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

Notes

Permissions

Chapter 1 Introduction

Chapter 2 Research objectives

The two main research objectives were to study the effects of two parenteral nutrition interventions (amino acid quantity and lipid composition) in extremely preterm infants.

Chapter 3 Methods

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

⁽²⁾ AT mass , kg = ( AT volume , l ) × 0.90.

Adipose tissue mass was used to determine percentage AT as:

⁽³⁾ Percentage AT mass = ( AT mass , kg ) / ( body mass , kg ) .

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.

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

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.

Whenever possible, and with time permitting, the following brain scans were also undertaken:

• 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.

• 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.

• 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:

• total cerebral volume: sum of the volumes of basal ganglia, thalami (deep grey matter), cerebrospinal fluid, grey matter, white matter and lateral ventricles

• whole-brain volume: sum of the volumes of basal ganglia, thalami (deep grey matter), grey matter and white matter

• 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.

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.

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.

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.

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:

1. Identifying a suitable manufacturer for the trial with an IMP licence to produce PN and the capacity to support the trial.

2. 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.

3. Obtaining NHS permission at each site was lengthy, the procurement process was a factor for this.

4. 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.

FIGURE 4.

Cumulative recruitment vs. target recruitment for the duration of the trial.

FIGURE 5.

Cumulative retention (number of MR scans) vs. target retention for the duration of the trial.

FIGURE 6.

Total 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.

TABLE 5 - Baseline characteristics for all infants randomiseda

IQR, interquartile range.

Data presented are means (SD) for continuous variables and frequency (percentage) for categorical variables.

Anthropometries measured at booking.

Data presented are medians (IQR, lower quartile, upper quartile).

Characteristic	Inc-AA/Intralipid (N = 42)	Inc-AA/SMOFlipid (N = 42)	Imm-RDI/Intralipid (N = 41)	Imm-RDI/SMOFlipid (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), mean (SD)	25.3 (2.0); n = 41	25.0 (3.0); n = 40	25.3 (1.9); n = 37	25.6 (2.9); n = 39
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), mean (SD)	–0.5 (0.9); n = 41	–0.3 (1.0); n = 39	–0.7 (0.9); n = 37	–0.2 (1.6); n = 41
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
Mother’s height (cm),b 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
Father’s weight (kg),b 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, (hours),c median (IQR)	18.4 (12.3–22.7); n = 42	19.5 (13.6–22.8); n = 41	20.4 (12.6–23.6); n = 40	17.7 (13.0–22.4); n = 43

TABLE 6 - Baseline characteristics for all infants completing MRI assessmenta

IQR, interquartile range.

Data presented are means (SD) for continuous variables and frequency (percentage) for categorical variables.

Anthropometries measured at booking.

Data presented are medians (IQR, lower quartile, upper quartile).

Characteristic	Inc-AA/Intralipid (N = 34)	Inc-AA/SMOFlipid (N = 28)	Imm-RDI/Intralipid (N = 34)	Imm-RDI/SMOFlipid (N = 37)
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), mean (SD)	25.3 (2.0); n = 34	25.6 (2.6); n = 26	25.5 (1.9); n = 32	25.7 (2.9); n = 34
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), mean (SD)	–0.5 (0.9); n = 34	–0.4 (1.0); n = 26	–0.7 (0.9); n = 32	–0.2 (1.7); n = 34
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),b mean (SD)	162.4 (7.2); n = 26	165.1 (7.3); n = 19	162.1 (9.1); n = 23	164.8 (9.1); n = 28
Father’s weight (kg),b 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, (hours),c median (IQR)	16.9 (10.5–22.3); n = 34	19.4 (12.1–22.3); n = 28	20.0 (12.4–23.5); n = 34	17.7 (13.2–22.4); n = 37

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

CoNS, coagulase-negative staphylococci.

There can be more than one reason for each infant.

Growth of a known pathogen on culture; data presented are the number of infants who had at least one positive result.

Characteristic	Inc-AA/Intralipid (N = 42)	Inc-AA/SMOFlipid (N = 42)	Imm-RDI/Intralipid (N = 41)	Imm-RDI/SMOFlipid (N = 43)
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 intake of 150 ml/kg/day for 24 hours	12 (9–17.5); n = 32	11.5 (9–16); n = 28	11 (10–14); n = 30	13 (9.5–18); n = 36
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 investigator manual	0 (0)	1 (2.4)	0 (0)	0 (0)
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)
Positive blood culture,b 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 CoNS	3 (20.0)	4 (30.8)	1 (11.1)	5 (35.7)
Gram-positive cocci not specified	2 (13.3)	1 (7.7)	1 (11.1)	0 (0)
Positive blood cultures (while on PN)b	8	9	8	8
Length of stay in hospital (days), median (IQR)	69.5 (52–95); n = 38	61 (45–88); n = 33	63 (45–95); n = 38	66.5 (44–98); n = 38

TABLE 8 - Parenteral nutrition details and blood culture results for all infants completing MRI assessment

CoNS, coagulase-negative staphylococci.

There can be more than one reason for each infant.

Growth of a known pathogen on culture; data presented are the number of infants who had at least one positive result.

Characteristic	Inc-AA/Intralipid (N = 34)	Inc-AA/SMOFlipid (N = 28)	Imm-RDI/Intralipid (N = 34)	Imm-RDI/SMOFlipid (N = 37)
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 intake of 150 ml/kg/day for 24 hours, median (IQR)	11 (9–16); n = 28	11.5 (9–16); n = 22	11 (10–13.5); n = 28	13 (10–18); n = 33
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 CoNS	2 (22.2)	0 (0)	0 (0)	5 (41.7)
Gram-positive cocci not specified	1 (11.1)	1 (25.0)	1 (20.0)	0 (0)
Positive blood cultures (while on PN)b	5	3	4	7
Length of stay in hospital (days), median (IQR)	69.5 (55–96); n = 34	59 (44–85); n = 28	60.5 (44–88); n = 34	67 (47–98.5); n = 36

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.

TABLE 9 - Trial PN intake during the first 7 days for all infants randomised

Day 1 is defined from birth to first 17.00.

Trial PN intake by day	Inc-AA/Intralipid (N = 42)	Inc-AA/SMOFlipid (N = 42)	Imm-RDI/Intralipid (N = 41)	Imm-RDI/SMOFlipid (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
Aqueous volume (ml/kg)	96.5 (20.8)	89.9 (31.1)	94.9 (16.8)	94.5 (20.9)
Lipid volume (ml/kg)	14.8 (9.9)	12.8 (5.2)	13.7 (2.6)	14.1 (4.0)
Protein (g/kg)	2.1 (0.5)	1.9 (0.7)	3.1 (0.5)	3.1 (0.7)
Carbohydrate (g/kg)	8.6 (1.8)	8.1 (2.8)	8.4 (1.4)	8.2 (1.8)
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
Aqueous volume (ml/kg)	114.4 (22.4)	112.8 (29.1)	112.6 (25.1)	114.0 (23.2)
Lipid volume (ml/kg)	14.3 (3.2)	14.7 (3.8)	13.4 (3.8)	13.4 (4.8)
Protein (g/kg)	2.5 (0.5)	2.5 (0.7)	3.1 (0.7)	3.1 (0.7)
Carbohydrate (g/kg)	8.5 (1.8)	8.3 (2.1)	8.4 (1.8)	8.3 (1.8)
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
Aqueous volume (ml/kg)	111.4 (26.0)	117.0 (21.5)	123.4 (14.7)	115.9 (20.2)
Lipid volume (ml/kg)	14.2 (5.3)	14.6 (4.2)	15.3 (4.4)	14.0 (4.1)
Protein (g/kg)	2.5 (0.6)	2.6 (0.5)	3.3 (0.4)	3.1 (0.5)
Carbohydrate (g/kg)	8.1 (2.1)	8.4 (1.6)	8.9 (1.1)	8.4 (1.4)
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
Aqueous volume (ml/kg)	106.5 (29.9)	111.3 (23.7)	114.0 (25.1)	112.4 (22.1)
Lipid volume (ml/kg)	14.4 (4.7)	14.9 (4.3)	14.8 (4.1)	13.7 (4.7)
Protein (g/kg)	2.4 (0.7)	2.5 (0.5)	3.0 (0.7)	3.0 (0.6)
Carbohydrate (g/kg)	7.7 (2.2)	8.1 (1.6)	8.2 (1.8)	8.1 (1.6)
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
Aqueous volume (ml/kg)	100.9 (34.0)	103.0 (31.6)	103.0 (31.3)	107.5 (27.8)
Lipid volume (ml/kg)	13.1 (5.4)	13.8 (4.9)	13.2 (6.0)	12.8 (5.7)
Protein (g/kg)	2.3 (0.8)	2.3 (0.7)	2.8 (0.8)	2.9 (0.7)
Carbohydrate (g/kg)	7.3 (2.5)	7.4 (2.3)	7.4 (2.3)	7.7 (2.0)
Fat (g/kg)	2.3 (1.0)	2.4 (0.9)	2.3 (1.0)	2.3 (1.0)
Day 7, mean (SD)	n = 41	n = 36	n = 36	n = 38
Aqueous volume (ml/kg)	93.4 (35.7)	92.4 (31.3)	101.3 (28.2)	100.9 (25.6)
Lipid volume (ml/kg)	11.1 (6.1)	11.1 (6.0)	13.1 (7.5)	11.6 (5.6)
Protein (g/kg)	2.1 (0.8)	2.1 (0.7)	2.7 (0.8)	2.7 (0.7)
Carbohydrate (g/kg)	6.8 (2.6)	6.7 (2.3)	7.3 (2.0)	7.3 (1.8)
Fat (g/kg)	2.0 (1.1)	1.9 (1.0)	2.3 (1.3)	2.1 (1.0)

TABLE 10 - Trial PN intake during the first 7 days for all infants completing MRI assessment

Day 1 is defined from birth to first 17.00.

Trial PN intake by day	Inc-AA/Intralipid (N = 34)	Inc-AA/SMOFlipid (N = 28)	Imm-RDI/Intralipid (N = 34)	Imm-RDI/SMOFlipid (N = 37)
Day 1,a mean (SD)	n = 32	n = 25	n = 31	n = 36
Aqueous volume (ml/kg)	69.5 (36.2)	70.1 (34.4)	70.9 (37.6)	67.8 (36.0)
Lipid volume (ml/kg)	8.4 (5.0)	8.7 (4.5)	8.9 (5.5)	8.1 (4.7)
Protein (g/kg)	1.2 (0.7)	1.2 (0.6)	2.5 (1.3)	2.4 (1.3)
Carbohydrate (g/kg)	6.7 (3.5)	6.7 (3.3)	6.8 (3.6)	6.4 (3.4)
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
Aqueous volume (ml/kg)	92.9 (18.5)	89.5 (22.6)	97.3 (15.3)	97.1 (20.2)
Lipid volume (ml/kg)	14.7 (10.7)	13.2 (4.4)	13.8 (2.6)	14.6 (3.8)
Protein (g/kg)	2.0 (0.4)	1.9 (0.5)	3.2 (0.5)	3.1 (0.6)
Carbohydrate (g/kg)	8.3 (1.7)	8.0 (1.9)	8.5 (1.3)	8.4 (1.7)
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
Aqueous volume (ml/kg)	116.8 (21.5)	112.2 (32.7)	111.8 (26.3)	117.5 (19.2)
Lipid volume (ml/kg)	14.6 (3.4)	14.4 (4.3)	13.3 (4.0)	13.9 (4.6)
Protein (g/kg)	2.6 (0.5)	2.5 (0.7)	3.1 (0.7)	3.2 (0.6)
Carbohydrate (g/kg)	8.6 (1.8)	8.3 (2.4)	8.4 (1.9)	8.6 (1.6)
Fat (g/kg)	2.8 (0.6)	2.8 (0.8)	2.6 (0.8)	2.7 (0.9)
Day 4, mean (SD)	n = 33	n = 28	n = 34	n = 35
Aqueous volume (ml/kg)	112.4 (26.8)	117.4 (24.4)	123.2 (14.9)	118.5 (14.6)
Lipid volume (ml/kg)	13.9 (5.6)	14.8 (4.7)	15.6 (3.7)	14.4 (3.5)
Protein (g/kg)	2.5 (0.6)	2.6 (0.5)	3.3 (0.4)	3.2 (0.3)
Carbohydrate (g/kg)	8.2 (2.1)	8.5 (1.8)	8.9 (1.1)	8.6 (0.9)
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
Aqueous volume (ml/kg)	107.6 (32.1)	111.0 (25.2)	113.3 (26.2)	112.3 (23.1)
Lipid volume (ml/kg)	14.3 (5.1)	15.2 (4.1)	14.7 (4.3)	13.5 (5.0)
Protein (g/kg)	2.4 (0.7)	2.5 (0.6)	3.0 (0.7)	3.0 (0.6)
Carbohydrate (g/kg)	7.7 (2.3)	8.0 (1.8)	8.2 (1.9)	8.1 (1.7)
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
Aqueous volume (ml/kg)	101.4 (36.8)	104.6 (33.7)	101.5 (32.5)	111.0 (24.1)
Lipid volume (ml/kg)	13.1 (5.8)	13.9 (5.4)	13.0 (6.1)	13.2 (5.6)
Protein (g/kg)	2.3 (0.8)	2.4 (0.8)	2.7 (0.9)	3.0 (0.6)
Carbohydrate (g/kg)	7.3 (2.7)	7.5 (2.4)	7.3 (2.3)	8.0 (1.7)
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
Aqueous volume (ml/kg)	93.8 (35.6)	93.2 (34.8)	101.9 (29.2)	103.1 (26.3)
Lipid volume (ml/kg)	11.3 (6.0)	11.4 (6.3)	13.3 (7.8)	12.3 (5.4)
Protein (g/kg)	2.1 (0.8)	2.1 (0.8)	2.7 (0.8)	2.8 (0.7)
Carbohydrate (g/kg)	6.8 (2.6)	6.7 (2.5)	7.3 (2.1)	7.4 (1.9)
Fat (g/kg)	2.0 (1.1)	2.0 (1.1)	2.4 (1.4)	2.2 (1.0)

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

Day 1 is defined from birth to first 17.00.

	Inc-AA/Intralipid (N = 42)	Inc-AA/SMOFlipid (N = 42)	Imm-RDI/Intralipid (N = 41)	Imm-RDI/SMOFlipid (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)
Carbohydrate (g/kg)	7.2 (3.6)	6.6 (3.6)	6.6 (3.9)	6.8 (3.6)
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)
Carbohydrate (g/kg)	9.1 (2.0)	8.7 (3.4)	9.1 (1.6)	9.0 (2.1)
Fat (g/kg)	3.1 (2.0)	2.9 (1.3)	3.0 (0.7)	3.1 (1.0)
Total energy (kcal/kg)	73.0 (21.7)	68.5 (27.4)	76.8 (13.4)	77.1 (18.7)
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)
Carbohydrate (g/kg)	9.7 (2.3)	9.6 (3.0)	9.8 (2.5)	9.7 (2.5)
Fat (g/kg)	3.2 (1.0)	3.5 (1.3)	3.3 (1.1)	3.3 (1.3)
Total energy (kcal/kg)	78.1 (17.5)	81.2 (26.0)	82.3 (21.9)	82.3 (23.1)
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)
Carbohydrate (g/kg)	9.8 (2.8)	10.3 (2.9)	10.7 (1.8)	9.9 (3.1)
Fat (g/kg)	3.3 (1.3)	3.7 (1.2)	3.7 (1.2)	3.3 (1.4)
Total energy (kcal/kg)	79.4 (22.5)	86.4 (24.4)	91.0 (18.7)	83.2 (28.6)
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)
Carbohydrate (g/kg)	9.9 (3.0)	10.8 (2.7)	11.1 (2.7)	11.0 (2.8)
Fat (g/kg)	3.7 (1.4)	4.1 (1.3)	4.1 (1.5)	3.7 (1.2)
Total energy (kcal/kg)	84.3 (27.1)	92.8 (24.9)	95.8 (26.9)	91.2 (20.1)
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)
Carbohydrate (g/kg)	10.4 (2.7)	11.2 (2.6)	11.2 (2.4)	10.9 (2.7)
Fat (g/kg)	3.9 (1.6)	4.4 (1.3)	4.3 (1.6)	3.9 (1.7)
Total energy (kcal/kg)	89.2 (26.6)	97.2 (24.2)	97.7 (25.3)	93.2 (28.1)
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)
Carbohydrate (g/kg)	10.5 (2.5)	11.0 (2.4)	11.5 (2.1)	11.2 (2.5)
Fat (g/kg)	3.9 (1.6)	4.3 (1.6)	4.6 (1.7)	4.1 (1.8)
Total energy (kcal/kg)	89.2 (25.2)	94.8 (24.6)	102.1 (23.3)	96.0 (27.6)
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)
Total energy (kcal)	3105 (1193)	2871 (1521)	3215 (1308)	3177 (1419)
Until 34 weeks postmenstrual age, mean (SD)	n = 42	n = 41	n = 40	n = 43
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)
Total energy (kcal)	5321 (1890)	4424 (2123)	4772 (1784)	5001 (1768)

TABLE 12 - Nutritional intake over the first 2 weeks for all infants randomiseda

Data presented are medians (IQR, lower quartile, upper quartile).

	Inc-AA/Intralipid (N = 42)	Inc-AA/SMOFlipid (N = 42)	Imm-RDI/Intralipid (N = 41)	Imm-RDI/SMOFlipid (N = 43)
Had maternal expressed breast milk, n (%)	40 (95.2)	38 (90.5)	38 (92.7)	42 (97.7)
Cumulative maternal expressed breast milk (l), median (IQR)	0.76 (0.42–1.56); n = 40	0.94 (0.21–1.33); n = 38	0.68 (0.34–1.44); n = 38	0.59 (0.17–1.11); n = 42
Had formula milk, n (%)	12 (28.6)	11 (26.2)	16 (39.0)	13 (30.2)
Cumulative formula intake (l), median (IQR)	0.03 (0.01–0.29); n = 12	0.16 (0.04–1.10); n = 11	0.47 (0.03–1.00); n = 16	0.37 (0.02–1.56); n = 13
Had trial PN, n (%)	42 (100.0)	41 (97.6)	40 (97.6)	43 (100.0)
Time on trial PN (days), median (IQR)	11.0 (8.0–14.0); n = 42	12.0 (9.0–14.0); n = 41	11.0 (9.0–13.0); n = 40	12.0 (9.0–14.0); n = 43
Cumulative trial PN intake, median (IQR)
Aqueous volume (l)	1.02 (0.74–1.29); n = 42	0.93 (0.70–1.30); n = 41	0.93 (0.81–1.18); n = 40	1.04 (0.81–1.36); n = 43
Lipid volume (l)	0.13 (0.09–0.17); n = 42	0.11 (0.08–0.17); n = 41	0.12 (0.10–0.15); n = 40	0.13 (0.09–0.18); n = 43
Protein (g)	22.4 (16.0–28.4); n = 42	20.9 (15.3–28.4); n = 41	25.9 (22.6–32.5); n = 40	29.5 (23.2–37.2); n = 43
Carbohydrate (g)	76.6 (57.7–96.1);n = 42	69.3 (53.1–96.6); n = 41	69.8 (61.1–87.8); n = 40	79.5 (62.7–100.4); n = 43
Fat (g)	23.6 (16.7–31.6); n = 42	21.0 (15.1–31.4); n = 41	21.4 (17.5–27.7); n = 40	24.6 (16.7–31.8); 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), median (IQR)	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
Cumulative non-trial PN intake, median (IQR)
Aqueous volume (l)	0.46 (0.40–0.47); n = 3	0.02 (0.02–0.02); n = 1	0.07 (0.07–0.07); n = 1	0.37 (0.20–0.44); n = 3
Lipid volume (l)	0.02 (0.01–0.04); n = 3	0.00 (0.00–0.00); n = 1	0.00 (0.00–0.00); n = 1	0.03 (0.01–0.03); n = 3
Protein (g)	1.34 (1.17–1.54); n = 3	0.07 (0.07–0.07); n = 1	0.15 (0.15–0.15); n = 1	1.04 (0.57–1.21); n = 3
Carbohydrate (g)	42.57 (40.82–45.99); n = 3	2.47 (2.47–2.47); n = 1	3.35 (3.35–3.35); n = 1	29.75 (17.04–35.48); n = 3
Fat (g)	4.40 (2.20–7.90); n = 3	0.30 (0.30–0.30); n = 1	0.53 (0.53–0.53); n = 1	4.44 (2.61–5.85); n = 3
Cumulative non-trial and trial PN intake, median (IQR)
Aqueous volume (l)	1.04 (0.81–1.29); n = 42	0.93 (0.70–10.30); n = 41	0.94 (0.81–1.18); n = 40	1.04 (0.81–1.36); n = 43
Lipid volume (l)	0.13 (0.09–0.17); n = 42	0.11 (0.08–0.17); n = 41	0.12 (0.10–0.15); n = 40	0.13 (0.09–0.18); n = 43
Protein (g)	22.4 (16.0–28.4); n = 42	20.9 (15.3–28.4); n = 41	25.9 (22.6–32.5); n = 40	29.5 (23.2–37.2); n = 43
Carbohydrate (g)	78.0 (61.8–96.1); n = 42	69.3 (53.1–96.6); n = 41	70.3 (61.1–87.8); n = 40	79.5 (62.7–100.4); n = 43
Fat (g)	23.6 (16.8–31.6); n = 42	21.0 (15.1–31.4); n = 41	21.4 (17.5–27.7); n = 40	24.6 (16.7–31.8); n = 43

TABLE 13 - Nutritional intake during the study period for all infants randomiseda

Data presented are medians (IQR, lower quartile, upper quartile) for continuous variables and frequency (percentage) for categorical variables.

	Inc-AA/Intralipid (N = 42)	Inc-AA/SMOFlipid (N = 42)	Imm-RDI/Intralipid (N = 41)	Imm-RDI/SMOFlipid (N = 43)
Glucose and insulin, median (IQR)
Cumulative intravenous dextrose (g)	15.2 (5.39–50.2); n = 42	18.1 (7.20–54.0); n = 41	11.8 (5.23–45.7); n = 40	22.2 (7.99–55.5); 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); n = 14	7.46 (5.12–18.4); n = 14	12.18 (9.01–42.7); n = 9	10.01 (5.44–22.5); n = 14
Additional potassium (mmol)	9.97 (6.79–11.73); n = 8	2.73 (1.47–4.88); n = 8	2.93 (1.73–8.24); n = 8	3.72 (1.64–7.14); n = 10
Received donor milk, n (%)	15 (35.7)	12 (28.6)	13 (31.7)	17 (39.5)
Cumulative donor milk (l, total volume per baby), median (IQR)	0.38 (0.08–1.07); n = 15	0.30 (0.02–0.47); n = 12	0.38 (0.05–1.10); n = 13	0.19 (0.03–0.36); n = 17
Received maternal breast milk, n (%)	41 (97.6)	38 (90.5)	38 (92.7)	43 (100.0)
Cumulative maternal expressed breast milk (l, total volume per baby), median (IQR)
During trial PN phase	0.72 (0.45–0.88); n = 41	0.61 (0.43–0.81); n = 38	0.57 (0.30–0.81); n = 38	0.63 (0.26–0.89); n = 43
During non-trial PN phase	1.92 (0.38–4.82); n = 8	4.01 (1.83–5.08); n = 8	1.79 (0.44–4.51); n = 8	2.78 (0.70–4.50); n = 8
Over study period	8.05 (4.50–13.2); n = 41	7.16 (1.94–12.1); n = 38	6.74 (1.27–12.1); n = 38	7.61 (1.73–15.7); n = 43
Number of days having Fortifier, median (IQR)	13.0 (10.0–24.0); n = 23	23.0 (14.5–41.8); n = 12	14.0 (1.0–29.0); n = 17	27.0 (13.0–44.0); 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)
During trial PN phase	0.02 (0.01–0.13); n = 10	0.24 (0.02–0.52); n = 10	0.28 (0.04–0.60); n = 15	0.30 (0.04–0.39); n = 14
During non-trial PN phase	2.05 (0.44–3.60); n = 6	2.09 (0.00–4.35); n = 4	0.78 (0.00–2.12); n = 7	2.30 (0.95–5.06); n = 6
Over study period	5.86 (1.22–10.87); n = 30	6.39 (0.41–9.36); n = 23	8.44 (2.21–13.04); n = 29	6.70 (3.86–12.63); n = 29
Received trial PN, n (%)	42 (100.0)	41 (97.6)	40 (97.6)	43 (100.0)
Time on trial PN (days), median (IQR)	11.0 (8.00–16.8); n = 42	12.0 (9.00–17.0); n = 41	11.0 (9.75–13.2); n = 40	12.0 (9.00–17.0); n = 43
Cumulative trial PN intake
Aqueous volume (l)	1.04 (0.77–1.56); n = 42	0.97 (0.71–1.39); n = 41	0.95 (0.81–1.33); n = 40	1.19 (0.86–1.56); n = 43
Lipid volume (l)	0.13 (0.09–0.19); n = 42	0.13 (0.08–0.18); n = 41	0.12 (0.10–0.16); n = 40	0.14 (0.10–0.19); n = 43
During trial PN phase, median (IQR)
Protein (g)	22.8 (16.9–34.4); n = 42	21.6 (15.3–30.5); n = 41	26.7 (23.2–37.5); n = 40	33.5 (24.0–42.5); n = 43
Carbohydrate (g)	78.0 (57.7–116.0); n = 42	72.7 (55.0–104.8); n = 41	71.9 (62.6–101.3); n = 40	90.4 (64.8–114.6); n = 43
Fat (g)	23.9 (16.7–35.7); n = 42	23.6 (15.1–33.0); n = 41	21.6 (18.0–29.5); n = 40	25.0 (18.6–35.0); n = 43
Received non-trial PN, n (%)	10 (23.8)	8 (19.0)	8 (19.5)	10 (23.3)
Time on non-trial PN (days), median (IQR)	5.5 (3.25–19.8); n = 10	22.0 (14.50–28.5); n = 8	28.5 (9.50–58.5); n = 8	12.0 (3.00–15.8); n = 10
Cumulative non-trial PN intake, median (IQR)
Aqueous volume (l)	0.47 (0.39–2.44); n = 10	2.12 (1.89–2.37); n = 8	2.91 (0.87–9.50); n = 8	1.23 (0.47–2.44); n = 10
Lipid volume (l)	0.06 (0.02–0.26); n = 10	0.33 (0.32–0.39); n = 8	0.38 (0.13–1.40); n = 8	0.16 (0.05–0.26); n = 10
During trial PN phase, median (IQR)
Protein (g)	0.0 (0.0–0.00); n = 10	0.0 (0.0–0.00); n = 8	0.0 (0.0–0.04); n = 8	0.0 (0.0–0.07); n = 10
Carbohydrate (g)	0.0 (0.0–0.00); n = 10	0.0 (0.0–0.00); n = 8	0.0 (0.0–0.84); n = 8	0.0 (0.0–3.25); n = 10
Fat (g)	0.0 (0.0–0.00); n = 10	0.0 (0.0–0.00); n = 8	0.0 (0.0–0.13); n = 8	0.0 (0.0–0.59); n = 10
Over study period, median (IQR)
Protein (g)	1.65 (1.34–6.54); n = 10	7.32 (5.60–7.66); n = 8	10.76 (4.60–31.95); n = 8	3.09 (0.39–4.28); n = 10
Carbohydrate (g)	49.4 (41.1–220.6); n = 10	222.7 (188.6–265.4); n = 8	338.7 (166.7–1136.4); n = 8	108.6 (11.3–146.3); n = 10
Fat (g)	10.5 (4.04–46.5); n = 10	58.8 (56.76–69.1); n = 8	66.0 (23.68–247.1); n = 8	28.3 (8.78–46.4); n = 10
Cumulative non-trial and trial PN intake, median (IQR)
Aqueous volume (l)	1.12 (0.85–1.85); n = 42	1.30 (0.83–1.64); n = 41	1.06 (0.87–1.48); n = 40	1.38 (0.95–2.04); n = 43
Lipid volume (l)	0.17 (0.10–0.24); n = 42	0.16 (0.10–0.21); n = 41	0.13 (0.11–0.20); n = 40	0.18 (0.12–0.24); n = 43
During trial PN phase, median (IQR)
Protein (g)	22.8 (16.9–34.4); n = 42	21.6 (15.3–30.5); n = 41	26.7 (23.2–37.5); n = 40	33.5 (24.0–42.5); n = 43
Carbohydrate (g)	78.0 (60.9–116.0); n = 42	72.7 (55.0–104.8); n = 41	71.9 (62.6–101.3); n = 40	90.4 (66.3–114.6); n = 43
Fat (g)	23.9 (16.8–35.7); n = 42	23.6 (15.1–33.0); n = 41	21.6 (18.0–29.5); n = 40	25.0 (18.6–35.0); n = 43
Over study period, median (IQR)
Protein (g)	23.7 (18.0–36.4); n = 42	23.9 (17.8–31.6); n = 41	29.5 (24.1–38.4); n = 40	33.5 (25.2–42.5); n = 43
Carbohydrate (g)	87.8 (64.5–139.6); n = 42	94.8 (64.0–136.5); n = 41	81.0 (65.1–113.4); n = 40	101.4 (70.8–149.7); n = 43
Fat (g)	31.3 (18.8–43.6); n = 42	28.2 (19.0–37.4); n = 41	23.2 (19.3–36.4); n = 40	32.0 (22.6–44.1); 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); n = 42	119.0 (91.1–161.0); n = 41	124.8 (103.1–175.3); n = 40	148.3 (122.1–170.7); n = 43
Carbohydrate (g)	565.6 (446.6–756.7); n = 42	557.9 (338.4–673.2); n = 41	508.0 (399.7–682.4); n = 40	590.4 (440.4–680.9); n = 43
Fat (g)	272.8 (217.2–346.5); n = 42	246.1 (157.6–306.1); n = 41	235.7 (182.7–295.7); n = 40	261.5 (201.9–313.0); n = 43

TABLE 14 - Nutritional intake over first 2 weeks for all infants completing MRI assessmenta

Data presented are medians (IQR, lower quartile, upper quartile).

	Inc-AA/Intralipid (N = 34)	Inc-AA/SMOFlipid (N = 28)	Imm-RDI/Intralipid (N = 34)	Imm-RDI/SMOFlipid (N = 37)
Had maternal expressed breast milk, n (%)	32 (94.1)	27 (96.4)	33 (97.1)	36 (97.3)
Cumulative maternal expressed breast milk (l), median (IQR)	0.79 (0.60–1.63); n = 32	0.97 (0.54–1.62); n = 27	0.77 (0.42–1.54); n = 33	0.76 (0.29–1.14); n = 36
Had formula milk, n (%)	10 (29.4)	9 (32.1)	14 (41.2)	11 (29.7)
Cumulative formula intake (l), median (IQR)	0.03 (0.01–0.19); n = 10	0.26 (0.09–1.11); n = 9	0.35 (0.03–1.12); n = 14	0.37 (0.05–1.32); n = 11
Had trial PN, n (%)	34 (100.0)	28 (100.0)	34 (100.0)	37 (100.0)
Time on trial PN (days), median (IQR)	11.0 (8.00–14.0); n = 34	12.0 (9.00–14.0); n = 28	11.0 (9.25–13.0); n = 34	12.0 (9.00–14.0); n = 37
Cumulative trial PN intake, median (IQR)
Aqueous volume (l)	1.02 (0.74–1.30); n = 34	1.03 (0.81–1.33); n = 28	0.94 (0.86–1.21); n = 34	1.07 (0.81–1.39); n = 37
Lipid volume (l)	0.13 (0.09–0.18); n = 34	0.13 (0.10–0.18); n = 28	0.12 (0.10–0.16); n = 34	0.13 (0.09–0.18); n = 37
Protein (g)	22.4 (16.0–28.9); n = 34	22.3 (17.6–29.1); n = 28	26.2 (23.9–33.2); n = 34	31.0 (23.0–37.9); n = 37
Carbohydrate (g)	76.6 (57.7–97.2); n = 34	79.1 (59.7–100.1); n = 28	70.6 (64.4–89.6); n = 34	83.7 (62.0–102.1); n = 37
Fat (g)	24.3 (17.0–32.2); n = 34	24.5 (18.1–32.2); n = 28	21.7 (18.1–28.4); n = 34	25.0 (17.1–32.9); n = 37
Received non-trial PN, n (%)	2 (5.9)	1 (3.6)	1 (2.9)	2 (5.4)
Time on non-trial PN (days)	4.0 (4.0–4.0); n = 2	1.0 (1.0–1.0); n = 1	1.0 (1.0–1.0); n = 1	5.0 (3.5–6.5); n = 2
Cumulative non-trial PN intake, median (IQR)
Aqueous volume (l)	0.47 (0.47–0.48); n = 2	0.02 (0.02–0.02); n = 1	0.07 (0.07–0.07); n = 1	0.44 (0.41–0.47); n = 2
Lipid volume (l)	0.01 (0.01–0.02); n = 2	0.00 (0.00–0.00); n = 1	0.00 (0.00–0.00); n = 1	0.03 (0.03–0.04); n = 2
Protein (g)	1.37 (1.18–1.55); n = 2	0.07 (0.07–0.07); n = 1	0.15 (0.15–0.15); n = 1	1.21 (1.12–1.29); n = 2
Carbohydrate (g)	45.99 (44.28–47.69); n = 2	2.47 (2.47–2.47); n = 1	3.35 (3.35–3.35); n = 1	35.48 (32.62–38.34); n = 2
Fat (g)	2.20 (1.10–3.30); n = 2	0.30 (0.30–0.30); n = 1	0.53 (0.53–0.53); n = 1	5.85 (5.14–6.56); n = 2
Cumulative non-trial and trial PN intake, median (IQR)
Aqueous volume (l)	1.04 (0.82–1.30); n = 34	1.03 (0.81–1.33); n = 28	0.95 (0.86–1.21); n = 34	1.07 (0.81–1.39); n = 37
Lipid volume (l)	0.13 (0.09–0.18); n = 34	0.13 (0.10–0.18); n = 28	0.12 (0.10–0.16); n = 34	0.13 (0.09–0.18); n = 37
Protein (g)	22.4 (16.0–28.9); n = 34	22.3 (17.6–29.1); n = 28	26.2 (23.9–33.2); n = 34	31.0 (23.0–37.9); n = 37
Carbohydrate (g)	78.0 (63.0–97.2); n = 34	79.1 (60.5–100.1); n = 28	71.7 (64.4–89.6); n = 34	83.7 (62.0–102.1); n = 37
Fat (g)	24.3 (17.0–32.2); n = 34	24.5 (18.4–32.2); n = 28	21.7 (18.1–28.4); n = 34	25.0 (17.1–32.9); n = 37

TABLE 15 - Nutritional intake during the study period for all infants completing MRI assessmenta

Data presented are medians (IQR, lower quartile, upper quartile) for continuous variables and frequency (percentage) for categorical variables.

	Inc-AA/Intralipid (N = 34)	Inc-AA/SMOFlipid (N = 28)	Imm-RDI/Intralipid (N = 34)	Imm-RDI/SMOFlipid (N = 37)
Glucose and insulin, median (IQR)
Cumulative intravenous dextrose (g), median (IQR)	9.27 (4.56–54.8); n = 34	23.64 (7.04–75.1); n = 28	11.81 (5.68–25.9); n = 34	22.20 (8.89–53.4); n = 37
Insulin (number of babies who received insulin), n (%)	7 (20.6)	6 (21.4)	4 (11.8)	7 (18.9)
Electrolytes, median (IQR)
Additional sodium (mmol)	20.93 (11.81–34.2); n = 10	11.13 (5.00–22.6); n = 11	9.85 (8.87–13.5); n = 7	10.01 (5.44–22.5); n = 14
Additional potassium (mmol)	10.38 (6.45–12.29); n = 7	3.71 (1.62–5.58); n = 7	2.18 (1.46–5.00); n = 7	4.19 (1.60–7.85); n = 9
Received donor milk, n (%)	11 (32.4)	9 (32.1)	12 (35.3)	15 (40.5)
Cumulative donor milk (l, total volume per baby), median (IQR)	0.38 (0.08–1.07); n = 11	0.27 (0.02–0.44); n = 9	0.24 (0.04–1.15); n = 12	0.19 (0.03–0.84); n = 15
Received maternal expressed breast milk, n (%)	33 (97.1)	27 (96.4)	33 (97.1)	37 (100.0)
Cumulative maternal expressed breast milk (l), median (IQR)
During trial PN phase	0.73 (0.57–0.88); n = 33	0.62 (0.45–0.83); n = 27	0.62 (0.32–0.82); n = 33	0.65 (0.36–0.90); n = 37
During non-trial PN phase	1.75 (0.33–4.46); n = 5	4.16 (2.86–5.13); n = 7	1.79 (0.50–3.16); n = 6	2.78 (0.70–4.50); n = 8
Over study period	8.46 (5.90–14.6); n = 33	7.86 (2.79–13.5); n = 27	5.97 (1.53–12.3); n = 33	8.86 (3.32–15.9); n = 37
Number of days having fortifier, median (IQR)	13.0 (10.0–21.0); n = 21	20.0 (16.0–36.0); n = 9	13.0 (1.0–28.2); n = 16	25.5 (13.0–45.0); n = 16
Received formula milk, n (%)	25 (73.5)	19 (67.9)	25 (73.5)	26 (70.3)
Cumulative formula intake (l), median (IQR)
During trial PN phase	0.02 (0.00–0.16); n = 9	0.41 (0.13–0.53); n = 8	0.13 (0.04–0.66); n = 13	0.28 (0.05–0.38); n = 11
During non-trial PN phase	2.34 (1.76–4.02); n = 5	2.09 (0.00–4.35); n = 4	0.78 (0.00–3.45); n = 5	2.30 (0.95–5.06); n = 6
Over study period	2.81 (0.99–14.37); n = 25	8.04 (2.96–9.93); n = 19	8.44 (2.21–13.04); n = 25	7.26 (3.90–12.61); n = 26
Received trial PN (number of babies), n (%)	34 (100.0)	28 (100.0)	34 (100.0)	37 (100.0)
Time on trial PN (days), median (IQR)	11.0 (8.0–16.0); n = 34	12.5 (9.0–16.0); n = 28	11.0 (10.0–13.0); n = 34	13.0 (9.0–17.0); n = 37
Cumulative trial PN intake, median (IQR)
Aqueous volume (l)	1.04 (0.77–1.56); n = 34	1.06 (0.83–1.39); n = 28	0.96 (0.86–1.32); n = 34	1.19 (0.82–1.57); n = 37
Lipid volume (l)	0.14 (0.09–0.19); n = 34	0.13 (0.10–0.19); n = 28	0.12 (0.10–0.16); n = 34	0.14 (0.10–0.20); n = 37
During trial PN phase, median (IQR)
Protein (g)	22.8 (16.9–34.4); n = 34	23.0 (18.2–30.6); n = 28	27.2 (23.9–36.9); n = 34	33.5 (23.5–42.5); n = 37
Carbohydrate (g)	78.0 (57.7–118.5); n = 34	81.6 (62.1–105.2); n = 28	73.4 (64.4–99.6); n = 34	90.4 (63.4–114.6); n = 37
Fat (g)	25.0 (17.0–35.7); n = 34	24.9 (18.1–33.9); n = 28	21.7 (19.1–28.4); n = 34	25.0 (19.0–35.6); n = 37
Received non-trial PN (number of babies), n (%)	6 (17.6)	7 (25.0)	6 (17.6)	9 (24.3)
Time on non-trial PN (days), median (IQR)	13.5 (4.0–27.5); n = 6	19.0 (13.0–26.0); n = 7	43.5 (25.8–69.5); n = 6	12.0 (3.0–16.0); n = 9
Cumulative non-trial PN intake, median (IQR)
Aqueous volume (l)	1.77 (0.40–3.76); n = 6	2.13 (1.86–2.48); n = 7	6.26 (2.52–10.83); n = 6	1.24 (0.74–2.76); n = 9
Lipid volume (l)	0.18 (0.02–0.58); n = 6	0.33 (0.32–0.35); n = 7	0.90 (0.32–1.58); n = 6	0.17 (0.07–0.29); n = 9
During trial PN phase, median (IQR)
Protein (g)	0.0 (0.0–0.00); n = 6	0.0 (0.0–0.00); n = 7	0.0 (0.0–0.12); n = 6	0.0 (0.0–0.00); n = 9
Carbohydrate (g)	0.0 (0.0–0.00); n = 6	0.0 (0.0–0.00); n = 7	0.0 (0.0–2.51); n = 6	0.0 (0.0–0.00); n = 9
Fat (g)	0.0 (0.0–0.0); n = 6	0.0 (0.0–0.0); n = 7	0.0 (0.0–0.4); n = 6	0.0 (0.0–0.0); n = 9
Over study period, median (IQR)
Protein (g)	4.88 (1.44–12.63); n = 6	7.44 (4.66–7.69); n = 7	19.86 (6.68–37.40); n = 6	3.98 (1.07–4.28); n = 9
Carbohydrate (g)	159.4 (47.9–431.8); n = 6	242.2 (186.9–267.0); n = 7	701.5 (239.8–1339.7); n = 6	139.6 (30.7–147.6); n = 9
Fat (g)	31.2 (2.96–101.7); n = 6	58.2 (56.47–62.2); n = 7	159.0 (55.66–278.5); n = 6	30.4 (13.16–50.8); n = 9
Cumulative non-trial and trial PN intake, median (IQR)
Aqueous volume (l)	1.11 (0.85–1.85); n = 34	1.34 (0.96–1.80); n = 28	1.12 (0.89–1.43); n = 34	1.53 (0.97–2.05); n = 37
Lipid volume (l)	0.17 (0.10–0.25); n = 34	0.18 (0.13–0.25); n = 28	0.13 (0.11–0.20); n = 34	0.18 (0.13–0.25); n = 37
During trial PN phase, median (IQR)
Protein (g)	22.8 (16.9–34.4); n = 34	23.0 (18.2–30.6); n = 28	27.2 (23.9–36.9); n = 34	33.5 (23.5–42.5); n = 37
Carbohydrate (g)	78.0 (61.2–118.5); n = 34	81.6 (62.1–105.2); n = 28	73.4 (64.4–99.6); n = 34	92.7 (66.4–114.6); n = 37
Fat (g)	25.0 (17.0–35.7); n = 34	24.9 (18.4–33.9); n = 28	21.7 (19.1–28.4); n = 34	25.0 (19.0–35.6); n = 37
Over study period, median (IQR)
Protein (g)	23.7 (18.0–36.5); n = 34	26.1 (20.5–31.6); n = 28	30.3 (24.9–37.9); n = 34	34.4 (25.8–42.5); n = 37
Carbohydrate (g)	85.5 (64.5–139.6); n = 34	101.9 (72.2–145.1); n = 28	83.5 (67.2–110.6); n = 34	112.0 (71.9–152.7); n = 37
Fat (g)	30.3 (18.8–45.5); n = 34	31.8 (23.5–46.0); n = 28	23.6 (20.2–35.7); n = 34	32.9 (22.8–45.6); 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); n = 34	144.3 (115.5–165.4); n = 28	124.2 (104.5–176.6); n = 34	154.7 (133.1–171.7); n = 37
Carbohydrate (g)	565.6 (489.3–767.7); n = 34	596.2 (476.4–703.0); n = 28	508.0 (402.8–684.2); n = 34	607.7 (508.9–690.0); n = 37
Fat (g)	284.3 (235.7–370.6); n = 34	282.2 (233.8–320.1); n = 28	245.2 (185.1–295.2); n = 34	279.1 (213.3–323.2); n = 37

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.

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

ALT, alanine transferase; NA, not applicable.

Biochemical indices were measured while the infant was on PN as part of routine monitoring whether on trial or non-trial PN.

SpAE	Inc-AA/Intralipid, (N = 42)	Inc-AA/SMOFlipid (N = 42)	Imm-RDI/Intralipid (N = 41)	Imm-RDI/SMOFlipid (N = 43)	p-value for amino acid	p-value 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
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

TABLE 17 - Safety data: summary of laboratory AEs by treatment for all infants completing MRI assessment

ALT, alanine transferase; NA, not applicable.

SpAE	Inc-AA/Intralipid, (N = 34)	Inc-AA/SMOFlipid (N = 28)	Imm-RDI/Intralipid (N = 34)	Imm-RDI/SMOFlipid (N = 37)	p-value for amino acid	p-value for lipid
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
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

TABLE 18 - Safety data: summary of SAEs by treatmenta

All SAEs were classified by local investigators on the SAE reporting form.

One infant may have more than one SAE.

Growth of known pathogen on culture; data presented are number of infants who had at least one positive result.

Variable	Inc-AA/Intralipid (N = 42)	Inc-AA/SMOFlipid (N = 42)	Imm-RDI/Intralipid (N = 41)	Imm-RDI/SMOFlipid (N = 43)
Number of infants who experience a SAE, n (%)	6 (14.3)	12 (28.6)	8 (19.5)	9 (20.9)
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 inpatient hospitalisation, n (%)	2 (4.8)	2 (4.8)	1 (2.4)	3 (7.0)
Persistent or significant disability or incapacity, n (%)	0 (0.0)	1 (2.4)	0 (0.0)	1 (2.3)
Sepsis (diagnosis on SAE form), n (%)c	2 (4.8)	6 (14.3)	1 (2.4)	1 (2.3)
Necrotising enterocolitis, n (%)c	3 (7.1)	4 (9.5)	1 (2.4)	4 (9.3)

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).

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.

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.

TABLE 19 - Baseline characteristics and trial outcomes (all infants completing primary outcome assessments)

ALT, alanine transferase.

Adjusted for: age at MRI, sex, gestational age, birthweight and centre; body mass components are derived from body mass volumes.

Log-transformation was used in the regression model with the results transformed back from the log-scale.

Total of basal ganglia, thalami (deep grey matter), cerebrospinal fluid, grey matter, white matter and lateral ventricles volumes.

Total of basal ganglia, thalami (deep grey matter), grey matter and white matter.

Total of cerebellum and brainstem volumes.

Percentages of babies.

Logistic regression was used for modelling and odds ratios are reported.

	Inc-AA/Intralipid (N = 34)	Inc-AA/SMOFlipid (N = 28)	Imm-RDI/Intralipid (N = 34)	Imm-RDI/SMOFlipid (N = 37)	Adjusted mean difference (Imm-RDI minus Inc-AA)a	Adjusted mean difference (SMOFlipid minus Intralipid)a	Interaction (p-value)
Gestational age (weeks), mean (95% CI)	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)	–	–	–
Birthweight (g), mean (95% CI)	1064 (962 to 1166)	1103 (979 to 1226)	1090 (993 to 1186)	1059 (962 to 1155)	–	–	–
Sex, male (%), mean (95% CI)	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)	–	–	–
Age (weeks) at scan, mean (95% CI)	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)	–	–	–
Primary outcomes
Non-adipose mass (g), mean (95% CI)	2450 (2246 to 2655)	2337 (2164 to 2510)	2344 (2244 to 2444)	2485 (2327 to 2643)	1 (–108 to 111); p = 0.98	–41 (–150 to 68); p = 0.46	216 (0 to 432); p = 0.05
IHCL, mean (95% CI)b	0.6 (0.4 to 0.9); n = 34	0.7 (0.5 to 1.0); n = 28	0.5 (0.4 to 0.6); n = 34	0.5 (0.3 to 0.7); n = 36	0.7 (0.5 to 1.1); n = 132 p = 0.11	1.1 (0.8 to 1.6); n = 132 p = 0.58	0.8 (0.4 to 1.7); n = 132 p = 0.53
Secondary outcomes
Total cerebral volume (cm3), mean (95% CI)c	468 (419 to 518); n = 13	480 (425 to 534); n = 10	468 (414 to 523); n = 11	511 (440 to 583); n = 15	15 (–42 to 71); n = 49 p = 0.61	24 (–32 to 80); n = 49 p = 0.40	–26 (–142 to 90); n = 49 p = 0.66
Whole-brain volume (cm3), mean (95% CI)d	339 (304 to 373); n = 13	352 (319 to 385); n = 10	344 (296 to 393); n = 11	365 (321 to 410); n = 15	9 (–29 to 47); n = 49 p = 0.64	14 (–24 to 52); n = 49 p = 0.47	–29 (–107 to 49); n = 49 p = 0.46
Posterior fossa volume (cm3), mean (95% CI)e	30 (26 to 33); n = 13	31 (28 to 34); n = 10	30 (27 to 34); n = 11	35 (29 to 38); n = 15	1.44 (–1.99 to 4.87); n = 49 p = 0.41	2 (–2 to 5); n = 49 p = 0.35	–2 (–9 to 5); n = 49 p = 0.64
QUICKI, mean (95% CI)52	0.18 (0.17 to 0.19); n = 11	0.19 (0.18 to 0.20); n = 6	0.19 (0.18 to 0.20); n = 11	0.18 (0.17 to 0.20); n = 11	0.01 (0 to 0.02); n = 39p = 0.20	0.01 (–0.01 to 0.02); n = 39p = 0.28	–0.01 (–0.04 to 0.02); n = 39p = 0.46
Weight (g), mean (95% CI)	3060 (2780 to 3340)	2924 (2686 to 3162)	2932 (2780 to 3085)	3151 (2934 to 3368)	17 (–136 to 170); p = 0.83	–35 (–187 to 117); p = 0.65	293 (–8 to 593); p = 0.06
Length (cm), mean (95% CI)	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); p = 0.20	0.2 (–0.6 to 1.0); p = 0.56	0.5 (–1.1 to 2.1); p = 0.56
Head circumference (cm), mean (95% CI)	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); p = 0.02	–0.2 (–0.9 to 0.5); p = 0.56	1.1 (–0.2 to 2.5); p = 0.09
Superficial subcutaneous AT (g), mean (95% CI)	515 (437 to 593)	495 (431 to 559)	493 (431 to 554)	564 (499 to 629)	12 (–44 to 68); p = 0.67	9 (–46 to 64); p = 0.75	73 (–38 to 183); p = 0.20
Internal AT (g), mean (95% CI)	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); p = 0.62	–3.4 (–13.4 to 6.6); p = 0.50	0.1 (–19.9 to 20.1); p = 0.99
Deep subcutaneous abdominal AT (g), mean (95% CI)	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); p = 0.11	0.4 (–2.0 to 2.8); p = 0.74	3.5 (–1.3 to 8.3); p = 0.15
Internal abdominal AT (g), mean (95% CI)	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); p = 0.28	–0.8 (–3.4 to 1.8); p = 0.56	0.5 (–4.8 to 5.7); p = 0.86
Superficial subcutaneous AT tissue (g), mean (95% CI)	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); p = 0.38	5.0 (–6.7 to 16.6); p = 0.40	16.3 (–7.0 to 39.6); p = 0.17
Total AT (g), mean (95% CI)	610 (518 to 702)	587 (509 to 664)	589 (514 to 663)	666 (589 to 743)	16 (–51 to 82); p = 0.64	6 (–60 to 72); p = 0.85	77 (–55 to 208); p = 0.25
Total AT as a percentage of body weight (%), mean (95% CI)	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); p = 0.56	0.01 (–0.01 to 0.02); p = 0.45	0.01 (–0.02 to 0.03); p = 0.72
Safety outcomesf
Triglycerides > 2.5 mmol/l (%), mean (95% CI)g	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) p = 0.76	0.68 (0.28 to 1.62) p = 0.38	0.71 (0.12 to 4.06) p = 0.70
Total serum bilirubin > 150 µmol/l (%), mean (95% CI)g	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) p = 0.83	1.32 (0.59 to 2.94) p = 0.50	1.29 (0.26 to 6.47) p = 0.75
Conjugated bilirubin > 40 µmol/l (%), mean (95% CI)g	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) p = 0.28	0.93 (0.24 to 3.54) p = 0.92	1.65 (0.11 to 25.34) p = 0.72
ALT > 60 IU/l (%), mean (95% CI)g	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) p = 0.99	0.45 (0.09 to 2.12) p = 0.31	0.59 (0.03 to 12.69) p = 0.74

TABLE 20 - Adipose tissue compartments in litres for all infants completing MRI assessment

Adjusted for: age at scan, sex, gestational age, birthweight and centre.

	Inc-AA/20% Intralipid (n = 34)	Inc-AA/20% SMOFlipid (n = 28)	Imm-RDI/20% Intralipid (n = 34)	Imm-RDI/20% SMOFlipid (n = 37)	Adjusted mean difference (Imm-RDI – Inc-AA),a p-value	Adjusted mean difference (20% SMOFlipid – 20% Intralipid),a p-value	Interaction, p-value
Total internal AT (l) mean (95% CI)	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); p = 0.62	0 (–0.01 to 0.01); p = 0.50	0 (–0.02 to 0.02); p = 0.99
Superficial AT (l) mean (95% CI)	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); p = 0.67	0.01 (–0.05 to 0.07); p = 0.75	0.08 (–0.04 to 0.2); p = 0.20
Deep subcutaneous AT (l) mean (95% CI)	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); p = 0.24
Internal abdominal AT (l) mean (95% CI)	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); p = 0.86
Superficial subcutaneous abdominal AT (l) mean (95% CI)	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); p = 0.38	0.01 (–0.01 to 0.02); p = 0.40	0.02 (–0.01 to 0.04); p = 0.17
Deep subcutaneous abdominal AT (l) mean (95% CI)	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); p = 0.15
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); p = 0.64	0.01 (–0.07 to 0.08); p = 0.85	0.09 (–0.06 to 0.23); p = 0.25

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

MEBM, maternal expressed breast milk.

Covariates	Inc-AA/Intralipid (N = 34)	Inc-AA/SMOFlipid (N = 28)	RDI/Intralipid (N = 34)	RDI/SMOFlipid (N = 37)
Proportion of level 1 care, median (IQR)	0.11(0.07–0.21)	0.11 (0.06–0.35)	0.12 (0.08–0.31)	0.16 (0.07–0.32)
Proportion of level 2 care, median (IQR)	0.42 (0.21–0.59)	0.30 (0.20–0.46)	0.29 (0.20–0.44)	0.31 (0.21–0.42)
Proportion of MEBM in all milk intake, median (IQR)	0.80 (0.44–1.00); n = 33	0.95 (0.28–1.00); n = 27	0.65 (0.10–0.99); n = 33	0.70 (0.26–1.00); n = 37
Trial-PN phase protein intake (g), median (IQR)	31.5 (24.2–46.2)	34.4 (27.8–44.5)	42.8 (36.1–54.1)	48.5 (36.2–63.9)
Trial-PN phase carbohydrate intake (g), median (IQR)	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 (g), median (IQR)	52.8 (35.5–68.6)	51.6 (42.7–68.6)	51.8 (38.1–59.5)	53.1 (38.2–70.2)
Post-trial-PN phase protein intake (g), median (IQR)	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 carbohydrate intake (g), median (IQR)	969.0 (699.6–1786.1)	867.7 (643.5–1196.1)	1042.9 (644.8–1390.7)	1080.4 (696.3–1531.5)
Post-trial-PN phase fat intake (g), median (IQR)	479.8 (375.0–942.3)	434.8 (329.5–584.4)	515.9 (346.9–613.8)	550.5 (374.9–779.6)

TABLE 22 - Primary outcomes (pairwise comparisons)

Adjusted for age at MRI, sex, gestational age, birthweight and centre.

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.

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.

	Adjusted mean difference (Imm-RDI/Inc-AA)a	Adjusted mean difference (SMOFlipid/Intralipid)a	Interaction (p–value)	Adjusted mean difference (Imm-RDI/Inc-AA)b	Adjusted mean difference (SMOFlipid/Intralipid)b	Interaction (p-value)
Non-adipose mass (g), difference (95% CI)	1 (–108 to 111); p = 0.98	–41 (–150 to 68); p = 0.46	216 (0 to 432); p = 0.05	–44 (–226 to 139), n = 130; p = 0.64	–14 (–114 to 86), n = 130; p = 0.78	184 (–22 to 390); p = 0.08
IHCL content, difference (95% CI)c	0.7 (0.5 to 1.1), n = 132; p = 0.11	1.1 (0.8 to 1.6), n = 132; p = 0.58	0.8 (0.4 to 1.7), n = 132; p = 0.53	0.81 (0.37 to 1.80), n = 129; p = 0.61	0.89 (0.61 to 1.31) n = 129; p = 0.57	0.86 (0.39 to 1.92), n = 129; p = 0.71

Weight gain over the study period was similar across groups (Figure 8).

FIGURE 8.

Time trend for babies’ weights across four groups.

Trial PN protein intake was higher in the Imm-RDI arms in the first 2 weeks (Figure 9).

FIGURE 9.

Daily protein intake from all sources in the first 2 weeks across four groups.

Carbohydrate, lipid and energy intakes were similar across all four groups (Figures 10–12).

FIGURE 10.

Daily carbohydrate intake from all sources in the first 2 weeks across all four groups.

FIGURE 11.

Daily fat intake from all sources in the first 2 weeks across all four groups.

FIGURE 12.

Daily energy intake from all sources in the first 2 postnatal weeks.

Macronutrient and energy intake from milk and PN from the end of the second week onwards did not differ between groups (Figures 13–16).

FIGURE 13.

Daily protein intake from all sources after first 2 weeks across all four groups.

FIGURE 14.

Daily carbohydrate intake from all sources after first 2 weeks across all four groups.

FIGURE 15.

Daily fat intake from all sources after first 2 weeks across four groups.

FIGURE 16.

Daily energy intake from all sources after the first 2 postnatal weeks.

Chapter 5 Discussion

The 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.

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 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 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.

Chapter 6 Conclusions and recommendations

In 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.

Acknowledgements

The following people are thanked for participating in the trial:

• parents and babies

• 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

• Imperial Clinical Trials Unit

• Juan Gonzalez-Maffè and Alexina Mason

• Clinical Trials and Evaluation Unit at Royal Brompton and Harefield NHS Foundation Trust

• Sealed Envelope Ltd for providing the randomisation list and IVRS.

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;52:17-24. 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. 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. 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.
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 . 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;41:1-87. 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. 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 &Amp; Children n.d. 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 n.d. 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.
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. 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. 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, . 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.
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. 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 n.d.: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 n.d. 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&lt;407::AID-SIM742&gt;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;36:81S-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.
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. http://dx.doi.org/10.1542/peds.2013-2207.
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. http://dx.doi.org/10.1542/peds.2011-3379.
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.

Appendix 1 Distribution of primary and secondary outcomes after transformation

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.

Appendix 2 Parent information sheet, magnetic resonance information sheet and consent form

(PDF download)

List of abbreviations

AE

adverse event

AT

adipose tissue

CI

confidence interval

DMEC

Data Monitoring and Ethics Committee

eCRF

electronic case record form

EME

Efficacy and Mechanism Evaluation

FOV

field of view

ICTU

Imperial Clinical Trials Unit

IHCL

intrahepatocellular lipid

Imm-RDI

immediate introduction of the recommended daily intake of amino acids

Imm-RDI/Intralipid

immediate introduction of the recommended daily intake of amino acids and 20% Intralipid

Imm-RDI/SMOFlipid

immediate introduction of the recommended daily intake of amino acids and 20% SMOFlipid

IMP

investigational medicinal product

Inc-AA

incremental amino acids

Inc-AA/Intralipid

incremental amino acids and 20% Intralipid

Inc-AA/SMOFlipid

incremental amino acids and 20% SMOFlipid

IQR

interquartile range

IVRS

interactive voice recognition system

LBM

lean body mass

LCT

long-chain triglyceride

MCT

medium-chain triglyceride

MHRA

Medicines and Healthcare products Regulatory Agency

MR

magnetic resonance

MRI

magnetic resonance imaging

MRS

magnetic resonance spectroscopy

NEON

Nutritional Evaluation and Optimisation in Neonates

NICU

neonatal intensive care unit

NIHR

National Institute for Health Research

PIS

parent information sheet

PN

parenteral nutrition

QUICKI

quantitative insulin sensitivity check index

RCT

randomised controlled trial

RDI

recommended daily intake

SAE

serious adverse event

SD

standard deviation

SMOFlipid

a mixture of soya bean oil, MCTs, olive oil and fish oils, supplemented with vitamin E

SpAE

specific adverse event

SSCNAAT

superficial subcutaneous non-abdominal adipose tissue

TSC

Trial Steering Committee