Optimal timeframe to start parenteral nutrition

National Guideline Alliance (UK)

Optimal timeframe to start parenteral nutrition

Neonatal parenteral nutrition

Evidence review A2

NICE Guideline, No. 154

Authors

National Guideline Alliance (UK).

London: National Institute for Health and Care Excellence (NICE); 2020 Feb.

ISBN-13: 978-1-4731-3673-1

Optimal timeframe to starting parenteral nutrition

Review question

For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

Introduction

Where provision of parenteral nutrition (PN) support has been agreed, the optimal timeframe for starting such support is important. Delaying provision of PN may lead to increased nutritional deficits, especially for the preterm infant, where body stores are low. However, provision of early PN exposes infants to the recognised risks of PN administration, such as electrolyte imbalance, metabolic disturbance or fluid overload.

Summary of the protocol

Please see Table 1 for a summary of the Population, Intervention, Comparison and Outcome (PICO) characteristics of this review.

Table 1

Summary of the protocol (PICO table).

For full details see the review protocol in appendix A.

Clinical evidence

Included studies

Four studies were identified for inclusion in this review (Brownlee 1993, Dongming 2016, Ibrahim 2004, and van Puffelen 2018).

Three randomised controlled trials (RCTs) compared early to late PN in preterm babies. One study (n=129) defined early PN as before 36 hours, and late as 6 days (Brownlee 1993). One study (n=80) defined early PN as before 24 hours and late as 3 days (Dongming 2016). One study (n=32) defined early PN as before 2 hours and late as 48 hours PN (Ibrahim 2004). Due to the similarity in definitions for early PN in the Brownlee study (1993) and late PN in the Ibrahim study (2004), 36 hours and 48 hours respectively, it was not appropriate to pool outcome data.

One RCT (n=209) compared early (within 24 hours of admission) to late (1 week) PN in term babies (van Puffelen 2018).

The included studies are summarised in Table 2.

Table 2

Summary of included studies.

See the literature search strategy in appendix B, study selection flow chart in appendix C, study evidence tables in appendix D, forest plots in appendix E, and GRADE tables in appendix F.

Excluded studies

Studies not included in this review are listed, and reasons for their exclusions are provided, in appendix K.

Summary of clinical studies included in the evidence review

Summaries of the studies that were included in this review are presented Table 2.

See appendix D for the full evidence tables.

Quality assessment of clinical outcomes included in the evidence review

GRADE was conducted to assess the quality of outcomes. Evidence was identified for critical and important outcomes. The clinical evidence profiles can be found in appendix F.

Economic evidence

Included studies

A systematic review of the economic literature was conducted but no economic studies were identified which were applicable to this review question. A single economic search was undertaken for all topics included in the scope of this guideline. Please see supplementary material D for details.

Excluded studies

No studies were identified which were applicable to this review question.

Summary of studies included in the economic evidence review

No economic evaluations were identified which were applicable to this review question.

Economic model

This question was medium priority for economic evaluation. However, the identified clinical data was insufficient to inform de-novo economic modelling in this area.

Evidence statements

Clinical evidence statements

Early versus late parenteral nutrition in preterm babies

Weight gain

Very low quality evidence from 1 RCT (n=129) showed no clinically important difference in weight gain at 2 weeks in babies who had received early PN (36 hours) as compared to late PN (6 days). However, there was uncertainty around the effect: Mean difference (MD) 13g (95% CI −3.92 to 29.92).
Very low quality evidence from 1 RCT (n=129) showed no clinically important difference in weight gain per day until discharge between babies who received early PN (36 hours) as compared to late PN (6 days). However, there was uncertainty around the effect: MD 2.4g (95% CI −5.3 to 0.5).

Hyperglycaemia

Very low quality evidence from 1 RCT (n=80) showed a clinically important difference in the number of babies with hyperglycaemia between those who received early PN (24 hours) as compared to late PN (3 days), with fewer occurrences of hyperglycaemia in babies given early PN. However, there was high uncertainty around the effect: Relative risk (RR) 0.33 (95% CI 0.04 to 3.07)

Sepsis

Very low quality evidence from 1 RCT (n=29) showed no clinically important difference in the number of babies with sepsis between those who received early PN (within 2 hours) as compared to late PN (within 48 hours). However, there was high uncertainty around the effect: RR 0.92 (95% CI 0.41 to 2.07).

Mortality

Very low quality evidence from 1 RCT (n=129) showed no clinically important difference in mortality between those who received early PN (36 hours) as compared to late PN (6 days). However, there was high uncertainty around the effect: RR 0.82 (95% CI 0.4 to 1.67).
Very low quality evidence from 1 RCT (n=29) showed a clinically important difference in mortality of babies between those who received early PN (within 2 hours) as compared to late PN (within 48 hours), with fewer occurrences of mortality in those given early PN. However, there was high uncertainty around the effect: RR 0.5 (95% CI 0.05 to 4.98).

Caloric intake

Moderate quality evidence from 1 RCT (n=32) showed a clinically important difference in calorie intake between babies who received early PN (within 2 hours) as compared to those who received late PN (within 48 hours) with a greater intake associated with a later start of PN: MD 18.4kcal (95% CI 17.22 to 19.58)

Early versus late parenteral nutrition in critically ill, term babies

Paediatric intensive care unit (PICU) acquired infections

Low quality evidence from 1 RCT (n=209) showed a clinically important difference in the number of babies with PICU acquired infections between those who received early PN (within 24 hours) as compared to late PN (after 1 week), with more occurrences of infection in those given early PN. However, there was uncertainty around the effect: RR 1.89 (95% CI 1.13 to 3.17).
Low quality evidence from the same RCT (n=38) showed a clinically important difference in the number of babies with PICU acquired infections between those who received early PN (within 24 hours) as compared to late PN (after 1 week), with more occurrences of infection in those given early PN, in a subsample of babies who received no or minimal enteral nutrition. However, there was uncertainty around the effect: RR 1.49 (95% CI 0.81 to 2.73).

Hypoglycaemia during the first week

Low quality evidence from 1 RCT (n=209) showed a clinically important difference in the number of babies with hypoglycaemia between those who received early PN (within 24 hours) as compared to late PN (after 1 week), with fewer occurrences of hypoglycaemia in those given early PN. However, there was uncertainty around the effect: RR 0.61 (95% CI 0.34 to 1.10).
Low quality evidence from the same RCT (n=38) showed a clinically important difference in the number of babies with hypoglycaemia between those who received early PN (within 24 hours) as compared to late PN (after 1 week), with fewer occurrences of hypoglycaemia in those given early PN, in a subsample of babies who received no or minimal enteral nutrition. However, there was uncertainty around the effect: RR 0.33 (95% CI 0.12 to 0.89).

Mortality at 90 days

Low quality evidence from 1 RCT (n=209) showed a clinically important difference in mortality between those who received early PN (within 24 hours) as compare to late PN (after 1 week), with more occurrences of mortality in those given early PN. However, there was uncertainty around the effect: RR 2.83 (95% CI 1.14 to 7.01).
Very low quality evidence from 1 RCT (n=38) showed a clinically important difference in mortality between those who received early PN (within 24 hours) as compare to late PN (after 1 week), with more occurrences of mortality in those given early PN, in a subsample of babies who received no or minimal enteral nutrition. However, there was high uncertainty around the effect: RR 2.28 (95% CI 0.55 to 9.54).

Economic evidence statements

No studies were identified which were applicable to this review question.

The committee’s discussion of the evidence

Interpreting the evidence

The outcomes that matter most

The committee agreed that the critical outcomes were neurodevelopmental outcomes, growth and body composition. These were agreed as a delay in PN provision is likely to have direct effects on these anthropometric measures (i.e. growth and body composition) and also on the development of the brain. Infection and adverse events were also considered critical as these may occur as part of the practicalities of administration of PN. The committee agreed mortality, duration of hospital stay and nutrient intake should be considered as important outcomes; although influenced by PN other factors may impact on mortality and duration of hospital stay. The committee did not think the nutrient intake was as critical as the effect of intake on the individual baby (i.e. as long as the baby is growing, the detailed nutrient intake in itself is less important due to multiple influences on metabolism).

The quality of the evidence

The quality of evidence for this review was assessed using GRADE methodology. The evidence presented was mainly considered very low quality, indicating high uncertainty in the reliability of the data, with the exception of caloric intake which was considered moderate quality. Evidence was downgraded due to serious or very serious risk of bias associated with unclear methods of allocation, unclear concealment of allocation, and unclear blinding of assessors. In addition, the studies had small sample sizes leading to imprecision.

Another quality issue that the committee considered was the applicability or generalisability of the studies. The committee acknowledged that the evidence presented reflected how clinical practice has changed over time, as the earlier studies started PN later than what would now be considered good clinical practice.

Benefits and harms

The evidence on preterm babies presented in this review was heterogeneous regarding what was defined as early (ranged from 2 hours to 36 hours without very clear descriptions of the enteral feeding regimen). There was also no consistent pattern of findings. While none of the outcomes favoured late administration only a couple of them clearly indicated better outcomes associated with an earlier start of PN (fewer babies with hyperglycaemia and lower mortality). Other outcomes suggested no clear difference (weight gain and sepsis) and there was a higher caloric intake associated with later start (which is difficult to interpret as favouring either early or late). The evidence for all outcomes, apart from caloric intake, was rated as very low quality, and there were also limitations in its applicability to current clinical practice (most studies were old and were not considered to provide PN formulations that would now be considered optimal). Due to these issues, the committee had low confidence in the evidence and therefore made the recommendations using informal consensus, based on their clinical experience and expertise.

There was one study comparing early (within 24 hours) and late (after 1 week) provision of PN in term babies. There was evidence of greater rates of infection and mortality when PN was provided early compared with late, but lower rates of hypoglycaemia. These differences were observed both in the whole group of babies, and a subgroup who received no or minimal enteral nutrition. However, evidence was all low or very low quality.

The committee noted that some of the included studies specified a timeframe of two hours. They discussed that these were randomised controlled trials, which would have clear protocols and service arrangements in place that would make it possible for the participating centres to adhere to this schedule. The committee agreed that PN should be given as soon as possible once the decision to start PN (see evidence review A1 for the predictors of enteral feeding success) has been made and the earlier the better which was supported by some of the evidence. However, they decided that two hours would be unrealistic and likely to have a large resource impact, and that out-of-hours services could make such a timeframe unachievable. They therefore balanced the benefits of early administration (better nutrition) with the logistic challenges of adhering to a specific timeframe and decided that within eight hours would be both safe and achievable. Based on their experience the committee agreed that eight hours would not detrimentally affect growth outcomes. They also agreed that in certain circumstances, for example when a preterm baby requires PN or a term baby is critically unwell, or a baby has surgical problems, there are several competing treatment priorities; it may be the case that resuscitative measures must take precedence over the provision of PN, and therefore, a recommendation to give PN in under two hours could adversely affect essential treatment prioritisation. The committee provided examples of achievable timeframes from different levels of neonatal units, ranging from three hours to 17 hours; therefore, they agreed that a “within timeframe” recommendation was the most appropriate recommendation, and that based on their clinical experience and expertise, eight hours is both achievable and safe.

The committee discussed the possibility of giving a longer timeframe to start PN, but they were concerned that in practice this would be exceeded. It was agreed that stating a shorter timeframe would highlight the importance of starting PN as soon as possible. However, they agreed that this can vary in different situations, for example a moderately term baby may be tried on enteral feeds for a while before the decision is made for them to get PN. This would therefore be a delay in starting PN. They therefore intentionally phrased this to include the wording of ‘when a baby meets the indications for parenteral nutrition” to indicate that the decision is not always clear cut (as in 8 hours after the baby is born).

The committee discussed whether different timeframes should be recommended for preterm and term babies. They discussed the evidence on critically ill term babies showing an increased risk of infection and mortality, but decreased hypoglycaemia, associated with the earlier start They noted that this evidence was from a pre-planned subgroup analysis (which was small), the parenteral nutrition regimens used were not consistent across the different study sites and the intervention may not have been appropriate because parenteral nutrition would not normally be started on day 1 for critically ill term babies due to restricted fluid volumes and strain on organ systems. They therefore agreed that there was too much uncertainty to suggest a longer time to start based on these findings.

They agreed that without PN preterm babies will develop a nutritional deficit more rapidly than term babies; therefore, the committee decided that it may take longer to come to a decision whether PN in term babies is needed. Reaching this conclusion may take longer because term babies have greater nutritional reserves. However, once the decision is made that PN is needed, it should be started within eight hours, regardless of whether babies are term or preterm.

Having identified the limitations of the evidence, the committee agreed that there is a need for further research in this area in clearly defined populations (i.e. critically ill term babies and surgical babies). Identifying the timeframe for starting PN administration is crucial to ensure optimum care is provided and to improve outcomes for babies. They therefore made a research recommendation, by informal consensus, to address this topic in clearly defined populations.

Cost effectiveness and resource use

There was no existing economic evidence on the cost-effectiveness of an earlier versus later start of PN. The committee acknowledged the inconclusive clinical evidence in this area. However, according to the committee’s expertise alongside the limited clinical evidence presented it was concluded that early initiation of PN would have the benefit of adequate nutritional intake to support early development when compared to the late initiation of PN. In regards to costs, earlier initiation of PN is likely to reduce nutritional deficits arising which reduces the costs associated with dealing with those nutritional deficits. The total length of time PN is given for is closely tied to enteral feed tolerance and not just dependant on the time PN is started. Starting PN later would not necessarily result in a shorter duration of PN as the delayed nutritional delivery has the potential to reduce growth. Additionally, the recommendation to start PN earlier is likely to be most easily achieved through the use of standardised PN bags as opposed to bespoke PN bags. In general, standardised PN bags are cheaper than bespoke ones so this may lead to cost savings. Additionally, the committee felt the better outcomes for a baby initiated on earlier PN meant that this was the preferred strategy and therefore the benefits would outweigh the costs.

Other factors the committee took into account

The committee agreed that the use of standardised bags (which is recommended by the committee and discussed in evidence review E) would facilitate neonatal units to achieve this timeframe target because standardised bags would be readily available to administer when the decision has been made to start PN.

References

Brownlee 1993
Brownlee, K. G., Kelly, E. J., Ng, P. C., Kendall-Smith, S. C., Dear, P. R., Early or late parenteral nutrition for the sick preterm infant?, Archives of disease in childhood, 69, 281–3, 1993 [PMC free article: PMC1029492] [PubMed: 8215565]
Dongming 2016
Lang, Dongming, Zhou, Fengran, Zhang, Zhaojun, The study of early intravenous nutrition therapy in very low birth weight infants, Pakistan journal of pharmaceutical sciences, 29, 2293–2295, 2016 [PubMed: 28167468]
Ibrahim 2004
Ibrahim, Hassan M., Jeroudi, Majied A., Baier, R. J., Dhanireddy, Ramasubbareddy, Krouskop, Richard W., Aggressive early total parental nutrition in low-birth-weight infants, Journal of perinatology : official journal of the California Perinatal Association, 24, 482–6, 2004 [PubMed: 15167885]
Van Puffelen 218
van Puffelen, E., Vanhorebeek, I., Joosten, K. F. M., Wouters, P. J., Van den Berghe, G., Verbruggen, Scat, Early versus late parenteral nutrition in critically ill, term neonates: a preplanned secondary subgroup analysis of the PEPaNIC multicentre, randomised controlled trial, The lancet child and adolescent health, 2, 505–515, 2018 [PubMed: 30169323]

Appendices

Appendix A. Review protocols

Review protocol for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

Table 3. Review protocol – optimal timeframe to starting PN

Appendix B. Literature search strategies

Literature search strategy for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

Databases: Medline; Medline EPub Ahead of Print; and Medline In-Process & Other Non-Indexed Citations

Databases: Embase; and Embase Classic

Databases: Cochrane Central Register of Controlled Trials; Cochrane Database of Systematic Reviews; Database of Abstracts of Reviews of Effects; and Health Technology Assessment

Appendix C. Clinical evidence study selection

Clinical study selection for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

Figure 1. PRISMA Flow chart of clinical article selection for review question, for those neonates where PN is required, what is the optimal timeframe?

Appendix D. Clinical evidence tables

Clinical evidence tables for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this? In babies on parenteral nutrition, what is the optimal frequency of blood sampling and monitoring?

Table 4. Clinical evidence tables for included studies (PDF, 429K)

Appendix E. Forest plots

Forest plots for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

No meta-analysis was conducted for this review; therefore there are no forest plots

Appendix F. GRADE tables

GRADE tables for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

Table 5. Clinical evidence profile for early versus late delivery of parenteral nutrition in preterm babies

Table 6. Clinical evidence profile for early versus late parenteral nutrition in critically ill, term babies

Appendix G. Economic evidence study selection

Economic evidence study selection for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

One global search was conducted for all review questions. See supplementary material D for further information.

Appendix H. Economic evidence tables

Economic evidence tables for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

No economic studies were identified which were applicable to this review question.

Appendix I. Economic evidence profiles

Economic evidence profiles for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

No economic studies were identified which were applicable to this review question.

Appendix J. Economic analysis

Economic analysis for review question: For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?

No economic analysis was undertaken for this review question.

Appendix K. Excluded studies

Excluded studies for review question: For those neonates where PN is required what is the optimal timeframe for doing this?

Clinical studies

Table 7. Excluded studies and reasons for their exclusion

Economic studies

No economic evidence was identified for this review question. See supplementary material D for further information.

Appendix L. Research recommendations

Research recommendations for review question: For those neonates where PN is required what is the optimal timeframe for doing this?

Research recommendation

What is the optimal timeframe for starting parenteral nutrition in term babies who are critically ill or who require surgery?

Why this is important

Where provision of parenteral nutrition (PN) has been agreed, the optimal timeframe for starting such support is important. Delaying the provision of PN may lead to increased nutritional deficits, and this is especially important in preterm babies who lack nutritional stores and for babies who are critically ill or babies who require surgery. Some evidence was identified for preterm babies; however, there was little evidence for term babies who are critically ill or babies requiring surgery.

Table 8. Research recommendation rationale

Table 9. Research recommendation modified PICO table

Final

Evidence reviews

These evidence reviews were developed by the National Guideline Alliance which is part of the Royal College of Obstetricians and Gynaecologists

Disclaimer: The recommendations in this guideline represent the view of NICE, arrived at after careful consideration of the evidence available. When exercising their judgement, professionals are expected to take this guideline fully into account, alongside the individual needs, preferences and values of their patients or service users. The recommendations in this guideline are not mandatory and the guideline does not override the responsibility of healthcare professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patient and/or their carer or guardian.

Local commissioners and/or providers have a responsibility to enable the guideline to be applied when individual health professionals and their patients or service users wish to use it. They should do so in the context of local and national priorities for funding and developing services, and in light of their duties to have due regard to the need to eliminate unlawful discrimination, to advance equality of opportunity and to reduce health inequalities. Nothing in this guideline should be interpreted in a way that would be inconsistent with compliance with those duties.

NICE guidelines cover health and care in England. Decisions on how they apply in other UK countries are made by ministers in the Welsh Government, Scottish Government, and Northern Ireland Executive. All NICE guidance is subject to regular review and may be updated or withdrawn.

Bookshelf ID: NBK555668PMID: 32282160

Table 1Summary of the protocol (PICO table)

Population	Babies born preterm, up to 28 days after their due birth date (preterm babies) Babies born at term, up to 28 days after their birth (term babies)
Intervention	Early start of PN^*
Comparison	Late start of PN^*
Outcomes	Critical Neurodevelopmental outcomes (general cognitive abilities at two years corrected age as measured by a validated scale) Growth: Weight gain (g/kg/d) Linear growth Head circumference (mm) Infection (including sepsis) Body composition (measured as lean mass, fat-free mass, fat mass, adipose tissue) Adverse effects of PN Hyperglycaemia Hypoglycaemia Hypertriglyceridemia Other PN associated liver disease Important Mortality Duration of hospital stay Nutritional intake (prescribed PN actually received)

*: The decision was made not to specifically define the timeframe of ‘early’ or ‘late’ in the protocol because it was recognised that this could be interpreted as the ‘time from birth to starting PN’ or as ‘the time from birth to the decision to start PN’, which could take into account an initial trial of enteral feeding. This could lead to very different timings and would mean what definitions of ‘early’ in one study may be ‘late’ in another. The timing would therefore be extracted directly as reported in the studies and the details of this taken into consideration in the discussion.

: PN: Parenteral nutrition

Table 2Summary of included studies

Study

Population

Intervention

Comparison

Outcomes

Brownlee 1993

RCT

N=129

Median GA (range):

Early: 29 weeks (23-33)

Late: 29 weeks (24 – 36)

Median BW (range)

Early: 1144g (539 – 1748)

Late: 1147g (415 – 1647)

Early PN (n= 63)

Received PN within the first 36 hours.

PN followed standard regimen - 20% intralipid, 0.5 g/kg/day glucose (increased daily to 3.5 g/kg/day)

Fluid regimen – started on 75ml/kg/day (increased to 165-180 ml.kg/day) of 10% dextrose solution

Late PN (n=66)

Received PN on the sixth complete day

PN followed standard regimen - 20% intralipid, 0.5 g/kg/day glucose (increased daily to 3.5 g/kg/day)

Fluid regimen – started on 75ml/kg/day (increased to 165-180 ml.kg/day) of 10% dextrose solution

Weight gain
Mortality
Nutrient intake

Dongming 2016

RCT

China

N=80

Mean age

Early: 30.3 weeks (SD 2.4)

Late: 30.4 weeks (SD 2.2)

Mean BW

Early: 1140g (SD 220)

Late: 1148g (SD 216)

Early PN (n=40)

Received PN within the first 24 hours after birth with 20% fat emulsion and 6% amino acid, initial dose 1.5 g/kg/day, daily increments of 0.5 g/kg/day up to 3 g/kg/day

Late PN (n=40)

Received PN within 3 days with 5-10% glucose

After day 3 received the same intravenous nutrition as the early group.

Proportion of body weight loss (%)
Hyperglycaemia

Ibrahim 2004

RCT

USA

N=32

Mean GA

Early: 27 weeks (SD 1.6)

Late: 26.8 weeks (SD 1.5)

Mean BW

Early: 846g (SD 261)

Late: 968 (SD 244)

Early PN (n=16)

Received PN within the first 2 hours after birth

3.5g/kg/day AA, 3g/kg/day of 20% lipid

Late PN (n=16)

Received a solution containing 5 to 10% glucose during the first 48 hours of life

After 48 hours PN included 2g/kg/day AA and 0.5g/kg/day lipid. These increased to a maximum of 3.5g/kg/day and 3g/kg/day respectively

Calorie intake
Mortality
Sepsis

van Puffelen 2018

RCT

Belgium, Canada & Netherlands

N=209

Mean GA

Early: 38.4 weeks (SD 1.4)

Late: 38.6 weeks (SD 1.6)

Mean BW

Early: 3193g (SD 538)

Late: 3238g (SD 510)

Early PN (n=98)

Received PN within 24 hours of admission

Late PN (n=111)

Received PN after 1 week

PICU acquired infections
Hypoglycaemia
Mortality

: AA: Amino acid; BW: Birth weight; GA: Gestational age; VLBW: Very low birth weight; PICU: Paediatric intensive care unit; PN: Parenteral nutrition; RCT: Randomised controlled trail; SD Standard deviation.

Table 3Review protocol – optimal timeframe to starting PN

Field (based on PRISMA-P)	Content
Review question	For those neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?
Type of review question	Intervention
Objective of the review	Where provision of parenteral nutrition (PN) support has been agreed, the optimal timeframe for starting such support is important.
Eligibility criteria – population/disease/condition/issue/domain	Where a decision has been made that PN is needed: Babies born preterm, up to 28 days after their due birth date (preterm babies) Babies born at term, up to 28 days after their birth (term babies).
Eligibility criteria – intervention(s)/exposure(s)/prognostic factor(s)	Early start of PN The decision was made not to specifically define the timeframe of ‘early’ or ‘late’ in the protocol because it was recognised that this could be interpreted as the ‘time from birth to starting PN’ or as ‘the time from birth to the decision to start PN’, which could take into account an initial trial of enteral feeding. This could lead to very different timings and would mean what definitions of ‘early’ in one study may be ‘late’ in another. The timing would therefore be extracted directly as reported in the studies and the details of this taken into consideration in the discussion.
Eligibility criteria – comparator(s)/control or reference (gold) standard	Late start of PN
Outcomes and prioritisation	Critical Neurodevelopmental outcomes (general cognitive abilities at two years corrected age as measured by a validated scale) Growth: Weight gain (g/kg/d) Linear growth Head circumference (mm) Infection (including sepsis) Body composition (measured as lean mass, fat-free mass, fat mass, adipose tissue) Adverse effects of PN Hyperglycaemia Hypoglycaemia Hypertriglyceridemia Other PN associated liver disease Important Mortality Duration of hospital stay Nutritional intake (prescribed PN actually received)
Eligibility criteria – study design	Systematic reviews of RCTs RCTs Comparative cohort studies (only if RCTs unavailable or limited data to inform decision making). The decision to include comparative cohort studies will be determined for each parameter according to available RCT data Conference abstracts of RCTs will only be considered if no evidence is available from full published RCTs (if no evidence from RCTs or comparative cohort studies available and are recent i.e., in the last 2 years-authors will be contacted for further information).
Other inclusion exclusion criteria	No sample size restriction No date restriction
Proposed sensitivity/sub-group analysis, or meta-regression	Stratified analysis Babies born preterm, up to 28 days after their due birth date (preterm babies) Babies born at term, up to 28 days after their birth (term babies) Babies who are critically ill or need surgery Where evidence exists, consideration will be given to the specific needs of population subgroups: Age of baby (first 2 weeks vs. later) Preterm (Extremely preterm <28 weeks GA; very preterm: 28-31 weeks GA; moderately preterm: 32-36 weeks GA) Birth weight: Low birth weight (< 2500g); very low birth weight (< 1500g) and extremely low birth weight (< 1000g) Possible equality considerations Mothers aged 17 or below Parents or carers whose first language is not English Parents or carers who have learning difficulties Important confounders (when comparative observational studies are included for interventional reviews): Age of baby (first 2 weeks vs. later) Preterm (Very early <28 weeks GA; 28-31 weeks GA; 32-36 weeks GA) Birth weight: Low birth weight (< 2500g); very low birth weight (< 1500g) and extremely low birth weight (< 1000g) Sex of baby Gestation Neurodevelopmental outcomes: Biological (sex, small for gestational age, ethnicity) Neonatal (PVL, IVH, infarct, sepsis, ROP, NEC, antenatal/postnatal steroids, BPD at 36 weeks) Social (SES, substance abuse, alcohol abuse, multiple pregnancy, chorioamnionitis, neglect, maternal age, maternal mental health disorder) Postnatal (epilepsy, age of establishing feeding) Secure venous access
Selection process – duplicate screening/selection/analysis	Sifting, data extraction, appraisal of methodological quality and GRADE assessment will be performed by the systematic reviewer. Quality control will be performed by the senior systematic reviewer. A random sample of the references will be sifted by a second reviewer. This sample size will be 10% of the total, or 100 studies if the search identifies fewer than 1000 studies. All disagreements will be resolved by discussion between the two reviewers. The senior systematic reviewer or guideline lead will act as arbiter where necessary.
Data management (software)	Pairwise meta-analyses, if possible, will be performed using Cochrane Review Manager (RevMan5). ‘GRADEpro’ will be used to assess the quality of evidence for each outcome. Low income countries will be downgraded for indirectness. NGA STAR software will be used for generating bibliographies/citations, study sifting, data extraction and recording quality assessment using checklists (ROBIS (systematic reviews and meta-analyses); Cochrane risk of bias tool (RCTs or comparative cohort studies); Cochrane risk of bias tool (Non-randomised studies).
Information sources – databases and dates	Sources to be searched: Medline, Medline In-Process, CCTR, CDSR, DARE, HTA, Embase. Limits (e.g. date, study design): All study designs. Apply standard animal/non-English language filters. No date limit. Supplementary search techniques: No supplementary search techniques were used. See appendix B for full strategies.
Identify if an update	This is not an update
Author contacts	Developer: The National Guideline Alliance https://www.nice.org.uk/guidance/indevelopment/gid-ng10037
Highlight if amendment to previous protocol	For details please see section 4.5 of Developing NICE guidelines: the manual 2014.
Search strategy – for one database	For details please see appendix B.
Data collection process – forms/duplicate	A standardised evidence table format will be used, and published as appendix D (clinical evidence tables) or H (economic evidence tables).
Data items – define all variables to be collected	For details please see appendix B.
Methods for assessing bias at outcome/study level	Standard study checklists were used to critically appraise individual studies. For details please see section 6.2 of Developing NICE guidelines: the manual 2014. The risk of bias across all available evidence was evaluated for each outcome using an adaptation of the ‘Grading of Recommendations Assessment, Development and Evaluation (GRADE) toolbox’ developed by the international GRADE working group http://www.gradeworkinggroup.org/
Criteria for quantitative synthesis (where suitable)	For details please see section 6.4 of Developing NICE guidelines: the manual 2014.
Methods for analysis – combining studies and exploring (in)consistency	For details of the methods please see supplementary material C.
Meta-bias assessment – publication bias, selective reporting bias	For details please see section 6.2 of Developing NICE guidelines: the manual 2014. If sufficient relevant RCT evidence is available, publication bias will be explored using RevMan software to examine funnel plots. Trial registries will be examined to identify missing evidence: Clinical trials.gov, NIHR Clinical Trials Gateway.
Assessment of confidence in cumulative evidence	For details please see sections 6.4 and 9.1 of Developing NICE guidelines: the manual 2014.
Rationale/context – Current management	For details please see the introduction to the evidence review.
Describe contributions of authors and guarantor	A multidisciplinary committee developed the guideline. The committee was convened by The National Guideline Alliance and chaired by Joe Fawke (Consultant Neonatologist and Honorary Senior Lecturer, University Hospitals Leicester NHS Trust), in line with section 3 of Developing NICE guidelines: the manual 2014. Staff from The National Guideline Alliance, undertook systematic literature searches, appraised the evidence, conducted meta-analysis and cost-effectiveness analysis where appropriate, and drafted the guideline in collaboration with the committee. For details of the methods please see supplementary material C.
Sources of funding/support	The National Guideline Alliance is funded by NICE and hosted by The Royal College of Obstetricians and Gynaecologists
Name of sponsor	The National Guideline Alliance is funded by NICE and hosted by The Royal College of Obstetricians and Gynaecologists
Roles of sponsor	NICE funds the National Guideline Alliance to develop guidelines for those working in the NHS, public health, and social care in England
PROSPERO registration number	This review is not registered with PROSPERO

: BPD: bronchopulmonary dysplasia; CCTR: Cochrane controlled trials register; CDSR: Cochrane database of systematic reviews; DARE: database of abstracts of reviews of effects; GA: Gestational age; GRADE: Grading of Recommendations Assessment, Development and Evaluation; HTA: health technology assessment; IVH: intraventricular haemorrhage; NEC: necrotising enterocolitis; NGA: National Guidelines Alliance; NHS: National Health Service; NICE: National Institute of Clinical Excellence; NIHR: National Institute for Health Research; NGA: National Guideline Alliance; NHS: National Health Service; PN: Parenteral nutrition; PROSPERO: International prospective register of systematic reviews; PVL: periventricular leukomalacia; RCT: randomised controlled trial; ROBIS; risk of bias in systematic reviews; ROP: retinopathy of prematurity; SES: socioeconomic status.

Databases: Medline; Medline EPub Ahead of Print; and Medline In-Process & Other Non-Indexed Citations

#	Searches
1	INFANT, NEWBORN/
2	(neonat$ or newborn$ or new-born$ or baby or babies).ti,ab.
3	PREMATURE BIRTH/
4	((preterm$ or pre-term$ or prematur$ or pre-matur$) adj5 (birth? or born)).ab,ti.
5	exp INFANT, PREMATURE/
6	((preterm$ or pre-term$ or prematur$ or pre-matur$) adj5 infan$).ti,ab.
7	(pre#mie? or premie or premies).ti,ab.
8	exp INFANT, LOW BIRTH WEIGHT/
9	(low adj3 birth adj3 weigh$ adj5 infan$).ti,ab.
10	((LBW or VLBW) adj5 infan$).ti,ab.
11	INTENSIVE CARE, NEONATAL/
12	INTENSIVE CARE UNITS, NEONATAL/
13	NICU?.ti,ab.
14	or/1-13
15	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) adj5 parenteral$).ti,ab.
16	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) and parenteral$).ti.
17	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) adj5 (PN or SPN or IPN or TPN or STD-PN or IND-PN)).ti,ab.
18	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) adj5 (intravenous$ or intra-venous$ or IV or venous$ or infusion?) adj5 (nutrition$ or feed$ or fed$)).ti,ab.
19	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) and (intravenous$ or intra-venous$ or IV or venous$ or infusion?) and (nutrition$ or feed$ or fed$)).ti.
20	((24 h$ or 24h$) adj5 parenteral$).ti,ab.
21	or/15-20
22	*PARENTERAL NUTRITION/
23	*PARENTERAL NUTRITION, TOTAL/
24	*PARENTERAL NUTRITION SOLUTIONS/
25	or/22-24
26	TIME FACTORS/
27	(Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$).ti.
28	14 and 21
29	14 and 25 and 26
30	14 and 25 and 27
31	or/28-30
32	limit 31 to english language
33	LETTER/
34	EDITORIAL/
35	NEWS/
36	exp HISTORICAL ARTICLE/
37	ANECDOTES AS TOPIC/
38	COMMENT/
39	CASE REPORT/
40	(letter or comment*).ti.
41	or/33-40
42	RANDOMIZED CONTROLLED TRIAL/ or random*.ti,ab.
43	41 not 42
44	ANIMALS/ not HUMANS/
45	exp ANIMALS, LABORATORY/
46	exp ANIMAL EXPERIMENTATION/
47	exp MODELS, ANIMAL/
48	exp RODENTIA/
49	(rat or rats or mouse or mice).ti.
50	or/43-49
51	32 not 50

Databases: Embase; and Embase Classic

#	Searches
1	NEWBORN/
2	(neonat$ or newborn$ or new-born$ or baby or babies).ti,ab.
3	PREMATURITY/
4	((preterm$ or pre-term$ or prematur$ or pre-matur$) adj5 (birth? or born)).ab,ti.
5	((preterm$ or pre-term$ or prematur$ or pre-matur$) adj5 infan$).ti,ab.
6	(pre#mie? or premie or premies).ti,ab.
7	exp LOW BIRTH WEIGHT/
8	(low adj3 birth adj3 weigh$ adj5 infan$).ti,ab.
9	((LBW or VLBW) adj5 infan$).ti,ab.
10	NEWBORN INTENSIVE CARE/
11	NEONATAL INTENSIVE CARE UNIT/
12	NICU?.ti,ab.
13	or/1-12
14	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) adj5 parenteral$).ti,ab.
15	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) and parenteral$).ti.
16	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) adj5 (PN or SPN or IPN or TPN or STD-PN or IND-PN)).ti,ab.
17	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) adj5 (intravenous$ or intra-venous$ or IV or venous$ or infusion?) adj5 (nutrition$ or feed$ or fed$)).ti,ab.
18	((Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$) and (intravenous$ or intra-venous$ or IV or venous$ or infusion?) and (nutrition$ or feed$ or fed$)).ti.
19	((24 h$ or 24h$) adj5 parenteral$).ti,ab.
20	or/14-19
21	*PARENTERAL NUTRITION/
22	*TOTAL PARENTERAL NUTRITION/
23	PERIPHERAL PARENTERAL NUTRITION/
24	PARENTERAL SOLUTIONS/
25	INTRAVENOUS FEEDING/
26	or/21-25
27	TIME FACTOR/
28	(Initiat$ or Start$ or Introduc$ or begin$ or commenc$ or inaugurat$ or launch$ or instigat$ or establish$ or set$ up or timing or early or earlier or prompt$ or quick$ or speed$ or rapid$ or fast or delay$ or late? or wait$ or slow$ or postpon$ or put$ off or defer$ or push$ back or belated$).ti.
29	13 and 20
30	13 and 26 and 27
31	13 and 26 and 28
32	or/29-31
33	limit 32 to english language
34	letter.pt. or LETTER/
35	note.pt.
36	editorial.pt.
37	CASE REPORT/ or CASE STUDY/
38	(letter or comment*).ti.
39	or/34-38
40	RANDOMIZED CONTROLLED TRIAL/ or random*.ti,ab.
41	39 not 40
42	ANIMAL/ not HUMAN/
43	NONHUMAN/
44	exp ANIMAL EXPERIMENT/
45	exp EXPERIMENTAL ANIMAL/
46	ANIMAL MODEL/
47	exp RODENT/
48	(rat or rats or mouse or mice).ti.
49	or/41-48
50	33 not 49

Databases: Cochrane Central Register of Controlled Trials; Cochrane Database of Systematic Reviews; Database of Abstracts of Reviews of Effects; and Health Technology Assessment

#	Searches
1	MeSH descriptor: [INFANT, NEWBORN] this term only
2	(neonat* or newborn* or new-born* or baby or babies):ti,ab
3	MeSH descriptor: [PREMATURE BIRTH] this term only
4	((preterm* or pre-term* or prematur* or pre-matur*) near/5 (birth? or born)).ab,ti.
5	MeSH descriptor: [INFANT, PREMATURE/
6	((preterm* or pre-term* or prematur* or pre-matur) near/5 infan):ti,ab
7	(pre#mie? or premie or premies):ti,ab
8	MeSH descriptor: [INFANT, LOW BIRTH WEIGHT/
9	(low near/3 birth near/3 weigh* near/5 infan*):ti,ab
10	((LBW or VLBW) near/5 infan*):ti,ab
11	MeSH descriptor: [INTENSIVE CARE, NEONATAL] this term only
12	MeSH descriptor: [INTENSIVE CARE UNITS, NEONATAL] this term only
13	NICU?:ti,ab
14	#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13
15	((Initiat* or Start* or Introduc* or begin* or commenc* or inaugurat* or launch* or instigat* or establish* or set* up or timing or early or earlier or prompt* or quick* or speed* or rapid* or fast or delay* or late? or wait* or slow* or postpon* or put* off or defer* or push* back or belated) near/5 parenteral):ti,ab
16	((Initiat* or Start* or Introduc* or begin* or commenc* or inaugurat* or launch* or instigat* or establish* or set* up or timing or early or earlier or prompt* or quick* or speed* or rapid* or fast or delay* or late? or wait* or slow* or postpon* or put* off or defer* or push* back or belated) near/10 parenteral).ti.
17	((Initiat* or Start* or Introduc* or begin* or commenc* or inaugurat* or launch* or instigat* or establish* or set* up or timing or early or earlier or prompt* or quick* or speed* or rapid* or fast or delay* or late? or wait* or slow* or postpon* or put* off or defer* or push* back or belated*) near/5 (PN or SPN or IPN or TPN or STD-PN or IND-PN)):ti,ab
18	((Initiat* or Start* or Introduc* or begin* or commenc* or inaugurat* or launch* or instigat* or establish* or set* up or timing or early or earlier or prompt* or quick* or speed* or rapid* or fast or delay* or late? or wait* or slow* or postpon* or put* off or defer* or push* back or belated) near/5 (intravenous or intra-venous* or IV or venous* or infusion?) near/5 (nutrition* or feed* or fed*)):ti,ab
19	((Initiat* or Start* or Introduc* or begin* or commenc* or inaugurat* or launch* or instigat* or establish* or set* up or timing or early or earlier or prompt* or quick* or speed* or rapid* or fast or delay* or late? or wait* or slow* or postpon* or put* off or defer* or push* back or belated) near/10 (intravenous or intra-venous* or IV or venous* or infusion?) near/10 (nutrition* or feed* or fed*)).ti.
20	((24 h* or 24h) near/5 parenteral):ti,ab
21	#15 or #16 or #17 or #18 or #19 or #20
22	MeSH descriptor: [PARENTERAL NUTRITION] this term only
23	MeSH descriptor: [PARENTERAL NUTRITION, TOTAL] this term only
24	MeSH descriptor: [PARENTERAL NUTRITION SOLUTIONS] this term only
25	#22 or #23 or #24
26	MeSH descriptor: [TIME FACTORS] this term only
27	(Initiat* or Start* or Introduc* or begin* or commenc* or inaugurat* or launch* or instigat* or establish* or set* up or timing or early or earlier or prompt* or quick* or speed* or rapid* or fast or delay* or late? or wait* or slow* or postpon* or put* off or defer* or push* back or belated*).ti.
28	#14 and #21
29	#14 and #25 and #26
30	#14 and #25 and #27
31	#28 or #29 or #30

Figure 1PRISMA Flow chart of clinical article selection for review question, for those neonates where PN is required, what is the optimal timeframe?

Table 5Clinical evidence profile for early versus late delivery of parenteral nutrition in preterm babies

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	Early PN	Late PN	Relative (95% CI)	Absolute	Quality	Importance
Weight gain at 2 weeks (g) (follow-up mean 2 weeks; Better indicated by higher values)
1	randomised trials	very serious¹	no serious inconsistency	no serious indirectness	serious²	none	63	66	-	MD 13 higher (3.92 lower to 29.92 higher)	⨁◯◯◯ VERY LOW	CRITICAL
Weight gain per day to discharge (Better indicated by higher values)
1	randomised trials	very serious¹	no serious inconsistency	no serious indirectness	serious³	none	63	66	-	MD 2.4 lower (5.3 lower to 0.5 higher)	⨁◯◯◯ VERY LOW	CRITICAL
Hyperglycaemia
1	randomised trials	very serious¹	no serious inconsistency	no serious indirectness	very serious⁴	none	1/40 (2.5%)	3/40 (7.5%)	RR 0.33 (0.04 to 3.07)	50 fewer per 1000 (from 72 fewer to 155 more)	⨁◯◯◯ VERY LOW	CRITICAL
Sepsis
1	randomised trials	serious⁵	no serious inconsistency	no serious indirectness	very serious⁴	none	6/14 (42.9%)	7/15 (46.7%)	RR 0.92 (0.41 to 2.07)	37 fewer per 1000 (from 275 fewer to 499 more)	⨁◯◯◯ VERY LOW	CRITICAL
Mortality - 36 hours versus 6 days (follow-up median 8 days)
1	randomised trials	very serious¹	no serious inconsistency	no serious indirectness	very serious⁴	none	11/63 (17.5%)	14/66 (21.2%)	RR 0.82 (0.4 to 1.67)	38 fewer per 1000 (from 127 fewer to 142 more)	⨁◯◯◯ VERY LOW	IMPORTANT
Mortality - 2 hours versus 48 hours
1	randomised trials	serious⁵	no serious inconsistency	no serious indirectness	very serious⁴	none	1/16 (6.3%)	2/16 (12.5%)	RR 0.5 (0.05 to 4.98)	62 fewer per 1000 (from 119 fewer to 498 more)	⨁◯◯◯ VERY LOW	IMPORTANT
Caloric intake in first five days of life (kcal/kg/day) (follow-up mean 5 days; Better indicated by higher values)
1	randomised trials	serious⁵	no serious inconsistency	no serious indirectness	no serious imprecision	none	16	16	-	MD 18.4 higher (17.22 to 19.58 higher)	⨁⨁⨁◯ MODERATE	IMPORTANT

: CI: confidence interval; PN: parenteral nutrition; RR: risk ratio.

1: Very serious risk of bias, evidence downgraded due to unclear risk of allocation and selection bias. No details provided regarding methods of allocation or randomisation. Unclear risk of performance and detection bias, unclear if care personnel or assessors were blinded to treatment.

2: Evidence was downgraded by 1 due to serious imprecision, 95% confidence interval crosses one default MID for continuous outcomes, calculated as 0.5 x SD control at baseline (24)

3: Evidence was downgraded by 1 due to serious imprecision, 95% confidence interval crosses one default MID for continuous outcomes, calculated as 0.5 x SD control at baseline (4.6)

4: Evidence was downgraded by 2 due to very serious imprecision, 95% confidence interval crosses two default MID for dichotomous outcomes, (0.8 and 1.25).

5: Serious risk of bias, evidence downgraded due to unclear risk of selection bias, no details provided regarding methods of randomisation. Unclear risk of performance and detection bias, unclear if care personnel or assessors were blinded to treatment.

Table 6Clinical evidence profile for early versus late parenteral nutrition in critically ill, term babies

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	Early PN	Late PN	Relative (95% CI)	Absolute	Quality	Importance
PICU acquired infection - whole sample
1	randomised trials	serious¹	no serious inconsistency	no serious indirectness	serious²	none	30/98 (30.6%)	18/111 (16.2%)	RR 1.89 (1.13 to 3.17)	144 more per 1000 (from 21 more to 352 more)	⨁⨁◯◯ LOW	CRITICAL
PICU acquired infection - no/minimal EN
1	randomised trials	serious¹	no serious inconsistency	no serious indirectness	serious²	none	16/23 (69.6%)	7/15 (46.7%)	RR 1.49 (0.81 to 2.73)	229 more per 1000 (from 89 fewer to 807 more)	⨁⨁◯◯ LOW	CRITICAL
Hypoglycaemia in first week - whole sample
1	randomised trials	serious¹	no serious inconsistency	no serious indirectness	serious²	none	14/98 (14.3%)	26/111 (23.4%)	RR 0.61 (0.34 to 1.1)	91 fewer per 1000 (from 155 fewer to 23 more)	⨁⨁◯◯ LOW	CRITICAL
Hypoglycaemia in first week - no/minimal EN
1	randomised trials	serious¹	no serious inconsistency	no serious indirectness	serious²	none	4/23 (17.4%)	8/15 (53.3%)	RR 0.33 (0.12 to 0.89)	357 fewer per 1000 (from 59 fewer to 469 fewer)	⨁⨁◯◯ LOW	CRITICAL
Mortality at 90 days - whole sample
1	randomised trials	serious¹	no serious inconsistency	no serious indirectness	serious²	none	15/98 (15.3%)	6/111 (5.4%)	RR 2.83 (1.14 to 7.01)	99 more per 1000 (from 8 more to 325 more)	⨁⨁◯◯ LOW	IMPORTANT
Mortality at 90 days - no/minimal EN
1	randomised trials	serious¹	no serious inconsistency	no serious indirectness	very serious³	none	7/23 (30.4%)	2/15 (13.3%)	RR 2.28 (0.55 to 9.54)	171 more per 1000 (from 60 fewer to 1000 more)	⨁◯◯◯ VERY LOW	IMPORTANT

: CI: confidence interval; EN: enteral nutrition; PICU: paediatric intensive care unit; PN: parenteral nutrition; RR: risk ratio.

1: Serious risk of bias, evidence downgraded due to unclear risk of allocation concealment and performance bias.

2: Evidence was downgraded by 1 due to serious imprecision, 95% confidence interval crosses one default MID for dichotomous outcomes, (0.8 or 1.25).

3: Evidence was downgraded by 2 due to very serious imprecision, 95% confidence interval crosses two default MID for dichotomous outcomes, (0.8 and 1.25).

Table 7Excluded studies and reasons for their exclusion

Study	Reason for Exclusion
Abdou, R. M., Weheiba, H. M. I., The effect of early versus late lipid infusion in parenteral nutrition on the biochemical and cortical auditory evoked potential parameters in preterm neonates, Egyptian Pediatric Association Gazette, 2018	Study intervention does not meet protocol eligibility criteria - early vs late lipids; infants receive glucose and dextrose from birth and amino acids from day 2.
Aroor, Amitha R., Krishnan, Lalitha, Reyes, Zenaida, Fazallulah, Muhammed, Ahmed, Masood, Khan, Ashfaq A., Al-Farsi, Yahya, Early versus Late Parenteral Nutrition in Very Low Birthweight Neonates: A retrospective study from Oman, Sultan Qaboos University medical journal, 12, 33–40, 2012 [PMC free article: PMC3286714] [PubMed: 22375256]	Study design does not meet the inclusion criteria, this is a retrospective cohort study. Additionally, only AA are started early compared to late, all other constituents are given at the same time point.
Bishay, M., Lakshminarayanan, B., Arnaud, A., Garriboli, M., Cross, K. M., Curry, J. I., Drake, D., Kiely, E. M., De Coppi, P., Pierro, A., Eaton, S., The role of parenteral nutrition following surgery for duodenal atresia or stenosis, Pediatric Surgery International, 29, 191–5, 2013 [PubMed: 23187894]	Population does not meet the inclusion criteria.
Blanco, Cynthia Liudmilla, Falck, Alison, Green, Belinda Kay, Cornell, John E., Gong, Alice Kim, Metabolic responses to early and high protein supplementation in a randomized trial evaluating the prevention of hyperkalemia in extremely low birth weight infants, The Journal of pediatrics, 153, 535–40, 2008 [PubMed: 18589451]	Intervention does not meet the inclusion criteria; study focus is on amino acid dose and starting time.
Blanco, Cynthia Liudmilla, Gong, Alice Kim, Green, Belinda Kay, Falck, Alison, Schoolfield, John, Liechty, Edward A., Early changes in plasma amino acid concentrations during aggressive nutritional therapy in extremely low birth weight infants, The Journal of pediatrics, 158, 543–548.e1, 2011 [PubMed: 21129755]	Intervention does not meet the inclusion criteria; study focus is on amino acid dose and starting time.
Bulbul, Ali, Okan, Fusun, Bulbul, Lida, Nuhoglu, Asiye, Effect of low versus high early parenteral nutrition on plasma amino acid profiles in very low birth-weight infants, The journal of maternal-fetal & neonatal medicine : the official journal of the European Association of Perinatal Medicine, the Federation of Asia and Oceania Perinatal Societies, the International Society of Perinatal Obstetricians, 25, 770–6, 2012 [PubMed: 21770835]	Intervention does not meet the inclusion criteria; both groups received early PN.
Can, E., Bulbul, A., Uslu, S., Comert, S., Bolat, F., Nuhoglu, A., Evaluation of two different types of parenteral nutrition on early growth of preterm infants, Early Human Development, 86, S85, 2010	Conference abstract.
Can, Emrah, Bulbul, Ali, Uslu, Sinan, Bolat, Fatih, Comert, Serdar, Nuhoglu, Asiye, Early Aggressive Parenteral Nutrition Induced High Insulin-like growth factor 1 (IGF-1) and insulin-like growth factor binding protein 3 (IGFBP3) Levels Can Prevent Risk of Retinopathy of Prematurity, Iranian journal of pediatrics, 23, 403–10, 2013 [PMC free article: PMC3883369] [PubMed: 24427493]	Intervention does not meet the inclusion criteria; both groups started PN at the same time.
Chan, S. H. T., Johnson, M. J., Vollmer, B., Early nutrition and neurodevelopmental outcomes in very preterm infants: A systematic review and meta-analysis, Developmental Medicine and Child Neurology, 56, 42–43, 2014	Conference abstract.
Dinerstein, A., Nieto, R. M., Solana, C. L., Perez, G. P., Otheguy, L. E., Larguia, A. M., Early and aggressive nutritional strategy (parenteral and enteral) decreases postnatal growth failure in very low birth weight infants, Journal of Perinatology, 26, 436–42, 2006 [PubMed: 16801958]	Intervention does not meet the inclusion criteria; the nutrition strategy includes EN.
Donovan, Ramona, Puppala, Bhagya, Angst, Denise, Coyle, Bryan W., Outcomes of early nutrition support in extremely low-birth-weight infants, Nutrition in clinical practice : official publication of the American Society for Parenteral and Enteral Nutrition, 21, 395–400, 2006 [PubMed: 16870808]	Study design does not meet inclusion criteria; retrospective chart review.
Elstgeest, Liset E., Martens, Shirley E., Lopriore, Enrico, Walther, Frans J., te Pas, Arjan B., Does parenteral nutrition influence electrolyte and fluid balance in preterm infants in the first days after birth?, PLoS ONE, 5, e9033, 2010 [PMC free article: PMC2815790] [PubMed: 20140260]	Outcomes do not meet the inclusion criteria: Serum sodium, potassium levels and fluid balance.
Fivez, T, Kerklaan, D, Verbruggen, S, Vanhorebeek, I, Verstraete, S, Tibboel, D, Guerra, Gg, Wouters, Pj, Joffe, A, Joosten, K, Mesotten, D, Berghe, G, Impact of withholding early parenteral nutrition completing enteral nutrition in pediatric critically ill patients (PEPaNIC trial): study protocol for a randomized controlled trial, Trials, 16, 202, 2015 [PMC free article: PMC4422419] [PubMed: 25927936]	Protocol paper.
Geary, C. A., Fonseca, R. A., Caskey, M. A., Malloy, M. H., Improved growth and decreased morbidities in <1000 g neonates after early management changes, Journal of perinatology : official journal of the California Perinatal Association, 28, 347–53, 2008 [PubMed: 18337743]	Intervention does not meet the inclusion criteria; both groups received early PN at the same time.
Genoni, G., Binotti, M., Monzani, A., Bernascone, E., Stasi, I., Bona, G., Ferrero, F., Nonrandomised interventional study showed that early aggressive nutrition was effective in reducing postnatal growth restriction in preterm infants, Acta Paediatrica, International Journal of Paediatrics, 106, 1589–1595, 2017 [PubMed: 28632972]	Intervention does not meet the inclusion criteria; both intervention and comparative historical cohorts started PN on day 1 of life.
Haghedooren, R., Jenniskens, M., Guiza, F., Verbruggen, S., Guerra, G., Joosten, K., Langouche, L., Van Den Berghe, G., Prevalence and prognostic value of abnormal liver test results in critically ill children and the impact of nutrition hereon, Critical Care, 22, 2018 [PMC free article: PMC6282934] [PubMed: 30234740]	Study design does not meet protocol eligibility criteria - conference abstract.
Heimler, Ruth, Bamberger, Janine M., Sasidharan, Ponthenkandath, The effects of early parenteral amino acids on sick premature infants, Indian journal of pediatrics, 77, 1395–9, 2010 [PubMed: 20830534]	Intervention does not meet the inclusion criteria; study focus is on Amino Acids.
Ho, Man-Yau, Yen, Y. u-Hsuan, Hsieh, Mao-Chih, Chen, Hsiang-Yin, Chien, Shu-Chen, Hus-Lee, Shing-Mei, Early versus late nutrition support in premature neonates with respiratory distress syndrome, Nutrition (Burbank, Los Angeles County, Calif.), 19, 257–60, 2003 [PubMed: 12620530]	Intervention does not meet the inclusion criteria; study focus is on Amino Acids.
Jimenez, Lissette, Mehta, Nilesh M., Duggan, Christopher P., Timing of the initiation of parenteral nutrition in critically ill children, Current Opinion in Clinical Nutrition and Metabolic Care, 20, 227–231, 2017 [PMC free article: PMC5844227] [PubMed: 28376054]	Study design does not meet protocol eligibility criteria - not a systematic review.
Joffe, Ari, Anton, Natalie, Lequier, Laurance, Vandermeer, Ben, Tjosvold, Lisa, Larsen, Bodil, Hartling, Lisa, Nutritional support for critically ill children, Cochrane Database of Systematic Reviews, 2016 [PMC free article: PMC6517095] [PubMed: 27230550]	Participants do not meet the inclusion criteria; children aged 1 to 18 years.
Kennedy, K. A., Tyson, J. E., Chamnanvanikij, S., Early versus delayed initiation of progressive enteral feedings for parenterally fed low birth weight or preterm infants, The Cochrane database of systematic reviews, CD001970, 2000 [PubMed: 10796276]	Intervention does not fit the inclusion criteria; timing of EN feeding.
Kotsopoulos, K., Benadiba-Torch, A., Cuddy, A., Shah, P. S., Safety and efficacy of early amino acids in preterm <28 weeks’ gestation: Prospective observational comparison, Journal of Perinatology, 26, 749–754, 2006 [PubMed: 17024139]	Intervention does not meet the inclusion criteria; study focus is on Amino Acids.
Leenders, Erika K. S. M., de Waard, Marita, van Goudoever, Johannes B., Low- versus High-Dose and Early versus Late Parenteral Amino-Acid Administration in Very-Low-Birth-Weight Infants: A Systematic Review and Meta-Analysis, Neonatology, 113, 187–205, 2018 [PubMed: 29268262]	Systematic review - references cross-checked.
Loys, C. M., Maucort-Boulch, D., Guy, B., Putet, G., Picaud, J. C., Hays, S., Extremely low birthweight infants: how neonatal intensive care unit teams can reduce postnatal malnutrition and prevent growth retardation, Acta Paediatrica, 102, 242–8, 2013 [PubMed: 23167480]	Intervention does not meet the inclusion criteria; comparison includes earlier introduction of lipids and protein and EN.
Mamunes,P., Baden,M., Bass,J.W., Nelson,J., Early intravenous feeding of the low birth weight neonate, Pediatrics, 43, 241–250, 1969 [PubMed: 4974283]	Interventions does not meet the inclusion criteria; early intravenous (IV group) versus delayed oral feedings (fasted group).
Morgan, Jessie, Young, Lauren, McGuire, William, Delayed introduction of progressive enteral feeds to prevent necrotising enterocolitis in very low birth weight infants, Cochrane Database of Systematic Reviews, 2014 [PMC free article: PMC7063979] [PubMed: 25436902]	Intervention does not meet the inclusion criteria; timing of EN feeding.
Moyses, H. E., Johnson, M. J., Cornelius, V., Leaf, A. A., Is there any benefit to starting total parenteral nutrition early in very low birth weight infants? A systematic review, Proceedings of the Nutrition Society, 70, E259, 2011	Conference abstract.
Moyses, Helen E., Johnson, Mark J., Leaf, Alison A., Cornelius, Victoria R., Early parenteral nutrition and growth outcomes in preterm infants: a systematic review and meta-analysis, The American journal of clinical nutrition, 97, 816–26, 2013 [PubMed: 23446896]	Systematic review, references checked for inclusion. Article is in Chinese.
Radmacher, P. G., Lewis, S. L., Adamkin, D. H., Early amino acids and the metabolic response of ELBW infants (< or = 1000 g) in three time periods, Journal of perinatology : official journal of the California Perinatal Association, 29, 433–7, 2009 [PubMed: 19339983]	Intervention does not meet the inclusion criteria; the study focus is on Amino Acids.
Rao, S., Rao, U., Jape, G., Early versus late parnteral nutrition in critically ill full term neonates: A systematic review, Journal of Paediatrics and Child Health, 53, 49, 2017	Conference abstract.
Ribed Sanchez, A., Romero Jimenez, R. M., Sanchez De Orgaz, M. C., De Juan, A., Tovar Pozo, M., Diaz Garzon, J., Sanjurjo Saez, M., Early aggressive parenteral nutrition in preterm infants, International Journal of Clinical Pharmacy, 35, 983, 2013	Conference abstract.
Stensvold, Hans Jorgen, Strommen, Kenneth, Lang, Astri M., Abrahamsen, Tore G., Steen, Eline Kjorsvik, Pripp, Are H., Ronnestad, Arild E., Early Enhanced Parenteral Nutrition, Hyperglycemia, and Death Among Extremely Low-Birth-Weight Infants, JAMA pediatrics, 169, 1003–10, 2015 [PubMed: 26348113]	Study design does not meet the inclusion criteria; observational cohort study.
te Braake, F. W. J., van den Akker, C. H. P., Riedijk, M. A., van Goudoever, J. B., Parenteral amino acid and energy administration to premature infants in early life, Seminars in Fetal and Neonatal Medicine, 12, 11–18, 2007 [PubMed: 17142119]	Non-systematic review.
Tewari, Vishal Vishnu, Dubey, Sachin Kumar, Kumar, Reema, Vardhan, Shakti, Sreedhar, C. M., Gupta, Girish, Early versus Late Enteral Feeding in Preterm Intrauterine Growth Restricted Neonates with Antenatal Doppler Abnormalities: An Open-Label Randomized Trial, Journal of tropical pediatrics, 2017 [PubMed: 28369652]	Intervention does not meet the inclusion criteria; babies receive breast milk.
Tim-Aroon, Thipwimol, Harmon, Heidi M., Nock, Mary L., Viswanathan, Sreekanth K., McCandless, Shawn E., Stopping Parenteral Nutrition for 3 Hours Reduces False Positives in Newborn Screening, The Journal of pediatrics, 167, 312–6, 2015 [PubMed: 26003996]	Intervention does not meet the inclusion criteria; study focus is on Amino Acids.
Trintis,J., Donohue,P., Aucott,S., Outcomes of early parenteral nutrition for premature infants, Journal of Perinatology, 30, 403–407, 2010 [PubMed: 19865093]	Intervention does not meet the inclusion criteria; study focus is on Amino Acids.
Trivedi,A., Sinn,J.K., Early versus late administration of amino acids in preterm infants receiving parenteral nutrition, Cochrane Database of Systematic Reviews, -, 2013 [PubMed: 23881744]	Intervention does not meet the inclusion criteria; study focus is on Amino Acids.
Valentine, C. J., Fernandez, S., Rogers, L. K., Gulati, P., Hayes, J., Lore, P., Puthoff, T., Dumm, M., Jones, A., Collins, K., Curtiss, J., Hutson, K., Clark, K., Welty, S. E., Early amino-acid administration improves preterm infant weight, Journal of Perinatology, 29, 428–432, 2009 [PMC free article: PMC2834366] [PubMed: 19444236]	Intervention does not meet the inclusion criteria; study focus is on Amino Acids.
van Puffelen, Esther, Hulst, Jessie M., Vanhorebeek, Ilse, Dulfer, Karolijn, Van den Berghe, Greet, Verbruggen, Sascha C. A. T., Joosten, Koen F. M., Outcomes of Delaying Parenteral Nutrition for 1 Week vs Initiation Within 24 Hours Among Undernourished Children in Pediatric Intensive Care: A Subanalysis of the PEPaNIC Randomized Clinical Trial, JAMA network open, 1, e182668, 2018 [PMC free article: PMC6324499] [PubMed: 30646158]	Results not presented separately for neonates.
Vanhorebeek, Ilse, Verbruggen, Sascha, Casaer, Michael P., Gunst, Jan, Wouters, Pieter J., Hanot, Jan, Guerra, Gonzalo Garcia, Vlasselaers, Dirk, Joosten, Koen, Van den Berghe, Greet, Effect of early supplemental parenteral nutrition in the paediatric ICU: a preplanned observational study of post-randomisation treatments in the PEPaNIC trial, The Lancet. Respiratory medicine, 5, 475–483, 2017 [PubMed: 28522351]	Study design does not meet inclusion criteria; observational study (follow up study of PEPaNIC trial - PN in critically ill children).
Verstraete, Soren, Verbruggen, Sascha C., Hordijk, Jose A., Vanhorebeek, Ilse, Dulfer, Karolijn, Guiza, Fabian, van Puffelen, Esther, Jacobs, An, Leys, Sandra, Durt, Astrid, Van Cleemput, Hanna, Eveleens, Renate D., Garcia Guerra, Gonzalo, Wouters, Pieter J., Joosten, Koen F., Van den Berghe, Greet, Long-term developmental effects of withholding parenteral nutrition for 1 week in the paediatric intensive care unit: a 2-year follow-up of the PEPaNIC international, randomised, controlled trial, The Lancet. Respiratory medicine, 7, 141–153, 2019 [PubMed: 30224325]	Study intervention does not meet protocol eligibility criteria - PN administered as supplement to EN.
Weiler, Hope A., Fitzpatrick-Wong, Shirley C., Schellenberg, Jeannine M., Fair, Denise E., McCloy, Ursula R., Veitch, Rebecca R., Kovacs, Heather R., Seshia, Mary M., Minimal enteral feeding within 3 d of birth in prematurely born infants with birth weight < or = 1200 g improves bone mass by term age, The American journal of clinical nutrition, 83, 155–62, 2006 [PubMed: 16403735]	Intervention does not meet inclusion criteria; study focus is on Amino Acids.
Whitfield, M. F., Spitz, L., Milner, R. D., Clinical and metabolic consequences of two regimens of total parenteral nutrition in the newborn, Archives of Disease in Childhood, 58, 168–75, 1983 [PMC free article: PMC1627832] [PubMed: 6404225]	Interventions do not meet inclusion criteria; study compares sequential versus continuous regimens.

Table 8Research recommendation rationale

Research question	For neonates where parenteral nutrition is required, what is the optimal timeframe for doing this?
Why is this needed
Importance to ‘patients’ or the population	High: The timing of starting PN administration is crucial in order to avoid nutritional deficits and exposure to risks associated with PN administration.
Relevance to NICE guidance	High: Only four, randomised studies were identified for inclusion in this review. The evidence that was identified was limited in quality and did not reflect current good clinical practice. Findings from the included studies were inconsistent and the studies did not provide data to determine the optimal timing of starting PN administration in clearly defined populations (particularly in term babies who are critically ill or require surgery).
Relevance to the NHS	High: Identifying the timeframe for starting PN in clearly defined populations (i.e. critically ill term babies and surgical babies) is critical for the provision of optimum care and to ensure optimal growth, development and survival.
National priorities	The NHS Long term plan (launched in January 2019) for the next 10 years highlights ‘enabling everyone to get the best start in life’ as one of the main areas to improve the quality of patient care and health outcomes.
Current evidence base	The guideline identified that there is a gap in the evidence base. The four studies had small sample sizes (450 babies included across four studies), and were considered to be very low quality, with a high risk of bias. Furthermore, the studies were heterogeneous with regard to definitions for early PN and were not considered to provide PN formulations that would now be considered optimal.
Equality	Some parents and carers may need information in different languages and there may be cultural sensitivities around PN. Those with learning disabilities may need additional support.
Feasibility	This would require NHS ethical approval but would be feasible and safe to conduct.
Other comments

: NHS: National Health Service; NICE: National Institute for Health and Care Excellence; PN: Parenteral nutrition

Table 9Research recommendation modified PICO table

Criterion	Explanation
Population	Critically ill babies born > 37 weeks of gestational age, up to 28 days after their due date (term babies) including those requiring surgery.
Intervention	Early start of PN (defined as within 48 hours)
Comparator	Late start of PN (later than 2 days)
Outcomes	Survival to discharge Rates of nosocomial infections Incidence of hypoglycaemia Neurodevelopment Growth Infection Body composition Adverse effects of PN
Study design	Randomised controlled trial
Timeframe	From birth to discharge
Additional information	The decision was made not to specifically defined the timeframe of ‘early’ or ‘late’ in the protocol because it was recognised that this could be interpreted as the ‘time from birth to starting PN’ or as ‘the time from birth to the decision to start PN’, which could take into account an initial trial of enteral feeding. This could lead to very different timings and would mean what definitions of ‘early’ in one study may be ‘late’ in another. The timing would therefore be extracted directly as reported in the studies and the details of this taken into consideration in the discussion.

: PN: Parenteral nutrition