula milk, 25 (73.5) 19 (67.9) 25 (73.5) 26 (70.3)

n (%)
Cumulative formula intake (l), median (IQR)
n=9 n=8 n = 13 n = 11
During non-trial PN 2.34 (1.76–4.02); 2.09 (0.00–4.35); 0.78 (0.00–3.45); 2.30 (0.95–5.06);
phase n=5 n=4 n=5 n=6
n = 25 n = 19 n = 25 n = 26
Received trial PN 34 (100.0) 28 (100.0) 34 (100.0) 37 (100.0)

(number of babies),
n (%)
Time on trial PN (days), 11.0 (8.0–16.0); 12.5 (9.0–16.0); 11.0 (10.0–13.0); 13.0 (9.0–17.0);
median (IQR) n = 34 n = 28 n = 34 n = 37
continued
41
RESULTS
TABLE 15 Nutritional intake during the study period for all infants completing MRI assessmenta (continued )
Imm-RDI/
(N = 34) (N = 28) (N = 34) (N = 37)
Aqueous volume (l) 1.04 (0.77–1.56); 1.06 (0.83–1.39); 0.96 (0.86–1.32); 1.19 (0.82–1.57);
n = 34 n = 28 n = 34 n = 37
Lipid volume (l) 0.14 (0.09–0.19); 0.13 (0.10–0.19); 0.12 (0.10–0.16); 0.14 (0.10–0.20);
n = 34 n = 28 n = 34 n = 37
Protein (g) 22.8 (16.9–34.4); 23.0 (18.2–30.6); 27.2 (23.9–36.9); 33.5 (23.5–42.5);
n = 34 n = 28 n = 34 n = 37
Carbohydrate (g) 78.0 (57.7–118.5); 81.6 (62.1–105.2); 73.4 (64.4–99.6); 90.4 (63.4–114.6);
n = 34 n = 28 n = 34 n = 37
Fat (g) 25.0 (17.0–35.7); 24.9 (18.1–33.9); 21.7 (19.1–28.4); 25.0 (19.0–35.6);
n = 34 n = 28 n = 34 n = 37
Received non-trial PN 6 (17.6) 7 (25.0) 6 (17.6) 9 (24.3)

(number of babies),
n (%)
Time on non-trial PN 13.5 (4.0–27.5); 19.0 (13.0–26.0); 43.5 (25.8–69.5); 12.0 (3.0–16.0);
(days), median (IQR) n=6 n=7 n=6 n=9
Aqueous volume (l) 1.77 (0.40–3.76); 2.13 (1.86–2.48); 6.26 (2.52–10.83); 1.24 (0.74–2.76);
n=6 n=7 n=6 n=9
Lipid volume (l) 0.18 (0.02–0.58); 0.33 (0.32–0.35); 0.90 (0.32–1.58); 0.17 (0.07–0.29);
n=6 n=7 n=6 n=9
Protein (g) 0.0 (0.0–0.00); 0.0 (0.0–0.00); 0.0 (0.0–0.12); 0.0 (0.0–0.00);
n=6 n=7 n=6 n=9
Carbohydrate (g) 0.0 (0.0–0.00); 0.0 (0.0–0.00); 0.0 (0.0–2.51); 0.0 (0.0–0.00);
n=6 n=7 n=6 n=9
Fat (g) 0.0 (0.0–0.0); 0.0 (0.0–0.0); 0.0 (0.0–0.4); 0.0 (0.0–0.0);
n=6 n=7 n=6 n=9
Protein (g) 4.88 (1.44–12.63); 7.44 (4.66–7.69); 19.86 (6.68–37.40); 3.98 (1.07–4.28);
n=6 n=7 n=6 n=9
Carbohydrate (g) 159.4 (47.9–431.8); 242.2 (186.9–267.0); 701.5 (239.8–1339.7); 139.6 (30.7–147.6);
n=6 n=7 n=6 n=9
Fat (g) 31.2 (2.96–101.7); 58.2 (56.47–62.2); 159.0 (55.66–278.5); 30.4 (13.16–50.8);
n=6 n=7 n=6 n=9
Aqueous volume (l) 1.11 (0.85–1.85); 1.34 (0.96–1.80); 1.12 (0.89–1.43); 1.53 (0.97–2.05);
n = 34 n = 28 n = 34 n = 37
Lipid volume (l) 0.17 (0.10–0.25); 0.18 (0.13–0.25); 0.13 (0.11–0.20); 0.18 (0.13–0.25);
n = 34 n = 28 n = 34 n = 37
42
TABLE 15 Nutritional intake during the study period for all infants completing MRI assessmenta (continued )
Imm-RDI/
(N = 34) (N = 28) (N = 34) (N = 37)
Protein (g) 22.8 (16.9–34.4); 23.0 (18.2–30.6); 27.2 (23.9–36.9); 33.5 (23.5–42.5);
n = 34 n = 28 n = 34 n = 37
Carbohydrate (g) 78.0 (61.2–118.5); 81.6 (62.1–105.2); 73.4 (64.4–99.6); 92.7 (66.4–114.6);
n = 34 n = 28 n = 34 n = 37
Fat (g) 25.0 (17.0–35.7); 24.9 (18.4–33.9); 21.7 (19.1–28.4); 25.0 (19.0–35.6);
n = 34 n = 28 n = 34 n = 37

Protein (g) 23.7 (18.0–36.5); 26.1 (20.5–31.6); 30.3 (24.9–37.9); 34.4 (25.8–42.5);
n = 34 n = 28 n = 34 n = 37
Carbohydrate (g) 85.5 (64.5–139.6); 101.9 (72.2–145.1); 83.5 (67.2–110.6); 112.0 (71.9–152.7);
n = 34 n = 28 n = 34 n = 37
Fat (g) 30.3 (18.8–45.5); 31.8 (23.5–46.0); 23.6 (20.2–35.7); 32.9 (22.8–45.6);
n = 34 n = 28 n = 34 n = 37
Cumulative nutritional intake from birth until 34 weeks postmenstrual age (includes PN and milk intake), median (IQR)
Protein (g) 138.2 (115.8–185.8); 144.3 (115.5–165.4); 124.2 (104.5–176.6); 154.7 (133.1–171.7);
n = 34 n = 28 n = 34 n = 37
Carbohydrate (g) 565.6 (489.3–767.7); 596.2 (476.4–703.0); 508.0 (402.8–684.2); 607.7 (508.9–690.0);
n = 34 n = 28 n = 34 n = 37
Fat (g) 284.3 (235.7–370.6); 282.2 (233.8–320.1); 245.2 (185.1–295.2); 279.1 (213.3–323.2);
n = 34 n = 28 n = 34 n = 37
a Data presented are medians (IQR, lower quartile, upper quartile) for continuous variables and frequency (percentage) for
categorical variables.
There were no significant differences between the groups in the proportion of infants with abnormal
biochemical indices, namely serum glucose, worst base deficit in the previous 24 hours, total serum
bilirubin, conjugated bilirubin, serum cholesterol, serum triglycerides, serum sodium, serum potassium,
serum phosphate, serum calcium, serum creatinine and alanine transaminase. Tables 16–18 show infant
safety data, with Tables 16 and 18 including data for all infants randomised and Table 17 showing safety
data for only those infants who completed the MR scan.
However, there were significantly more infants with blood urea nitrogen levels > 7 mmol/l (50% and
47.6% in the groups Inc-AA/Intralipid and Inc-AA/SMOFlipid, respectively, vs. 70.7% and 79.1% in
Imm-RDI/Intralipid and Imm-RDI/SMOFlipid, respectively; p < 0.01) and > 10 mmol/l (14.3% and 21.4% in
the groups Inc-AA/Intralipid and Inc-AA/SMOFlipid respectively, vs. 43.9% and 53.5% in Imm-RDI/Intralipid
and Imm-RDI/SMOFlipid, respectively; p < 0.01).
There was a significant interaction (p = 0.05) between the two interventions for non-adipose
mass (Figure 7).
In relation to primary outcome measures, there were no significant differences in the quantity of
non-AT mass between the groups randomised to Inc-AA and the group randomised to the RDI of amino
acids (adjusted mean difference 1 g, 95% CI –108 g to 111 g; p = 0.98). For the lipid composition
intervention, there was no significant difference in IHCL content between the group randomised to receive
20% Intralipid and the group randomised to receive 20% SMOFlipid (adjusted geometric mean ratio of lipid
to water 1.1, 95% CI 0.8 to 1.6; p = 0.58). Primary and secondary outcomes for all infants randomised are
shown in Table 19. AT volumes are shown in Table 20.
43
RESULTS
TABLE 16 Safety data: summary of laboratory AEs by treatment for all infants randomiseda
Imm-RDI/ Imm-RDI/
Inc-AA/Intralipid, Inc-AA/SMOFlipid Intralipid SMOFlipid p-value for p-value
SpAE (N = 42) (N = 42) (N = 41) (N = 43) amino acid for lipid
Glucose, n (%)
Low (< 2.6 mmol/l) 12 (28.6) 19 (45.2) 15 (36.6) 16 (37.2) 1.0 0.32
High (> 15 mmol/l) 8 (19.0) 11 (26.2) 3 (7.3) 7 (16.3) 0.10 0.25
Worst base deficit in previous 24 hours, n (%)

> 15 mmol/l 5 (11.9) 3 (7.1) 5 (12.2) 8 (18.6) 0.35 1.0
Total serum bilirubin, n (%)
> 150 µmol/l 30 (71.4) 27 (64.3) 26 (63.4) 31 (72.1) 1.0 1.0
Conjugated bilirubin, n (%)
> 40 µmol/l 6 (14.3) 4 (9.5) 3 (7.3) 4 (9.3) 0.61 0.96
Cholesterol, n (%)
> 6 mmol/l 0 (0) 2 (4.8) 1 (2.4) 0 (0) 0.56 0.57
> 10 mmol/l 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) NA NA

Triglycerides, n (%)
> 2.5 mmol/l 15 (35.7) 13 (31.0) 14 (34.1) 12 (27.9) 0.87 0.55
> 5 mmol/l 2 (4.8) 2 (4.8) 1 (2.4) 2 (4.7) 0.70 0.72
Sodium, n (%)
Low (< 131 mmol/l) 9 (21.4) 7 (16.7) 10 (24.4) 9 (20.9) 0.70 0.65
High (> 150 mmol/l) 5 (11.9) 10 (23.8) 4 (9.8) 5 (11.6) 0.27 0.3
Potassium, n (%)
Low (< 3.2 mmol/l) 5 (11.9) 6 (14.3) 11 (26.8) 7 (16.3) 0.22 0.63
High (> 9 mmol/l) 0 (0) 0 (0) 0 (0) 0 (0) NA NA
Phosphate, n (%)
Low (< 1.5 mmol/l) 17 (40.5) 12 (28.6) 14 (34.1) 19 (44.2) 0.63 1.0
High (> 3 mmol/l) 4 (9.5) 5 (11.9) 5 (12.2) 4 (9.3) 1.0 1.0
Calcium, n (%)
Low (< 1 mmol/l) 0 (0) 1 (2.4) 0 (0) 2 (4.7) 0.56 0.08

High (> 3 mmol/l) 0 (0) 0 (0) 3 (7.3) 2 (4.7) 0.02 0.63
Urea, n (%)
Low (< 1.5 mmol/l) 13 (31.0) 11 (26.2) 5 (12.2) 8 (18.6) 0.15 0.55
High (> 7 mmol/l) 21 (50.0) 20 (47.6) 29 (70.7) 34 (79.1) < 0.01 0.78
High (> 10 mmol/l) 6 (14.3) 9 (21.4) 18 (43.9) 23 (53.5) < 0.01 0.30
Creatinine, n (%)
> 170 µmol/l 0 (0) 0 (0) 0 (0) 3 (7.0) 0.08 0.08

ALT, n (%)
> 60 IU/l 5 (11.9) 2 (4.8) 4 (9.8) 4 (9.3) 1.0 0.56
44
TABLE 16 Safety data: summary of laboratory AEs by treatment for all infants randomiseda (continued )
Imm-RDI/ Imm-RDI/
Zinc, n (%)
< 8 µmol/l 0 (0) 0 (0) 0 (0) 0 (0) NA NA
Copper, n (%)
< 2 µmol/l 0 (0) 0 (0) 0 (0) 0 (0) NA NA

Manganese, n (%)
> 30 nmol/l 0 (0) 0 (0) 0 (0) 0 (0) NA NA
Aluminium, n (%)
> 0.4 µmol/l 0 (0) 0 (0) 0 (0) 0 (0) NA NA
Selenium, n (%)
< 20 µg/l 0 (0) 1 (2.4) 0 (0) 0 (0) 0.32 0.32

ALT, alanine transferase; NA, not applicable.
a Biochemical indices were measured while the infant was on PN as part of routine monitoring whether on trial or
non-trial PN.
TABLE 17 Safety data: summary of laboratory AEs by treatment for all infants completing MRI assessment
Imm-RDI/ Imm-RDI/
Glucose, n (%)
Low (< 2.6 mmol/l) 8 (23.5) 12 (42.9) 13 (38.2) 14 (37.8) 1.0 0.32
High (> 15 mmol/l) 6 (17.6) 5 (17.9) 3 (8.8) 5 (13.5) 0.1 0.25
Worst base deficit in previous 24 hours, n (%)

> 15 mmol/l 2 (5.9) 0 (0.0) 2 (5.9) 6 (16.2) 0.35 1.0
Total serum bilirubin, n (%)
> 150 µmol/l 24 (70.6) 21 (75.0) 23 (67.6) 28 (75.7) 1.0 1.0
Conjugated bilirubin, n (%)
> 40 µmol/l 4 (11.8) 3 (10.7) 2 (5.9) 3 (8.1) 0.61 0.96
Cholesterol, n (%)
> 6 mmol/l 0 (0.0) 1 (3.6) 1 (2.9) 0 (0.0) 0.56 0.57
> 10 mmol/l 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) NA NA

Triglycerides, n (%)
> 2.5 mmol/l 10 (29.4) 7 (25.0) 11 (32.4) 10 (27.0) 0.87 0.55
> 5 mmol/l 2 (5.9) 0 (0.0) 1 (2.9) 2 (5.4) 0.7 0.72

continued
45
RESULTS
TABLE 17 Safety data: summary of laboratory AEs by treatment for all infants completing
MRI assessment (continued )
Imm-RDI/ Imm-RDI/
Sodium, n (%)
Low (< 131 mmol/l) 7 (20.6) 3 (10.7) 9 (26.5) 7 (18.9) 0.7 0.65
High (> 150 mmol/l) 3 (8.8) 3 (10.7) 4 (11.8) 4 (10.8) 0.27 0.3
Potassium, n (%)
Low (< 3.2 mmol/l) 2 (5.9) 4 (14.3) 9 (26.5) 6 (16.2) 0.22 0.63
High (> 9 mmol/l) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) NA NA
Phosphate, n (%)
Low (< 1.5 mmol/l) 12 (35.3) 8 (28.6) 13 (38.2) 17 (45.9) 0.63 1.0
High (> 3 mmol/l) 4 (11.8) 3 (10.7) 4 (11.8) 3 (8.1) 1.0 1.0
Calcium, n (%)
Low (< 1 mmol/l) 0 (0.0) 0 (0.0) 0 (0.0) 2 (5.4) 0.56 0.08
High (> 3 mmol/l) 0 (0.0) 0 (0.0) 3 (8.8) 1 (2.7) 0.02 0.63
Urea, n (%)
Low (< 1.5 mmol/l) 10 (29.4) 10 (35.7) 5 (14.7) 7 (18.9) 0.15 0.55
High (> 7 mmol/l) 14 (41.2) 14 (50.0) 25 (73.5) 29 (78.4) < 0.01 0.78
High (> 10 mmol/l) 1 (2.9) 4 (14.3) 16 (47.1) 18 (48.6) < 0.01 0.3
Creatinine, n (%)
> 170 µmol/l 0 (0.0) 0 (0.0) 0 (0.0) 2 (5.4) 0.08 0.08
ALT, n (%)
> 60 IU/l 3 (8.8) 2 (7.1) 3 (8.8) 2 (5.4) 1.0 0.56
Zinc, n (%)
< 8 µmol/l 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) NA NA
Copper, n (%)
< 2 µmol/l 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) NA NA
Manganese, n (%)
> 30 nmol/l 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) NA NA
Aluminium, n (%)
> 0.4 µmol/l 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) NA NA

Selenium, n (%)
< 20 µg/l 0 (0.0) 1 (3.6) 0 (0.0) 0 (0.0) 0.32 0.32

ALT, alanine transferase; NA, not applicable.
46
TABLE 18 Safety data: summary of SAEs by treatmenta

Variable (N = 42) (N = 42) (N = 41) (N = 43)
Number of infants who 6 (14.3) 12 (28.6) 8 (19.5) 9 (20.9)

experience a SAE, n (%)
SAE classification, n (%)b
Death, n (%) 3 (7.1) 7 (16.7) 3 (7.3) 3 (7.0)
Life-threatening, n (%) 3 (7.1) 3 (7.1) 4 (9.8) 4 (9.3)
Prolongation of existing 2 (4.8) 2 (4.8) 1 (2.4) 3 (7.0)
inpatient hospitalisation, n (%)
Persistent or significant 0 (0.0) 1 (2.4) 0 (0.0) 1 (2.3)
disability or incapacity, n (%)
Sepsis (diagnosis on SAE form), 2 (4.8) 6 (14.3) 1 (2.4) 1 (2.3)
n (%)c
Necrotising enterocolitis, n (%)c 3 (7.1) 4 (9.5) 1 (2.4) 4 (9.3)
a All SAEs were classified by local investigators on the SAE reporting form.
b One infant may have more than one SAE.
c Growth of known pathogen on culture; data presented are number of infants who had at least one positive result.
(a)
2600
Non-adipose body mass (g)
2400
Lipid
Intralipid
SMOFlipid
2200
2000
Inc-AA Imm-RDI
(b)
0.0
IHCL on natural log-scale
–0.5
Lipid
Intralipid
SMOFlipid
–1.0
–1.5
Inc-AA Imm-RDI
FIGURE 7 Means (95% CIs) of Inc-AA and Imm-RDI in two lipid subgroups for (a) non-adipose body mass; and
(b) IHCL content on a log-scale.
47
48
TABLE 19 Baseline characteristics and trial outcomes (all infants completing primary outcome assessments)
RESULTS
Adjusted mean Adjusted mean

difference difference
Inc-AA/Intralipid Inc-AA/SMOFlipid Imm-RDI/Intralipid Imm-RDI/SMOFlipid (Imm-RDI minus (SMOFlipid minus Interaction
(N = 34) (N = 28) (N = 34) (N = 37) Inc-AA)a Intralipid)a (p-value)
Gestational age (weeks), 28.0 (27.3 to 28.6) 28.0 (27.2 to 28.9) 28.4 (27.7 to 29.2) 27.7 (27.1 to 28.4) – – –
mean (95% CI)
Birthweight (g), mean 1064 (962 to 1166) 1103 (979 to 1226) 1090 (993 to 1186) 1059 (962 to 1155) – – –
(95% CI)
Sex, male (%), mean 58.8 (40.7 to 75.4) 64.3 (44.1 to 81.4) 50.0 (33.4 to 67.6) 51.4 (34.4 to 67.5) – – –

(95% CI)
Age (weeks) at scan, mean 12.5 (11.3 to 13.7) 12.4 (10.9 to 14.0) 12.1 (10.8 to 13.4) 13.3 (12.0 to 14.5) – – –
(95% CI)
Primary outcomes
Non-adipose mass (g), 2450 (2246 to 2655) 2337 (2164 to 2510) 2344 (2244 to 2444) 2485 (2327 to 2643) 1 (–108 to 111); –41 (–150 to 68); 216 (0 to 432);
mean (95% CI) p = 0.98 p = 0.46 p = 0.05
IHCL, mean (95% CI)b 0.6 (0.4 to 0.9); 0.7 (0.5 to 1.0); 0.5 (0.4 to 0.6); 0.5 (0.3 to 0.7); 0.7 (0.5 to 1.1); 1.1 (0.8 to 1.6); 0.8 (0.4 to 1.7);
n = 34 n = 28 n = 34 n = 36 n = 132 n = 132 n = 132
p = 0.11 p = 0.58 p = 0.53
Secondary outcomes
Total cerebral volume 468 (419 to 518); 480 (425 to 534); 468 (414 to 523); 511 (440 to 583); 15 (–42 to 71); 24 (–32 to 80); –26 (–142 to 90);
(cm3), mean (95% CI)c n = 13 n = 10 n = 11 n = 15 n = 49 n = 49 n = 49
p = 0.61 p = 0.40 p = 0.66
Whole-brain volume 339 (304 to 373); 352 (319 to 385); 344 (296 to 393); 365 (321 to 410); 9 (–29 to 47); 14 (–24 to 52); –29 (–107 to 49);
(cm3), mean (95% CI)d n = 13 n = 10 n = 11 n = 15 n = 49 n = 49 n = 49
p = 0.64 p = 0.47 p = 0.46
Posterior fossa volume 30 (26 to 33); 31 (28 to 34); 30 (27 to 34); 35 (29 to 38); 1.44 (–1.99 to 4.87); 2 (–2 to 5); –2 (–9 to 5);
(cm3), mean (95% CI)e n = 13 n = 10 n = 11 n = 15 n = 49 n = 49 n = 49
p = 0.41 p = 0.35 p = 0.64


QUICKI, mean (95% CI)52 0.18 (0.17 to 0.19); 0.19 (0.18 to 0.20); 0.19 (0.18 to 0.20); 0.18 (0.17 to 0.20); 0.01 (0 to 0.02); 0.01 (–0.01 to 0.02); –0.01 (–0.04 to 0.02);
DOI: 10.3310/eme03020
n = 11 n=6 n = 11 n = 11 n = 39 n = 39 n = 39
p = 0.20 p = 0.28 p = 0.46
Weight (g), mean 3060 (2780 to 3340) 2924 (2686 to 3162) 2932 (2780 to 3085) 3151 (2934 to 3368) 17 (–136 to 170); –35 (–187 to 117); 293 (–8 to 593);
(95% CI) p = 0.83 p = 0.65 p = 0.06
Length (cm), mean 47.7 (46.4 to 49.0) 48.0 (46.6 to 49.4) 48.2 (47.4 to 49.0) 49.1 (47.8 to 50.3) 0.5 (–0.3 to 1.3); 0.2 (–0.6 to 1.0); 0.5 (–1.1 to 2.1);
(95% CI) p = 0.20 p = 0.56 p = 0.56
Head circumference (cm), 36.0 (34.9 to 37.1) 35.3 (34.6 to 36.0) 34.8 (34.3 to 35.3) 35.2 (34.5 to 35.9) –0.8 (–1.5 to –0.1); –0.2 (–0.9 to 0.5); 1.1 (–0.2 to 2.5);
mean (95% CI) p = 0.02 p = 0.56 p = 0.09
Superficial subcutaneous 515 (437 to 593) 495 (431 to 559) 493 (431 to 554) 564 (499 to 629) 12 (–44 to 68); 9 (–46 to 64); 73 (–38 to 183);
AT (g), mean (95% CI) p = 0.67 p = 0.75 p = 0.20
Internal AT (g), mean 67.2 (55.5 to 79.0) 65.0 (52.4 to 77.5) 69.1 (57.2 to 81.0) 71.2 (59.8 to 82.5) 2.5 (–7.5 to 12.6); –3.4 (–13.4 to 6.6); 0.1 (–19.9 to 20.1);
(95% CI) p = 0.62 p = 0.50 p = 0.99
Deep subcutaneous 14.2 (11.0 to 17.3) 13.0 (10.7 to 15.2) 14.9 (12.3 to 17.5) 17.8 (14.8 to 20.7) 2.0 (–0.5 to 4.4); 0.4 (–2.0 to 2.8); 3.5 (–1.3 to 8.3);
abdominal AT (g), mean p = 0.11 p = 0.74 p = 0.15
(95% CI)
Internal abdominal AT (g), 14.8 (12.2 to 17.3) 14.1 (11.0 to 17.2) 15.9 (12.8 to 18.9) 16.5 (13.6 to 19.3) 1.4 (–1.2 to 4.1); –0.8 (–3.4 to 1.8); 0.5 (–4.8 to 5.7);
mean (95% CI) p = 0.28 p = 0.56 p = 0.86
Superficial subcutaneous 87.0 (72.4 to 101.6) 84.5 (72.8 to 96.2) 85.2 (73.8 to 96.5) 102.6 (87.2 to 118.1) 5.2 (–6.5 to 17.0); 5.0 (–6.7 to 16.6); 16.3 (–7.0 to 39.6);
AT tissue (g), mean p = 0.38 p = 0.40 p = 0.17
(95% CI)
Total AT (g), mean 610 (518 to 702) 587 (509 to 664) 589 (514 to 663) 666 (589 to 743) 16 (–51 to 82); 6 (–60 to 72); 77 (–55 to 208);
(95% CI) p = 0.64 p = 0.85 p = 0.25
Total AT as a percentage 19.4 (17.9 to 20.9) 19.7 (18.4 to 21.0) 19.6 (17.8 to 21.4) 20.8 (19.4 to 22.3) 0 (–0.01 to 0.02); 0.01 (–0.01 to 0.02); 0.01 (–0.02 to 0.03);
of body weight (%), p = 0.56 p = 0.45 p = 0.72
mean (95% CI)
continued
EFFICACY AND MECHANISM EVALUATION 2016 VOL. 3 NO. 2
49
50
RESULTS
TABLE 19 Baseline characteristics and trial outcomes (all infants completing primary outcome assessments) (continued )


Safety outcomesf
Triglycerides > 2.5 mmol/l 29.4 (15.1 to 47.5) 25.0 (10.7, 44.9) 32.4 (17.4 to 50.5) 27.0 (13.8 to 44.1) 1.15 (0.48 to 2.74) 0.68 (0.28 to 1.62) 0.71 (0.12 to 4.06)
g
(%), mean (95% CI) p = 0.76 p = 0.38 p = 0.70
Total serum bilirubin 70.6 (52.5 to 84.9) 75.0 (55.1 to 89.3) 67.6 (49.5 to 82.6) 75.7 (58.8 to 88.2) 0.92 (0.41 to 2.04) 1.32 (0.59 to 2.94) 1.29 (0.26 to 6.47)
> 150 µmol/l (%), mean p = 0.83 p = 0.50 p = 0.75
g
(95% CI)
Conjugated bilirubin 11.8 (3.3 to 27.5) 10.7 (2.3 to 28.2) 5.9 (0.7 to 19.7) 8.1 (1.7 to 21.9) 0.47 (0.12 to 1.85) 0.93 (0.24 to 3.54) 1.65 (0.11 to 25.34)
> 40 µmol/l (%), mean p = 0.28 p = 0.92 p = 0.72
g
(95% CI)
ALT > 60 IU/l (%), mean 8.8 (1.9 to 23.7) 7.1 (0.9 to 23.5) 8.8 (1.9 to 23.7) 5.4 (0.7 to 18.2) 0.99 (0.22 to 4.41) 0.45 (0.09 to 2.12) 0.59 (0.03 to 12.69)
g
(95% CI) p = 0.99 p = 0.31 p = 0.74
ALT, alanine transferase.
a Adjusted for: age at MRI, sex, gestational age, birthweight and centre; body mass components are derived from body mass volumes.
b Log-transformation was used in the regression model with the results transformed back from the log-scale.
c Total of basal ganglia, thalami (deep grey matter), cerebrospinal fluid, grey matter, white matter and lateral ventricles volumes.
d Total of basal ganglia, thalami (deep grey matter), grey matter and white matter.
e Total of cerebellum and brainstem volumes.
f Percentages of babies.
g Logistic regression was used for modelling and odds ratios are reported.
TABLE 20 Adipose tissue compartments in litres for all infants completing MRI assessment

DOI: 10.3310/eme03020
Adjusted mean
Adjusted mean difference
difference (20% SMOFlipid –
Inc-AA/20% Inc-AA/20% Imm-RDI/20% Imm-RDI/20% (Imm-RDI – Inc-AA),a 20% Intralipid),a Interaction,
Intralipid (n = 34) SMOFlipid (n = 28) Intralipid (n = 34) SMOFlipid (n = 37) p-value p-value p-value
Total internal AT (l) 0.07 (0.06 to 0.09) 0.07 (0.06 to 0.09) 0.08 (0.06 to 0.09) 0.08 (0.07 to 0.09) 0 (–0.01 to 0.01); 0 (–0.01 to 0.01); 0 (–0.02 to 0.02);
mean (95% CI) p = 0.62 p = 0.50 p = 0.99
Superficial AT (l) 0.57 (0.49 to 0.66) 0.55 (0.48 to 0.62) 0.55 (0.48 to 0.62) 0.63 (0.55 to 0.7) 0.01 (–0.05 to 0.08); 0.01 (–0.05 to 0.07); 0.08 (–0.04 to 0.2);
mean (95% CI) p = 0.67 p = 0.75 p = 0.20
Deep subcutaneous 0.03 (0.03 to 0.04) 0.03 (0.02 to 0.03) 0.03 (0.03 to 0.03) 0.03 (0.03 to 0.04) 0 (0 to 0); p = 0.55 0 (0 to 0); p = 0.71 0 (0 to 0.01);
AT (l) mean p = 0.24
(95% CI)
Internal 0.02 (0.01 to 0.02) 0.02 (0.01 to 0.02) 0.02 (0.01 to 0.02) 0.02 (0.02 to 0.02) 0 (0 to 0); p = 0.28 0 (0 to 0); p = 0.55 0 (–0.01 to 0.01);
abdominal AT (l) p = 0.86
mean (95% CI)
Superficial 0.1 (0.08 to 0.11) 0.09 (0.08 to 0.11) 0.09 (0.08 to 0.11) 0.11 (0.1 to 0.13) 0.01 (–0.01 to 0.02); 0.01 (–0.01 to 0.02); 0.02 (–0.01 to 0.04);
subcutaneous p = 0.38 p = 0.40 p = 0.17
abdominal AT (l)
mean (95% CI)
Deep subcutaneous 0.02 (0.01 to 0.02) 0.01 (0.01 to 0.02) 0.02 (0.01 to 0.02) 0.02 (0.02 to 0.02) 0 (0 to 0); p = 0.11 0 (0 to 0); p = 0.74 0 (0 to 0.01);
abdominal AT (l) p = 0.15
mean (95% CI)
Total AT (l) 0.68 (0.58 to 0.78) 0.65 (0.57 to 0.74) 0.65 (0.57 to 0.74) 0.74 (0.65 to 0.83) 0.02 (–0.06 to 0.09); 0.01 (–0.07 to 0.08); 0.09 (–0.06 to 0.23);
p = 0.64 p = 0.85 p = 0.25
a Adjusted for: age at scan, sex, gestational age, birthweight and centre.
51
RESULTS
There were no significant differences in secondary outcome measures of the quantity and distribution
of AT, measure of insulin sensitivity (as measured by the QUICKI), total cerebral volume, whole-brain volume,
weight and length at term age equivalent. There was, however, a significant difference in the mean head
circumference at term age equivalent between the group randomised to receive Inc-AA and that randomised to
receive the Imm-RDI (adjusted mean difference –0.8 cm, 95% CI –1.5 to –0.1 cm; p = 0.02).
In a secondary analysis, after adjusting for covariates, there were no significant differences in primary
outcomes (Tables 21 and 22).
TABLE 21 Summary of the covariates for secondary analysis
Inc-AA/Intralipid Inc-AA/SMOFlipid RDI/Intralipid RDI/SMOFlipid

Covariates (N = 34) (N = 28) (N = 34) (N = 37)
Proportion of level 1 care, 0.11(0.07–0.21) 0.11 (0.06–0.35) 0.12 (0.08–0.31) 0.16 (0.07–0.32)
median (IQR)
Proportion of level 2 care, 0.42 (0.21–0.59) 0.30 (0.20–0.46) 0.29 (0.20–0.44) 0.31 (0.21–0.42)
median (IQR)
Proportion of MEBM in all 0.80 (0.44–1.00); 0.95 (0.28–1.00); 0.65 (0.10–0.99); 0.70 (0.26–1.00);
milk intake, median (IQR) n = 33 n = 27 n = 33 n = 37
Trial-PN phase protein 31.5 (24.2–46.2) 34.4 (27.8–44.5) 42.8 (36.1–54.1) 48.5 (36.2–63.9)
intake (g), median (IQR)
Trial-PN phase carbohydrate 124.6 (98.5–178.7) 139.6 (110.8–173.2) 129.9 (109.6–158.9) 148.2 (111.8–182.6)
Trial-PN phase fat intake 52.8 (35.5–68.6) 51.6 (42.7–68.6) 51.8 (38.1–59.5) 53.1 (38.2–70.2)
(g), median (IQR)
Post-trial-PN phase protein 213.0 (168.7–449.4) 213.4 (154.0–302.2) 234.5 (155.3–335.4) 257.4 (187.3–365.2)
Post-trial-PN phase 969.0 (699.6–1786.1) 867.7 (643.5–1196.1) 1042.9 (644.8–1390.7) 1080.4 (696.3–1531.5)
carbohydrate intake (g),
median (IQR)
Post-trial-PN phase fat 479.8 (375.0–942.3) 434.8 (329.5–584.4) 515.9 (346.9–613.8) 550.5 (374.9–779.6)
MEBM, maternal expressed breast milk.
52
DOI: 10.3310/eme03020
TABLE 22 Primary outcomes (pairwise comparisons)
Adjusted mean Adjusted mean Adjusted mean Adjusted mean

difference difference difference difference
(Imm-RDI/Inc-AA)a (SMOFlipid/Intralipid)a Interaction (p–value) (Imm-RDI/Inc-AA)b (SMOFlipid/Intralipid)b Interaction (p-value)
Non-adipose mass (g), 1 (–108 to 111); –41 (–150 to 68); 216 (0 to 432); p = 0.05 –44 (–226 to 139), –14 (–114 to 86), 184 (–22 to 390);
difference (95% CI) p = 0.98 p = 0.46 n = 130; p = 0.64 n = 130; p = 0.78 p = 0.08
IHCL content, difference 0.7 (0.5 to 1.1), 1.1 (0.8 to 1.6), 0.8 (0.4 to 1.7), n=132; 0.81 (0.37 to 1.80), 0.89 (0.61 to 1.31) 0.86 (0.39 to 1.92),
(95% CI)c n = 132; p = 0.11 n = 132; p = 0.58 p=0.53 n = 129; p = 0.61 n = 129; p = 0.57 n = 129; p = 0.71
a Adjusted for age at MRI, sex, gestational age, birthweight and centre.
b Adjusted for age at scan, sex, gestational age, birthweight z-score (UK 1990 growth data),57 centre, level of care (% time spent receiving intensive and high-dependency care between
birth and assessment32) and nutritional intake.
c Log-transformation was used in the regression model, the results transformed back from the logarithmic scale and presented as the ratio of intervention to control.
53
RESULTS
Weight gain over the study period was similar across groups (Figure 8).
Trial PN protein intake was higher in the Imm-RDI arms in the first 2 weeks (Figure 9).
Carbohydrate, lipid and energy intakes were similar across all four groups (Figures 10–12).
Macronutrient and energy intake from milk and PN from the end of the second week onwards did not
differ between groups (Figures 13–16).
3 Inc-AA/Intralipid
Weight (kg)
Inc-AA/SMOFlipid
Imm-RDI/Intralipid
2 Imm-RDI/SMOFlipid
27 29 31 33 35 37 39
Postmenstrual age (weeks)
FIGURE 8 Time trend for babies’ weights across four groups.
4
Mean protein (g/kg)
3 Inc-AA/Intralipid
Inc-AA/SMOFlipid
Imm-RDI/Intralipid
2 Imm-RDI/SMOFlipid
0 2 4 6 8 10 12 14
Age (days)
FIGURE 9 Daily protein intake from all sources in the first 2 weeks across four groups.
54
15
Mean carbohydrate (g/kg)
10 Inc-AA/Intralipid
Inc-AA/SMOFlipid
Imm-RDI/Intralipid
Imm-RDI/SMOFlipid
5
0 2 4 6 8 10 12 14
Age (days)
FIGURE 10 Daily carbohydrate intake from all sources in the first 2 weeks across all four groups.
6
Mean fat (g/kg)
Inc-AA/Intralipid
4 Inc-AA/SMOFlipid
Imm-RDI/Intralipid
Imm-RDI/SMOFlipid
0 2 4 6 8 10 12 14
Age (days)
FIGURE 11 Daily fat intake from all sources in the first 2 weeks across all four groups.
55
RESULTS
150
Mean energy (kcal/kg)
100
Inc-AA/Intralipid
Inc-AA/SMOFlipid
Imm-RDI/Intralipid
Imm-RDI/SMOFlipid
50
0 2 4 6 8 10 12 14
Age (days)
FIGURE 12 Daily energy intake from all sources in the first 2 postnatal weeks.
4
Mean protein (g/kg)
3 Inc-AA/Intralipid
Inc-AA/SMOFlipid
Imm-RDI/Intralipid
2 Imm-RDI/SMOFlipid
15 22 29 36 43 50 57 64 71 78 85
Age (days)
FIGURE 13 Daily protein intake from all sources after first 2 weeks across all four groups.
56
15
Mean carbohydrate (g/kg)
10
Inc-AA/Intralipid
Inc-AA/SMOFlipid
Imm-RDI/Intralipid
Imm-RDI/SMOFlipid
5
15 22 29 36 43 50 57 64 71 78 85
Age (days)
FIGURE 14 Daily carbohydrate intake from all sources after first 2 weeks across all four groups.
4
Mean fat (g/kg)
Inc-AA/Intralipid
Inc-AA/SMOFlipid
Imm-RDI/Intralipid
Imm-RDI/SMOFlipid
2
15 22 29 36 43 50 57 64 71 78 85
Age (days)
FIGURE 15 Daily fat intake from all sources after first 2 weeks across four groups.
57
RESULTS
150
Mean energy (kcal/kg)
100
Inc-AA/Intralipid
Inc-AA/SMOFlipid
Imm-RDI/Intralipid
Imm-RDI/SMOFlipid
50
15 22 29 36 43 50 57 64 71 78 85
Age (days)
FIGURE 16 Daily energy intake from all sources after the first 2 postnatal weeks.
58
Chapter 5 Discussion
T he key strength of the NEON trial was the excellent trial protocol adherence, including the introduction
of milk feeds within 24 hours of birth and a prespecified approach to the management of electrolyte
disturbances despite clinician blinding to group allocation. The need for central venous access can limit
early commencement, hence the composition of trial PN permitted delivery by peripheral vein. Both
gestational age strata (23–26 weeks and 27–31 weeks) were broadly equal across groups making the trial
results applicable to the most immature infants. The NEON trial was adequately powered, as the CIs for
the mean differences in non-adipose mass and SMOFlipid-to-Intralipid IHCL ratio exclude, respectively, the
prespecified difference of 200 g,19 and decrease of 40%. As we know of no biological reason for lipid type
to influence the quantity of non-AT we consider it likely that the between-intervention interaction we
detected is due to chance.
This is the first RCT of the impact of amino acids intake in PN on body composition in extremely preterm
infants. Despite several guidelines and reviews recommending that extremely preterm infants be given the
RDI of protein and calling for more aggressive nutritional management, especially in the early postnatal
period, we have shown that this does not have an impact on LBM at term age equivalent. Similarly,
the use of 20% SMOFlipid as the primary lipid composition resulted in similar IHCL levels as those found in
infants who received 20% Intralipid.
This is the first RCT of SMOFlipid compared with Intralipid in preterm infants to study the impact of lipid
composition on IHCL content. Preterm infants are known to have elevated IHCL content compared with
term-born infants, and this is correlated with early lipid intake.58 IHCL content measured with MRS in
adults has been shown to have good diagnostic accuracy and compares favourably with the gold standard
of liver biopsy for the quantitative measurement of hepatic steatosis.59 SMOFlipid has been shown to be
liver protective in the context of intestinal failure and PN in children and adults. Increasingly, the use of
SMOFlipid has been adopted for use in neonates with liver impairment. However, to date, there have been
no studies showing benefit for its use to prevent hepatic impairment. Previous studies of SMOFlipid in
preterm infants published to date have focused mainly on lipid profiles,60–63 including one small study on
the incidence of retinopathy of prematurity62 and a further one on the impact on growth outcomes.60
In contrast to Vlaardingerbroek et al.,60 we did not find any difference in growth outcomes in either weight
or head circumference in this study between the group receiving SMOFlipid and the one receiving Intralipid.
Although this study was not powered to detect a significant difference in rates of sepsis, there was a
higher rate of sepsis associated with SAE reports in the SMOFlipid group (15.5% vs. 3.6%), although this
finding could be due to chance. A systematic review, comparing soya bean with non-soya bean lipid
preparations, found a trend towards a lower incidence of sepsis, which did not reach statistical
significance, in the group receiving the non-soya bean lipid preparation. In the current study, 20% SMOFlipid
did not result in a reduction in IHCL content in preterm infants at term when used as a primary lipid
composition. Although there were no differences between the groups, the quantity of IHCLs in the cohort of
babies was similar to that in our previously published work comparing preterm infants with term-born
infants.18,64 We utilised IHCL content as a mechanistic marker of lipid tolerance, as levels at term correlate with
early lipid intake58 and are higher in very preterm than in full-term infants18 and young adults.65 We also
monitored liver function using conventional biochemical markers and identified no between-group
differences. Overall, our data support the conclusion of a systematic review and meta-analysis that fish
oil-based lipid emulsions do not prevent PN-associated cholestasis31 although the possibility that other
formulations, including those with higher fish oil content, may be beneficial is not precluded.
59
DISCUSSION
Other strengths of this study are that there was a prospective collection of detailed data of nutritional
intake from birth to discharge and the proportion of missing nutritional data was < 0.1% for data on trial
PN and < 5% for non-trial PN. The cohort, for this reason, lends itself to the long-term study of early
nutritional intervention on outcomes such as neurodevelopment and later metabolic health. As evident
from Tables 9 and 10, the trial interventions were delivered according to protocol both in the
commencement of PN within 24 hours (barring a few protocol violations) and in the subsequent immediate
postnatal period. By reducing the concentration of glucose, it was possible to commence PN without the need
for central access, which can be a rate-limiting factor in the early commencement of PN. Additionally, we
were able to demonstrate adherence to the use of a standardised regimen with a standardised prespecified
approach to the management of electrolyte disturbances. Previous non-randomised studies comparing
standardised with individualised PN have been inconclusive on the effect of these regimens on delivery of the
required amounts of nutrition. We have demonstrated that using a standardised PN regimen in the context of
a RCT is feasible. This is an important outcome, as increasingly there is recognition that current practices in
relation to the prescription, preparation and use of PN pose a potential clinical risk to patients. There are
several standard bags and regimens commercially available on the market. However, none of these regimens
has been subjected to the rigour of a large RCT with clinically meaningful outcomes and the concurrent
collection of a host of safety data. Data collection for this study included daily electrolyte and biochemical data
while infants were on trial PN and weekly thereafter.
The study was carried out in four neonatal units in London and the south-east of England. Two of the
units were designated NICUs in nature, whereas the remaining two were designated local neonatal units
or level 2 units that cared for infants of > 27 weeks of completed gestation. All units serve a varied
population in terms of both ethnic and socioeconomic backgrounds. Owing to the lack of MRI facilities
for research use on site, the original trial protocol dictated that infants had to be discharged from hospital
before the measurement of the primary outcome, as the MRI facility was located at a site separate from
the location of the four hospitals and it would have been unethical to transfer a baby for MRI purely for
the purpose of research if the baby was not fit to be discharged. This potentially could have resulted in a
bias, with the sickest babies being excluded from the measurement of primary outcome. However, early on
in the trial the hospital where the majority of infants were recruited was able to scan infants who were still
inpatients, resulting in nine infants who were otherwise not fit for discharge to be safely scanned on site.
This allowed these infants who would have otherwise been excluded from the primary outcome measure
analysis to be included. Four babies were excluded from the analysis of primary outcome measure because
they were still hospital inpatients during the window of measurement. There was also concern during the
trial design stage that the most immature infants at the highest risk of death may be under-represented in
the final results. Therefore, the TSC considered that it might be necessary to stop recruiting to the stratum
of infants in the higher gestational age category if this was found to be an issue in the interim analysis
(i.e. if more mature babies were being recruited and completing the scans there might have been over
representation of the more mature babies). However, this was not found to be the case and, hence, the
results are generalisable to not just the sicker infants but the most immature as well. In the group of
infants born between 31 and 33 weeks of gestation, or the very growth-restricted, but more mature,
infant in whom the use and justification of PN remains uncertain, definite recommendations cannot be
made from the results of this study.
There has been previous concern about the use of aggressive nutrition in a study comparing standard
intake with ‘aggressive’ nutrition when the intervention included higher protein and energy intake.66 The
authors terminated the study early as there was an increased rate of sepsis in the intervention group and
an association between low serum phosphate levels in the intervention group (despite increased phosphate
delivery) and sepsis. In our study both groups received similar intake of electrolytes and there was no
significant difference in the incidence of abnormalities in electrolytes.
Previous studies have suggested that the provision of increased early amino acids in PN is safe and not
associated with an increased incidence of metabolic acidosis or elevation in blood urea nitrogen.40,41 We
found a significantly higher incidence of elevated blood urea concentrations in the groups receiving the
60
RDI of amino acids. This is also in keeping with the studies of Vlaardingerbroek et al.,35,67 although there
was no associated increased incidence of metabolic acidosis. The significance of elevated blood urea in the
early postnatal period in preterm infants is unclear. It may reflect increased amino acid oxidation, but is
also dependent on renal function and hydration status. The long-term outcomes of providing increased
intake of amino acids from birth require evaluation before this approach can be recommended in practice.
We noted a smaller head circumference at term in babies receiving the higher amino acid intake. The
observation is at odds with the Standardised, Concentrated Additional Macronutrients, Parenteral nutrition
in very preterm infants (SCAMP) study, in which very preterm neonates randomised to receive higher PN
from birth had a larger head circumference at 28 days.68 Of note is that, although the SCAMP study aimed
to deliver large amounts of PN, randomisation occurred up to 120 hours of age (compared with 24 hours
in the NEON trial) and hence infants received a lower average energy and amino acid intake over the first
3 postnatal days than the NEON trial infants. The NEON trial was not powered to detect a difference in
head circumference, but our observation is a concern as the possibility of adverse effects from higher PN
has been raised previously. Choudri et al.69 found smaller brain growth and compromised neurodevelopment
despite equivalent weight gain in preterm piglets receiving total parenteral in comparison with total enteral
nutrition. Blanco et al.70 found that infants receiving an immediate parenteral amino acid intake of 2 g/kg/day
increasing to 4 g/kg/day, compared with a group randomised to receive a lower intake, had a lower mean
Mental Development Index at 18 months and lower mean z-scores for weight, length and head circumference.
Reassuringly, in the NEON trial we identified no between-group differences in brain volume.
Several studies have shown growth failure in preterm infants in the postnatal period and continuing to
adulthood.2,71 There are calls for a more aggressive approach to early postnatal nutrition to prevent this
growth failure.42,72 Various published guidelines recommend the early introduction of amino acids, with
recommended intakes of up to 4 g/kg/day.23 However, these recommendations are based on limited
evidence, and there are no long-term data to support the safety of such an approach. A recent paper with
a similar intervention of amino acids demonstrated that the early introduction of parenteral amino acids
given in conjunction with lipids improved nitrogen balance. However, higher intake of amino acids from
day 1 did not further improve the nitrogen balance, but led to increased amino acid oxidation.35 Although
there is concern that undernutrition is associated with adverse neurodevelopmental outcomes, there is also
some suggestion that ‘overnutrition’ may also be detrimental to neurodevelopment.73 As noted above,
there are animal data reporting an association between PN and adverse neurodevelopment when compared
with enteral nutrition.69 Interestingly, in that study, the pigs fed on the enteral diet showed a slowing of
growth before recovery of growth rate to match the PN-fed pigs. The PN-fed pigs showed a positive growth
trajectory in the immediate postnatal period, which excludes poor postnatal growth as being the cause of
adverse neurodevelopment. This study shows that commencing amino acids within the first 24 and increasing
the quantity to a maximum of 2.7 g/kg/day when accompanied by the early introduction of enteral feeds
results in an increase in LBM compared with historical controls and no significant difference between the
intervention and controls. Our sample size calculation was based on our previous work.19 Practice in neonatal
PN has changed with the emphasis on commencing PN earlier. Data from the UK National Data Analysis Unit
show that year on year more infants born before 30 weeks of gestation are started on PN within the first
48 hours after birth, but currently up to one-fifth to one-quarter of babies do not receive PN until day 3.37
In the incremental group the mean protein intake from parenteral and enteral intake by day 3 was 3.4 g/kg/day.
The difference in mean protein intake between the incremental and RDI groups was significant only in the first
2 weeks, when the infants were becoming established on enteral nutrition. It is of note that the CI for the mean
difference in lean mass between the groups excludes the deviation of prespecified difference in lean mass on
which the sample size calculation was based. The trial was, therefore, adequately powered to detect any
clinically important differences between the groups.
Our own work and that of others has shown that LBM in preterm infants at term age equivalent is
significantly lower than in healthy term-born infants. A systematic review, including our work, has shown
that the magnitude of mean difference between preterm and term-born infants is about 460 g.74 The
mean LBM in the NEON trial cohort was 2.41 kg (SD 0.46 kg). These values of lean mass are higher than
61
DISCUSSION
the mean of our previous cohort of preterm infants of 2.1 kg (SD 0.4 kg), on which the sample size
calculation was based, and closer to the mean of our term-born cohort of 2.6 kg (SD 0.21 kg).19
Furthermore, the values are comparable to the mean lean mass seen in a more recent cohort of preterm
infants of 2.49 kg (95% CI 2.45 to 2.54 kg), published after the commencement of this study.11 As is
evident from the National Data Analysis Unit data, babies are increasingly receiving PN earlier. Babies
randomised to the standard arm in this trial commenced PN earlier than is routine practice and, hence,
were not exposed to deficits that may arise from delaying PN. Despite the difference in early protein intake
in the first week, this did not result in differences in LBM at term age equivalent. This suggests that,
provided PN is initiated early and established according to a standardised regimen, accompanied by early
introduction and advancement of milk feeds, it is possible to achieve LBM in preterm infants at term age
equivalent that are closer to that seen in healthy term-born infants.
Parenteral nutrition is a high-cost and widely used neonatal intensive care intervention, yet there have
been few previous RCTs and none that has evaluated effects on body composition.37 We achieved a clear
difference in amino acid intake between the Imm-RDI and Inc-AA groups. The possible reasons why this
did not translate into a difference in body composition or weight at term merit consideration. First, the
incidence of elevated blood urea concentrations was significantly higher the Imm-RDI groups. This suggests
that increased delivery above a threshold results in increased amino acid oxidation with no improvement in
nitrogen retention or growth, as suggested previously.35,75,76 We consider it unlikely that impaired utilisation
of amino acids was attributable to inadequate non-protein energy delivery, as there were no significant
differences between the groups in non-protein energy intakes. Second, trial interventions may have
resulted in a short-term difference in body composition that was attenuated when babies transitioned to
self-regulated suck feeds. Embleton and Cooke77 found that babies fed a higher-protein formula by
nasogastric tube achieved an increase in LBM that did not persist after a period of self-regulated feeding
by bottle.
One of the weaknesses of this study was that the appropriate MR images of the brain required to measure
volumes were not availabe for a significant proportion of babies. The brain volumes were derived from
T2-weighted images. This sequence followed the longer T1-weighted sequence in the scanning protocol
and babies often woke up at the end of the T1-weighted scan. This was a missed opportunity to test the
hypothesis that amino acid intake and SMOFlipid influences brain growth and volume by using a direct
measure of brain growth instead of the previously used surrogate measure of head circumference.68
However, in one-third of infants, for whom there were images of sufficient quality to analyse, there were
no differences seen between groups in relation to either intervention. If early nutritional intervention at a
period of rapid brain development has a long-term impact on neurodevelopment, then it is plausible that a
difference in total and regional brain volumes persists beyond the neonatal period. Follow-up of this cohort
in which detailed nutritional intake has been captured offers the unique opportunity of studying the
long-term impact of early nutrition on brain development as well as neurodevelopment. Establishing the
long-term safety of the introduction of higher amino acid intake is of particular importance given the calls
for early aggressive nutrition without the accompanying evidence of the lack of harm, both in the short
and long term. The QUICKI data were not able to be captured on a number of infants because of the
inability of non-lead sites to carry out this assessment, as well as babies being transferred to non-trial sites
before reaching 37 weeks postmenstrual age.
62
Chapter 6 Conclusions and recommendations

I n conclusion, commencement within 24 hours of birth of an incremental amino acid regimen providing
a maximum of 2.7 g/kg/day together with the early introduction of milk feeds, compared with the
immediate provision of an amino acid intake of 3.6 g/kg/day, does not appear to be detrimental to body
composition and may be safer, and SMOFlipid does not reduce intrahepatic lipid accumulation.
Extremely preterm infants at term age equivalent can achieve a body composition close to that of healthy
term-born infants if provided at an early stage with PN in accordance with a standardised regimen.
The results do not support the calls for more aggressive nutrition in the extremely preterm infant or the
routine use of SMOFlipid as reflected in international consensus statements (higher amounts of amino
acids) or as is increasingly seen in current practice. In the light of the unexpected finding of a smaller head
circumference in those randomised to receive immediate RDI of amino acids, we recommend that large
amounts of amino acids be used only in the context of randomised clinical trials. Optimal amino acid
intakes and intravenous lipid formulations for extremely preterm infants remains to be established.
Health-care recommendations
A key ancillary observation of this trial was that the use of standard PN regimens is feasible, acceptable to
clinicians even when blinded, can deliver desired nutritional intake without manipulation and is safe.
In our opinion, standardised regimens that have been tested in the context of a RCT should be adopted in
routine clinical practice to reduce the clinical risk to infants from variation in practice.
Research recommendations
1. We recommend that long-term follow-up of functional outcomes of neurodevelopment as well as

long-term body composition and metabolic health of both the trial interventions is essential before
either of the interventions studied in this trial can be recommended as routine practice.
2. It is essential to develop early biomarkers of nutritional intervention in neonates that are predictive of
long-term body composition and health. Demonstration of stability in body composition and IHCL
content between term age equivalent and childhood would provide justification for the use of body
composition and IHCL content as short-term surrogate outcomes in neonatal nutrition research.
3. This study did not address the role of carbohydrate in PN. The optimal intake of carbohydrate and the
management of hyperglycaemia whether by reducing intake of glucose or the use of insulin during PN
should be established.
63
Acknowledgements
T he following people are thanked for participating in the trial:
l parents and babies

l Professor Mary Rutherford, Professor of Perinatal Imaging, Centre for the Developing Brain,
King’s College London, London, UK; and Chris Gale for medical supervision of scans
l Imperial Clinical Trials Unit
l Juan Gonzalez-Maffè and Alexina Mason
l Clinical Trials and Evaluation Unit at Royal Brompton and Harefield NHS Foundation Trust
l Sealed Envelope Ltd for providing the randomisation list and IVRS.
Trial Steering Committee members
Richard Cooke, Lorraine Dob, Paul Clarke and Robert Hume.
Data Monitoring and Ethics Committee members
Peter Brocklehurst, Tim Cole, Tony Nunn and Helen Mactier.
Participating sites
Chelsea and Westminster Hospital
l Sabita Uthaya, Izabela Andrzejewska, Sara Abdula, Laura Read, Elaine Smith and Kathryn McCormick.
Medway Maritime Hospital
l Aung Soe, Helen McElroy, Abimbola Ojo, Helen Harizaj and Parool Darbar.
Northwick Park Hospital
l Richard Nicholl, Rosemond Owoo, Cecilia Lam, Matilda Lang and Theo Emmanuel.
West Middlesex Hospital
l Nour Elhadi, Dr Hashir Ariff, Siew Koay, Mylene Erese and Isabel Munoz.
65
ACKNOWLEDGEMENTS
Contributions of authors
Sabita Uthaya conceived and designed the study; analysed and interpreted the data; and drafted, revised
and approved the report.
Xinxue Liu analysed the data, and drafted, revised and approved the report.
Daphne Babalis designed the study, analysed and interpreted the data; and drafted, revised and
approved the report.
Caroline Dore designed the study, analysed and interpreted the data; and drafted, revised and approved
the report.
Jane Warwick analysed the data and approved the report.
Jimmy Bell acquired and analysed the data, and drafted, revised and approved the report.
Louise Thomas analysed the data, and drafted, revised and approved the report.
Deborah Ashby analysed the data, and drafted, revised and approved the report.
Giuliana Durighel acquired the data, and drafted, revised and approved the report.
Ash Ederies acquired and analysed the data, and drafted, revised and approved the report.
Monica Yanez-Lopez analysed the data, and revised and approved the report.
Neena Modi conceived and designed the study, analysed and interpreted the data; and drafted, revised
and approved the report.
Data sharing statement
All available data can be obtained from the corresponding author.
66
References
1. Dusick AM, Poindexter BB, Ehrenkranz RA, Lemons JA. Growth failure in the preterm infant: can
we catch up? Semin Perinatol 2003;27:302–10. http://dx.doi.org/10.1016/S0146-0005(03)00044-2
2. Hack M, Schluchter M, Cartar L, Rahman M, Cuttler L, Borawski E. Growth of very low birth weight
infants to age 20 years. Pediatrics 2003;112:e30–8. http://dx.doi.org/10.1542/peds.112.1.e30
3. Eriksson JG, Forsen T, Tuomilehto J, Winter PD, Osmond C, Barker DJ. Catch-up growth in
childhood and death from coronary heart disease: longitudinal study. BMJ 1999;318:427–31.
http://dx.doi.org/10.1136/bmj.318.7181.427
4. Ong KK, Ahmed ML, Emmett PM, Preece MA, Dunger DB. Association between postnatal catch-up
growth and obesity in childhood: prospective cohort study. BMJ 2000;320:967–71.
http://dx.doi.org/10.1136/bmj.320.7240.967
5. Astbury J, Orgill AA, Bajuk B, Yu VY. Sequelae of growth failure in appropriate for gestational age,
very low-birthweight infants. Dev Med Child Neurol 1986;28:472–9. http://dx.doi.org/10.1111/
j.1469-8749.1986.tb14285.x
6. Latal-Hajnal B, von Siebenthal K, Kovari H, Bucher HU, Largo RH. Postnatal growth in VLBW
infants: significant association with neurodevelopmental outcome. J Pediatr 2003;143:163–70.
http://dx.doi.org/10.1067/S0022-3476(03)00243-9
7. Putet G, Senterre J, Rigo J, Salle B. Energy balance and composition of body weight. Biol Neonate
1987;52(Suppl. 1):17–24. http://dx.doi.org/10.1159/000242736
8. Regan FM, Cutfield WS, Jefferies C, Robinson E, Hofman PL. The impact of early nutrition in
premature infants on later childhood insulin sensitivity and growth. Pediatrics 2006;118:1943–9.
http://dx.doi.org/10.1542/peds.2006-0733
9. Crump C. Medical history taking in adults should include questions about preterm birth.
BMJ 2014;349:g4860. http://dx.doi.org/10.1136/bmj.g4860
10. Embleton NE, Pang N, Cooke RJ. Postnatal malnutrition and growth retardation: an inevitable
consequence of current recommendations in preterm infants? Pediatrics 2001;107:270–3.
http://dx.doi.org/10.1542/peds.107.2.270
11. Vasu V, Durighel G, Thomas L, Malamateniou C, Bell JD, Rutherford MA, et al. Preterm nutritional
intake and MRI phenotype at term age: a prospective observational study. BMJ Open
2014;4:e005390. http://dx.doi.org/10.1136/bmjopen-2014-005390
12. Hofman PL, Regan F, Jackson WE, Jefferies C, Knight DB, Robinson EM, et al. Premature birth and
later insulin resistance. N Engl J Med 2004;351:2179–86. http://dx.doi.org/10.1056/NEJMoa042275
13. Irving RJ, Belton NR, Elton RA, Walker BR. Adult cardiovascular risk factors in premature babies.
Lancet 2000;355:2135–6. http://dx.doi.org/10.1016/S0140-6736(00)02384-9
14. Rotteveel J, van Weissenbruch MM, Twisk JW, Delemarre-Van de Waal HA. Infant and childhood
growth patterns, insulin sensitivity, and blood pressure in prematurely born young adults. Pediatrics
2008;122:313–21. http://dx.doi.org/10.1542/peds.2007-2012
15. Hovi P, Andersson S, Eriksson JG, Järvenpää AL, Strang-Karlsson S, Mäkitie O, et al. Glucose
regulation in young adults with very low birth weight. N Engl J Med 2007;356:2053–63.
http://dx.doi.org/10.1056/NEJMoa067187
67
REFERENCES
16. Rotteveel J, van Weissenbruch MM, Twisk JW, Delemarre-Van de Waal HA. Abnormal lipid profile
and hyperinsulinaemia after a mixed meal: additional cardiovascular risk factors in young adults
born preterm. Diabetologia 2008;51:1269–75. http://dx.doi.org/10.1007/s00125-008-1029-5
17. Singhal A, Fewtrell M, Cole TJ, Lucas A. Low nutrient intake and early growth for later insulin
resistance in adolescents born preterm. Lancet 2003;361:1089–97. http://dx.doi.org/10.1016/
S0140-6736(03)12895-4
18. Thomas EL, Uthaya S, Vasu V, McCarthy JP, McEwan P, Hamilton G, et al. Neonatal
intrahepatocellular lipid. Arch Dis Child Fetal Neonatal Ed 2008;93:F382–3. http://dx.doi.org/
10.1136/adc.2007.127431
19. Uthaya S, Thomas EL, Hamilton G, Dore CJ, Bell J, Modi N. Altered adiposity after extremely
preterm birth. Pediatr Res 2005;57:211–15. http://dx.doi.org/10.1203/01.PDR.0000148284.
58934.1C
20. Petrou S, Henderson J, Bracewell M, Hockley C, Wolke D, Marlow N. Pushing the boundaries of
viability: the economic impact of extreme preterm birth. Early Hum Dev 2006;82:77–84.
http://dx.doi.org/10.1016/j.earlhumdev.2006.01.002
21. Fomon SJ, Nelson SE. Body composition of the male and female reference infants. Annu Rev Nutr
2002;22:1–17. http://dx.doi.org/10.1146/annurev.nutr.22.111401.145049
22. Ziegler EE, O’Donnell AM, Nelson SE, Fomon SJ. Body composition of the reference fetus.
Growth 1976;40:329–41.
23. Koletzko B, Goulet O, Hunt J, Krohn K, Shamir R, Parenteral Nutrition Guidelines Working Group,
et al. 1. Guidelines on paediatric parenteral nutrition of the European Society of Paediatric
Gastroenterology, Hepatology and Nutrition (ESPGHAN) and the European Society for Clinical
Nutrition and Metabolism (ESPEN), supported by the European Society of Paediatric Research
(ESPR). J Pediatr Gastroenterol Nutr 2005;41(Suppl. 2):1–87. http://dx.doi.org/10.1097/
01.mpg.0000181841.07090.f4
24. Tsang RC. Nutrition of the Preterm Infant: Scientific Basis and Practical Guidelines. Cincinnati, OH:
Digital Educational Publishing; 2005.
25. Kleinman RE. Pediatric Nutrition Handbook. 5th edn. Elk Grove Village, IL: American Academy of
Pediatrics; 2004.
26. Roggero P, Giannì ML, Amato O, Orsi A, Piemontese P, Puricelli V, et al. Influence of protein and
energy intakes on body composition of formula-fed preterm infants after term. J Pediatr
Gastroenterol Nutr 2008;47:375–8. http://dx.doi.org/10.1097/MPG.0b013e3181615cba
27. Simmer K. Aggressive nutrition for preterm infants – benefits and risks. Early Hum Dev
2007;83:631–4. http://dx.doi.org/10.1016/j.earlhumdev.2007.07.013
28. Ehrenkranz RA. Early, aggressive nutritional management for very low birth weight infants: what is
the evidence? Semin Perinatol 2007;31:48–55. http://dx.doi.org/10.1053/j.semperi.2007.02.001
29. Paedeatric Chief Pharmacits Group. Improving Practice and Reducing Risk in the Provision
of Parenteral Nutrition for Neonates & Children. URL: www.rpharms.com/support-pdfs/
minimising-risk-pn-children-%286%29.pdf? (accessed 24 November 2015).
30. National Confidential Enquiry into Patient Outcome and Death. A Mixed Bag: An Enquiry into the
Care of Hospital Patients Receiving Parenteral Nutrition. URL: www.ncepod.org.uk/2010pn.html
(accessed 24 November 2015).
31. Park HW, Lee NM, Kim JH, Kim KS, Kim SN. Parenteral fish oil-containing lipid emulsions may
reverse parenteral nutrition-associated cholestasis in neonates: a systematic review and
meta-analysis. J Nutr 2015;145:277–83. http://dx.doi.org/10.3945/jn.114.204974
68
32. Tan M, Abernethy L, Cooke R. Improving head growth in preterm infants – a randomised
controlled trial II: MRI and developmental outcomes in the first year. Arch Dis Child Fetal Neonatal
Ed 2008;93:F342–6. http://dx.doi.org/10.1136/adc.2007.124255
33. Tan MJ, Cooke RW. Improving head growth in very preterm infants – a randomised controlled trial I:
neonatal outcomes. Arch Dis Child Fetal Neonatal Ed 2008;93:F337–41. http://dx.doi.org/
10.1136/adc.2007.124230
34. Moyses HE, Johnson MJ, Leaf AA, Cornelius VR. Early parenteral nutrition and growth outcomes in
preterm infants: a systematic review and meta-analysis. Am J Clin Nutr 2013;97:816–26.
http://dx.doi.org/10.3945/ajcn.112.042028
35. Vlaardingerbroek H, Vermeulen MJ, Rook D, van den Akker CH, Dorst K, Wattimena JL. Safety and
efficacy of early parenteral lipid and high-dose amino acid administration to very low birth weight
infants. J Pediatr 2013;163:638–44.e5. http://dx.doi.org/10.1016/j.jpeds.2013.03.059
36. Vlaardingerbroek H, Veldhorst MA, Spronk S, van den Akker CH, van Goudoever JB. Parenteral
lipid administration to very-low-birth-weight infants – early introduction of lipids and use of new
lipid emulsions: a systematic review and meta-analysis. Am J Clin Nutr 2012;96:255–68.
http://dx.doi.org/10.3945/ajcn.112.040717
37. Uthaya S, Modi N. Practical preterm parenteral nutrition: systematic literature review and
recommendations for practice. Early Hum Dev 2014;90:747–53. http://dx.doi.org/10.1016/
j.earlhumdev.2014.09.002
38. Holmes A, Doré CJ, Saraswatula A, Bamford KB, Richards MS, Coello R, et al. Risk factors and
recommendations for rate stratification for surveillance of neonatal healthcare-associated
bloodstream infection. J Hosp Infect 2008;68:66–72. http://dx.doi.org/10.1016/j.jhin.2007.08.019
39. Robinson DT, Ehrenkranz RA. Parenteral nutrition-associated cholestasis in small for gestational age
infants. J Pediatr 2008;152:59–62. http://dx.doi.org/10.1016/j.jpeds.2007.06.002
40. Ibrahim HM, Jeroudi MA, Baier RJ, Dhanireddy R, Krouskop RW. Aggressive early total parental
nutrition in low-birth-weight infants. J Perinatol 2004;24:482–6. http://dx.doi.org/10.1038/
sj.jp.7211114
41. Thureen PJ, Melara D, Fennessey PV, Hay WW Jr. Effect of low versus high intravenous amino acid
intake on very low birth weight infants in the early neonatal period. Pediatr Res 2003;53:24–32.
http://dx.doi.org/10.1203/00006450-200301000-00008
42. Denne SC, Poindexter BB. Evidence supporting early nutritional support with parenteral amino acid
infusion. Semin Perinatol 2007;31:56–60. http://dx.doi.org/10.1053/j.semperi.2007.02.005
43. Waitzberg DL, Torrinhas RS, Jacintho TM. New parenteral lipid emulsions for clinical use. J Parenter
Enteral Nutr 2006;30:351–67. http://dx.doi.org/10.1177/0148607106030004351
44. Antébi H1, Mansoor O, Ferrier C, Tétégan M, Morvan C, Rangaraj J, et al. Liver function and
plasma antioxidant status in intensive care unit patients requiring total parenteral nutrition:
comparison of 2 fat emulsions. J Parenter Enteral Nutr 2004;28:142–8. http://dx.doi.org/10.1177/
0148607104028003142
45. Modi N, Thomas EL, Uthaya SN, Umranikar S, Bell JD, Yajnik C. Whole body magnetic resonance
imaging of healthy newborn infants demonstrates increased central adiposity in Asian Indians.
Pediatr Res 2009;65:584–7. http://dx.doi.org/10.1203/PDR.0b013e31819d98be
46. Ross R, Leger L, Guardo R, De Guise J, Pike BG. Adipose tissue volume measured by magnetic
resonance imaging and computerized tomography in rats. J Appl Physiol 1991;70:2164–72.
47. Martin AD, Daniel MZ, Drinkwater DT, Clarys JP. Adipose tissue density, estimated adipose lipid
fraction and whole body adiposity in male cadavers. Int J Obes Relat Metab Disord 1994;18:79–83.
69
REFERENCES
48. Naressi A, Couturier C, Castang I, de Beer R, Graveron-Demilly D. Java-based graphical user

interface for MRUI, a software package for quantitation of in vivo/medical magnetic resonance
spectroscopy signals. Comput Biol Med 2001;31:269–86. http://dx.doi.org/10.1016/S0010-4825
(01)00006-3
49. Stefan D, Cesare FD, Andrasescu A, Popa E, Lazariev A, Vescovo, et al. Quantitation of magnetic
resonance spectroscopy signals: the jMRUI software package. Meas Sci Technol 2009;20:104035.
http://dx.doi.org/10.1088/0957-0233/20/10/104035
50. Vanhamme L, van den Boogaart A, Van Huffel S. Improved method for accurate and efficient
quantification of MRS data with use of prior knowledge. J Magn Reson 1997;129:35–43.
http://dx.doi.org/10.1006/jmre.1997.1244
51. Makropoulos A, Gousias IS, Ledig C, Aljabar P, Serag A, Hajnal JV, et al. Automatic whole brain
MRI segmentation of the developing neonatal brain. IEEE Trans Med Imaging 2014;33:1818–31.
http://dx.doi.org/10.1109/TMI.2014.2322280
52. Katz A, Nambi SS, Mather K, Baron AD, Follmann DA, Sullivan G, et al. Quantitative insulin
sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans.
J Clin Endocrinol Metab 2000;85:2402–10. http://dx.doi.org/10.1210/jcem.85.7.6661
53. Jackson L, Burchell A, McGeechan A, Hume R. An inadequate glycaemic response to glucagon is
linked to insulin resistance in preterm infants? Arch Dis Child Fetal Neonatal Ed 2003;88:F62–6.
http://dx.doi.org/10.1136/fn.88.1.F62
54. Pocock SJ. Clinical Trials: A Practical Approach. Chichester: Wiley; 1983.
55. McAlister FA, Straus SE, Sackett DL, Altman DG. Analysis and reporting of factorial trials: a
systematic review. JAMA 2003;289:2545–53. http://dx.doi.org/10.1001/jama.289.19.2545
56. UK Government. The Medicines for Human Use (Clinical Trials) Regulations 2004.
URL: www.legislation.gov.uk/uksi/2004/1031/contents/made (accessed 24 November 2015).
57. Cole TJ, Freeman JV, Preece MA. British 1990 growth reference centiles for weight, height, body
mass index and head circumference fitted by maximum penalized likelihood. Stat Med 1998;17:407–29.
http://dx.doi.org/10.1002/(SICI)1097-0258(19980228)17:4<407::AID-SIM742>3.0.CO;2-L
58. Vasu V, Thomas EL, Durighel G, Hyde MJ, Bell JD, Modi N. Early nutritional determinants of
intrahepatocellular lipid deposition in preterm infants at term age. Int J Obes 2013;37:500–4.
http://dx.doi.org/10.1038/ijo.2012.213
59. Georgoff P, Thomasson D, Louie A, Fleischman E, Dutcher L, Mani H, et al. Hydrogen-1 MR
spectroscopy for measurement and diagnosis of hepatic steatosis. Am J Roentgenol 2012;199:2–7.
http://dx.doi.org/10.2214/AJR.11.7384
60. Vlaardingerbroek H, Vermeulen MJ, Carnielli VP, Vaz FM, van den Akker CH, van Goudoever JB.
Growth and fatty acid profiles of VLBW infants receiving a multicomponent lipid emulsion
from birth. J Pediatr Gastroenterol Nutr 2014;58:417–27. http://dx.doi.org/10.1097/
MPG.0000000000000280
61. Rayyan M, Devlieger H, Jochum F, Allegaert K. Short-term use of parenteral nutrition with a
lipid emulsion containing a mixture of soybean oil, olive oil, medium-chain triglycerides,
and fish oil: a randomized double-blind study in preterm infants. J Parenter Enteral Nutr
2012;36(Suppl. 1):81S–94S. http://dx.doi.org/10.1177/0148607111424411
62. Beken S, Dilli D, Fettah ND, Kabatas EU, Zenciroglu A, Okumus N. The influence of fish-oil lipid
emulsions on retinopathy of prematurity in very low birth weight infants: a randomized controlled
trial. Early Hum Dev 2014;90:27–31. http://dx.doi.org/10.1016/j.earlhumdev.2013.11.002
70
63. Tomsits E, Pataki M, Tolgyesi A, Fekete G, Rischak K, Szollar L. Safety and efficacy of a lipid
emulsion containing a mixture of soybean oil, medium-chain triglycerides, olive oil, and fish oil:
a randomised, double-blind clinical trial in premature infants requiring parenteral nutrition.
J Pediatr Gastroenterol Nutr 2010;51:514–21. http://dx.doi.org/10.1097/MPG.0b013e3181de210c
64. Vasu V, Modi N. Assessing the impact of preterm nutrition. Early Hum Dev 2007;83:813–8.
http://dx.doi.org/10.1016/j.earlhumdev.2007.09.008
65. Thomas EL, Parkinson JR, Hyde MJ, Yap IK, Holmes E, Doré CJ, et al. Aberrant adiposity and
ectopic lipid deposition characterize the adult phenotype of the preterm infant. Pediatr Res
2011;70:507–12. http://dx.doi.org/10.1203/PDR.0b013e31822d7860
66. Moltu SJ, Strømmen K, Blakstad EW, Almaas AN, Westerberg AC, Brække K, et al. Enhanced
feeding in very-low-birth-weight infants may cause electrolyte disturbances and septicemia –
a randomized, controlled trial. Clin Nutr 2013;32:207–12. http://dx.doi.org/10.1016/j.clnu.2012.09.004
67. Vlaardingerbroek H, Roelants JA, Rook D, Dorst K, Schierbeek H, Vermes A, et al. Adaptive
regulation of amino acid metabolism on early parenteral lipid and high-dose amino acid
administration in VLBW infants – a randomized, controlled trial. Clin Nutr 2014;33:982–90.
http://dx.doi.org/10.1016/j.clnu.2014.01.002
68. Morgan C, McGowan P, Herwitker S, Hart AE, Turner MA. Postnatal head growth in preterm
infants: a randomized controlled parenteral nutrition study. Pediatrics 2014;133:e120–8.
69. Choudhri AF, Sable HJ, Chizhikov VV, Buddington KK, Buddington RK. Parenteral nutrition
compromises neurodevelopment of preterm pigs. J Nutr 2014;144:1920–7. http://dx.doi.org/
10.3945/jn.114.197145
70. Blanco CL, Gong AK, Schoolfield J, Green BK, Daniels W, Liechty EA, et al. Impact of early and
high amino acid supplementation on ELBW infants at 2 years. J Ped Gastroenterol Nutr
2012;54:601–7.
71. Ehrenkranz RA, Younes N, Lemons JA, Fanaroff AA, Donovan EF, Wright LL, et al. Longitudinal
growth of hospitalized very low birth weight infants. Pediatrics 1999;104:280–9. http://dx.doi.org/
10.1542/peds.104.2.280
72. Poindexter B. Approaches to growth faltering. World Rev Nutr Diet 2014;110:228–38.
http://dx.doi.org/10.1159/000358471
73. Fewtrell MS, Abbott RA, Kennedy K, Singhal A, Morley R, Caine E, et al. Randomized, double-blind
trial of long-chain polyunsaturated fatty acid supplementation with fish oil and borage oil in
preterm infants. J Pediatr 2004;144:471–9. http://dx.doi.org/10.1016/j.jpeds.2004.01.034
74. Johnson MJ, Wootton SA, Leaf AA, Jackson AA. Preterm birth and body composition at term
equivalent age: a systematic review and meta-analysis. Pediatrics 2012;130:e640–9.
75. Clark RH, Chace DH, Spitzer AR, Pediatrix Amino Acid Study Group. Effects of two different doses
of amino acid supplementation on growth and blood amino acid levels in premature neonates
admitted to the neonatal intensive care unit: a randomized, controlled trial. Pediatrics
2007;120:1286–96. http://dx.doi.org/10.1542/peds.2007-0545
76. Burattini I, Bellagamba MP, Spagnoli C, D'Ascenzo R, Mazzoni N, Peretti A, et al. Targeting 2.5
versus 4 g/kg/day of amino acids for extremely low birth weight infants: a randomized clinical trial.
J Pediatr 2013;163:1278–82.e1. http://dx.doi.org/10.1016/j.jpeds.2013.06.075
77. Embleton ND, Cooke RJ. Protein requirements in preterm infants: effect of different levels of
protein intake on growth and body composition. Pediatr Res 2005;58:855–60. http://dx.doi.org/
10.1203/01.PDR.0000182586.46532.7C
71
Appendix 1 Distribution of primary and

secondary outcomes after transformation
(a) (b) (c) (d)
6 4
8
Frequency
Frequency
Frequency
Frequency
6
4
4 3 2
2
0 0 0 0
1500 3000 –3 –1 1 400 600 250 350 450
Non-adipose IHCL on Total cerebral Whole-brain
body mass log-scale volume volume
(e) (f) (g) (h)

Frequency
Frequency
Frequency
Frequency
4 4
8 8
2 2 4 4
0 0 0 0
25 35 45 0.16 0.20 2000 3500 40 50 60
Posterior fossa QUICKI EOS weight EOS length
volume
(i) (j) (k) (l)
20
Frequency
Frequency
Frequency
Frequency
12 8
8
10 6 4
4
0 0 0 0
35 40 45 200 600 1200 50 150 0 10 30
EOS head SSCAT IAT DSCAAT
circumference
(m) (n) (o) (p)
15
8
Frequency
Frequency
Frequency
Frequency
12 15
6 4
5 5
0 0 0 0
10 30 50 150 250 0 400 1000 0.05 0.20 0.35
IAAT SSCAAT TAT % ATM
(q) (r)
15
Frequency
Frequency
5 4
0 0
0.02 0.05 0.08 0.10 0.20
Ratio of active Ratio of IAT
to inactive to SSCAT
FIGURE 17 Distribution of primary and secondary outcomes after transformation. IHCL values are log-transformed.
ATM, adipose tissue mass; DSCAAT, deep subcutaneous abdominal adipose tissue; EOS, end of study; IAAT, internal
non-abdominal adipose tissue; IAT, internal adipose tissue; SSCAAT, superficial subcutaneous abdominal adipose
tissue; SSCAT, superficial subcutaneous adipose tissue; TAT, total adipose tissue.
73
Appendix 2 Parent information sheet, magnetic

resonance information sheet and consent form
INFORMATION SHEET FOR PARENTS
Study title: Amino acid regimen and intravenous lipid composition in preterm
parenteral nutrition: a randomised controlled trial of Nutritional Evaluation and
Optimisation in Neonates (NEON)
Invitation to participate
We would like to invite you to consider giving your consent to include your baby in a
research study. Please take time to read this information carefully and discuss it with
others if you wish. A member of our team will go through the information sheet with
you. Please ask if there is anything that is not clear or if you would like more
information.
What is the purpose of the study?

Extremely preterm infants (born less than 31 weeks of gestation) spend several
weeks and months in hospital. Feeding babies born so early is difficult. By the time
they reach their due date their weight is typically about 1 kg (2 lbs) less than that of a
full-term healthy baby.
Food is initially provided as a fluid called parenteral nutrition (PN) that is given into a
vein. As extremely preterm babies may have other medical problems, traditionally,
the amount of nutrition provided in PN has been gradually increased in a cautious,
stepwise manner. This means that it can take several days to reach the full
recommended nutritional intake to enable them to grow.
Though necessary, PN has complications, especially if used for several weeks. One
complication is damage to the liver. The type of fat used in PN may affect this.
Recent studies have shown that giving preterm babies the recommended amount of
nutrition straight away without the stepwise approach, and using a new type of fat
(SMOF lipid) that contains soybean oil, olive and fish oil rather than the fat we
currently use (Intralipid) which has soybean oil alone is safe. Although these
approaches to feeding are used by doctors in day to day practice, we do not know for
sure if one has benefit over the other in preterm babies. Before this can be
introduced into everyday practice as recommendation we need to make sure this
approach is beneficial.
The purpose of this study is to improve the growth and health of preterm babies. We
will do this by:
1) comparing “immediate” introduction of Parenteral Nutrition with “stepwise”
introduction
2) comparing the currently used fat in PN, with a newer type of fat that we hope
is less harmful to the liver.
Why have I been invited?

You have been invited because your baby has been born prematurely (at less than
31 weeks of gestation) and needs Parenteral Nutrition.
Does my baby have to take part?

It is entirely up to you to decide whether or not you wish your baby to take part. If you
do decide to take part you will be given this information sheet to keep and be asked
to sign a consent form, a copy of which will be given to you. If you decide to take part
75
APPENDIX 2
you are still free to withdraw your baby from the study at any time and without giving
a reason. We would ask that you allow us to use any information collected up to that
point. A decision not to take part will not affect the standard of care that your baby
receives.
What will the study involve?

We will be enrolling 128 babies into this study. Babies will receive one of four
different combinations of parenteral nutrition treatment.
Group 1: Stepwise introduction of PN and currently used fat

Group 2: Stepwise introduction of PN and newer lipid
Group 3: Immediate introduction of PN and currently used fat
Group 4: Immediate introduction of PN and newer lipid
The process of allocating which treatment a baby receives is done by

‘randomisation’.
Randomisation means that a specially designed computer programme will determine

the choice. There is a 50% chance that your baby will receive either type of treatment
(like tossing a coin). Randomisation is done in order to ensure that every baby has
the same chance of receiving either one or the other treatment. We will not know
which treatment your baby receives until the end of the study. This is to prevent any
bias in the results of the study.
Your baby will start milk feeds and the study PN within 24 hours as is normal
practice. We recommend you provide your own expressed breast milk to your baby.
When your baby is tolerating milk feeds well and no longer requires PN this will be
stopped. We will collect the following information on your baby:
Routinely collected information

This is collected for any baby receiving care on a neonatal unit. This includes
measurement of growth, recording the amount of nutrition (milk or PN) a baby
receives and blood tests that show the effect of nutrition on the body.
Information collected for research

This is information that will be collected in addition to routine tests and information.
The blood tests will be done at the same time as other routine blood tests and after
the tests are done the samples will be destroyed. The samples will be labelled with a
unique trial identification number.
The following additional tests will be done:
1. We will take 3 drops of blood in the first week, and additionally, once a week
during your baby’s stay in hospital, we will collect a few drops of urine (10 drops)
and stool from the nappy to measure metabolite levels. The test uses a new
technique called magnetic resonance (a method that uses a magnetic field) which
allows a large number of metabolites (waste products of food) to be measured in
very small quantities of blood or urine.
2. If your baby is born at Chelsea and Westminster hospital, we will take a few
drops of blood (0.5 – 1 ml) to measure the type of fat present in the blood on the
first and fifth day after birth.
3. When your baby reaches his /her due date we will take a few drops (1ml) of blood
to measure sugar, insulin and metabolite levels.
76
4. In order to determine the results of the treatments on the development of the

body, brain and liver, we will arrange a magnetic resonance (MR) scan to obtain
pictures of your baby’s body and brain. This will be done when you baby has
reached his or her due date and has gone home. We will give you more
information about this nearer the time of this scan.
Other than the MR scans the study samples may not be taken if your baby is
transferred to another hospital. If this is the case we will take a sample of urine when
your baby has the MR scan.
After your baby has had his or her scan, involvement in this study will end. Your baby
will continue to receive routine care and follow up. If you agree, we may contact you
about future research studies looking at how nutrition affects babies in later life.
Expenses and payments

The MR scan is done at the Hammersmith Hospital where we will ask you to come
for a morning or an afternoon. We will arrange for a taxi or reimburse your travel and
parking costs.
What are the alternatives for diagnosis or treatment?

If you choose not to enter your baby in to the study then your baby will receive
standard care which will include PN. It is not routine practice to do MR scanning of
the body, liver or brain.
What are the possible disadvantages and risks of taking part?

Parenteral nutrition (PN) is usually unavoidable for extremely preterm infants. The
benefits of PN in neonatal intensive care are believed to outweigh the risks. The
additional risk from using “immediate” PN from day one is minimal. Previous studies
have not shown an increase in complications. SMOF lipid has been used in other
studies and is often used in preterm infants receiving prolonged PN.
You will need to travel to the Hammersmith Hospital after discharge for the MR
scans. They are not being carried out for clinical diagnosis but there is a possibility
that they might show something unexpected. If this occurs, a senior doctor will
explain this to you and notify your GP, and discuss whether any further action is
necessary.
What are the possible benefits of taking part?

As we do not know if one treatment has benefit over the other there is no direct
benefit to your baby. By following a standardised approach to milk feeding as in this
study, there may be benefits to your baby. However, the information we obtain from
this study may help us to improve nutrition of preterm babies in the future.
What if there is a problem?

All the treatments used in this study are currently used in day to day practice and it is
not anticipated that there will be problems related directly to the study. As with all
studies, Imperial College London holds insurance policies which apply to this study.
If your baby experiences harm or injury as a result of taking part in this study, you
may be eligible to claim compensation without having to prove that Imperial College
is at fault. This does not affect your legal rights to seek compensation.
If your baby is harmed due to someone’s negligence, then you may have grounds for
a legal action. Regardless of this, if you wish to complain, or have any concerns
about any aspect of the way you have been treated during the course of this study
then you should immediately inform the Investigator (Insert name and contact
77
APPENDIX 2
details). The normal National Health Service complaint complaints mechanisms are
also available to you. If you are still not satisfied with the response, you may contact
the Imperial AHSC Joint Research Office.
Will my baby’s taking part in the study be kept confidential?

All information which is collected about your baby during the course of the research
will be kept strictly confidential, and any information about your baby which leaves
the hospital will have your baby’s name and address removed so that he /she cannot
be recognised. They may be looked at by authorised people to check that the study
is being carried out correctly. All will have a duty of confidentiality to your baby as a
research participant and we will do our best to meet this duty.
What if relevant new information becomes available?

Sometimes we get new information about the treatment being studied. If this
happens, your research doctor will tell you and discuss your baby’s options.
What will happen to the results of the research study?

We will publish the results in a scientific journal. No participant will be identified in
any publication. We will send a letter summarising the results to the parents of the
babies who took part. At this stage should you wish to know which group your baby
was in we would be happy to provide you this information.
Who is organising and funding the research?

The research is being organised by Imperial College London. The study is being
funded by the Efficacy and Mechanism Programme of the National Institute for
Health Research.
Who has reviewed the study?

The study has been reviewed by independent doctors, specialists and parent
representatives. All research in the NHS is looked at by independent group of
people, called a National Research Ethics Committee, to protect the interests of
participants. This study has been reviewed and given favourable opinion by the
Hammersmith Hospital Research Ethics Committee
Further information and contact details

If you would like further information about the study please contact
Insert local PI details.
Nutritional Evaluation and Optimisation in Neonates Study: the NEON study.

Version 4: 28 Oct 2010
Nutritional Evaluation and Optimisation in Neonates Study: The NEON Study:
Information Sheet for parents on the Magnetic Resonance Scan:
78
We thank you for including your baby in the Nutritional Evaluation and Optimisation
in Neonates (NEON) Study. This information sheet gives you additional information
about the magnetic resonance (MR) scan which is the final part of this study.
As your baby is now preparing to go home we are in a position to arrange the MR

scan which will look at how the body, brain and liver have developed. The results of
this scan will be compared between the babies who have received different
treatments in order to see whether or not one treatment has any benefits over the
other. The scan will be done within roughly 2 weeks of your baby going home.
You will need to travel to the Hammersmith Hospital for this scan. We will arrange
transport for you and your baby to and from the hospital or reimburse you for parking
if you choose to drive yourself.
MR imaging is a technique widely used in infants and we have studied several

hundred infants with MR. A MR scanner uses a magnet to take detailed pictures of
the body and brain and measures the amount of fat in the liver.
The scan is carried out whilst your baby is in natural sleep without the use of
sedatives. The scan normally takes no more than 40 minutes but sometimes
additional time is required to settle a baby. You are welcome to be in the adjacent
control room and watch your baby during the scan. During the scan your baby will be
under the care of a doctor. As the MR scanner makes some noise we use baby ear
muffs to protect your baby’s ears. After the scan is complete we will measure your
baby’s growth and blood pressure.
We will be happy to show you the pictures taken of your baby. The scan is not being
carried out for clinical diagnosis but there is a possibility that they might show
something unexpected. If this occurs, a senior doctor will explain this to you and
notify your GP, and discuss whether any further action is necessary. The brain scan
however will be reported and the results will be sent to your baby’s doctor who will be
able to discuss this with you.
NEON Study Information Sheet for parents on MR scan Version 1 261109
79
APPENDIX 2
CONSENT FORM FOR PARENTS

Version 2 dated 28th October 2010
Study title: Amino acid regimen and intravenous lipid composition in preterm
parenteral nutrition: a randomised controlled trial of Nutritional Evaluation and
Optimisation in Neonates (NEON)
Patient’s name and hospital sticker
Please initial
The parent should complete this sheet himself or herself.
boxes
I confirm that I have read and understand the parents information sheet
dated 28th October 2010 (version 4) for the above study. I have had the
1.
opportunity to consider the information, ask questions and have had
these answered satisfactorily.
I understand that I am free to withdraw my baby from the study at any

2. time without giving any reason, without my baby’s medical care or legal
rights being affected.
I understand that relevant sections of any of my baby’s medical

notes and data collected during the study may be looked at by
responsible individuals from the Clinical Trials and Evaluation Unit
3.
or staff from regulatory authorities where it is relevant to my baby
taking part in research; I give permission for these individuals to
have access to my baby’s records.
I understand that routine information about my baby’s care may
be collected for the purposes of the study if my baby is transferred
4
to another hospital prior to discharge home. I agree to this
information being collected.
I agree to be contacted in the future to be informed about follow up

5.
studies that may take place.
6. I agree to my baby being included in the above study.
__________________ ___________ __________________

NAME IN BLOCK CAPITALS Date Signature
Relationship to patient: ______________________________
Investigator’s signature ___________________________Date _____________
(INVESTIGATOR’S NAME IN BLOCK CAPITALS)__________________________
When completed, 1 for infant’s parent; 1 for researcher file; 1 (original) to be kept in medical notes
th
NEON Consent Form v2.0 dated 28 October 2010
80
EME
HS&DR
HTA
PGfAR
PHR
Part of the NIHR Journals Library
www.journalslibrary.nihr.ac.uk
This report presents independent research funded by the

National Institute for Health Research (NIHR). The views
expressed are those of the author(s) and not necessarily
those of the NHS, the NIHR or the Department of Health
Published by the NIHR Journals Library

Bookshelf NBK350515

Uploaded by

Copyright:

Available Formats

Bookshelf NBK350515

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bookshelf NBK350515

Uploaded by

Copyright:

Available Formats

EFFICACY AND MECHANISM EVALUATION

VOLUME 3 ISSUE 2 MARCH 2016

Nutritional Evaluation and Optimisation in Neonates

Sabita Uthaya,1,2* Xinxue Liu,3 Daphne Babalis,3,4

Foundation Trust, London, UK

Clinical Science Centre, Imperial College London, London, UK

Council Clinical Sciences Centre, Hammersmith Hospital, London, UK

Published March 2016

This report should be referenced as follows:

ISSN 2050-4373 (Online)

Editorial contact: nihredit@southampton.ac.uk

Criteria for inclusion in the Efficacy and Mechanism Evaluation journal

NIHR Journals Library Editor-in-Chief

NIHR Journals Library Editors

Dr Martin Ashton-Key Consultant in Public Health Medicine/Consultant Advisor, NETSCC, UK

Dr Tessa Crilly Director, Crystal Blue Consulting Ltd, UK

Dr Peter Davidson Director of NETSCC, HTA, UK

Ms Tara Lamont Scientific Advisor, NETSCC, UK

Professor John Norrie Health Services Research Unit, University of Aberdeen, UK

Dr Rob Riemsma Reviews Manager, Kleijnen Systematic Reviews Ltd, UK

Editorial contact: nihredit@southampton.ac.uk

NIHR Journals Library www.journalslibrary.nihr.ac.uk

Nutritional Evaluation and Optimisation in Neonates (NEON)

Sabita Uthaya,1,2* Xinxue Liu,3 Daphne Babalis,3,4 Caroline Dore,5

Centre, Imperial College London, London, UK

Sciences Centre, Hammersmith Hospital, London, UK

*Corresponding author s.uthaya@imperial.ac.uk

List of abbreviations xiii

Plain English summary xv

Scientific summary xvii

Chapter 2 Research objectives 7

Chapter 6 Conclusions and recommendations 63

Appendix 1 Distribution of primary and secondary outcomes after transformation 73

Appendix 2 Parent information sheet, magnetic resonance information sheet

TABLE 2 Summary of tests and investigations 12

TABLE 3 Definitions for assessment of causality 16

TABLE 4 Definitions of SpAEs including thresholds for reporting to the DMEC 17

TABLE 5 Baseline characteristics for all infants randomised 28

TABLE 6 Baseline characteristics for all infants completing MRI assessment 29

TABLE 16 Safety data: summary of laboratory AEs by treatment for all

TABLE 18 Safety data: summary of SAEs by treatment 47

TABLE 19 Baseline characteristics and trial outcomes (all infants completing

TABLE 20 Adipose tissue compartments in litres for all infants completing

TABLE 21 Summary of the covariates for secondary analysis 52

TABLE 22 Primary outcomes (pairwise comparisons) 53

FIGURE 2 The Consolidated Standards of Reporting Trials diagram 23

FIGURE 3 Summary of screening data for all trial sites 24

FIGURE 4 Cumulative recruitment vs. target recruitment for the duration

FIGURE 6 Total recruitment and retention per centre 27

FIGURE 8 Time trend for babies’ weights across four groups 54

FIGURE 17 Distribution of primary and secondary outcomes after transformation 73

IMP investigational medicinal product SAE serious adverse event

Inc-AA incremental amino acids SD standard deviation

Plain English summary

Amino acid intervention

l greater accrual of non-adipose (lean) body mass at term (primary objective)

l reduced IHCL content at term age equivalent (primary objective)

There were four randomised groups:

1. Inc-AA and 20% Intralipid (Inc-AA/Intralipid)

l Major congenital or life-threatening abnormalities.