Topic 1: Parenteral Nutrition

Justification

There is no evidence of benefit for key clinical outcomes such as mortality, sepsis, necrotising enterocolitis, length of hospital stay or breastfeeding at discharge when parenteral nutrition is used instead of 10% glucose in this population. Parenteral nutrition is invasive, increases resource use and costs and may negatively influence the transition to exclusive maternal breastmilk, particularly in pēpi Māori. The evidence does not support the routine use of parenteral nutrition in this population without clear clinical indication. In contrast, 10% glucose is inexpensive, widely available and safer for short-term use while awaiting maternal milk. Standardising the use of 10% glucose and delaying formula or parenteral nutrition unless clinically necessary may reduce variation and support equitable outcomes. A consistent approach using 10% glucose while actively supporting lactation may protect maternal milk feeding and improve equity for whānau Māori. Supporting early breastmilk expression and timely transition to enteral feeds should remain the priority.

Subgroup considerations

Families from under-represented backgrounds may struggle with the complexity of early feeding decisions. Clear, culturally responsive communication about the temporary use of IV fluids, the benefits of maternal breastmilk and plans to transition to enteral feeding can help ensure informed parental involvement and support equitable care.

Implementation considerations

Ensure that all infants have prompt access to IV glucose if needed. Consider initiating parenteral nutrition if breastfeeding is contraindicated or has not commenced by 96 hours after birth. Prioritise culturally safe, whānau-led lactation support to reduce early use of formula, particularly for pēpi Māori.

Monitoring and evaluation

Track use of 10% glucose and parenteral nutrition by ethnicity, gestation and birthweight. Evaluate equity impacts by ethnicity in feeding transitions, exposure to alternative nutrition and discharge feeding outcomes.

Research priorities

• Culturally responsive strategies to optimise early lactation and reduce time to mother's milk availability.
• Long-term outcomes associated with early 10% glucose vs. parenteral nutrition in moderate-to-late preterm infants. 
• Impact of early nutritional decisions on exclusive breastmilk feeding at discharge across ethnic groups.

Implementation Considerations

Requires streamlined processes to ensure vascular access is secured and parenteral nutrition can be started within 6–8 hours of birth.

Monitoring and Evaluation

Track time from birth to initiation of parenteral nutrition.

Research Priorities

• Early vs delayed parenteral nutrition initiation stratified by gestation and birthweight.

Subgroup Considerations

Particularly relevant to infants born <28 weeks or <1,000 g who may have a prolonged transition to full enteral feeding.

Implementation

Ensure availability of early vascular access protocols, standard parenteral nutrition solutions, education for staff on timing targets and coordination between clinical teams.

Monitoring and Evaluation

Audit timing of parenteral nutrition initiation and correlate with short-term metabolic outcomes and infection rates.

Research Priorities

• Research comparing parenteral nutrition initiation at <8 hours vs 8–12 hours in extremely preterm infants

Justification

While some older studies suggest that lower parenteral fluid volumes may reduce the risk of patent ductus arteriosus and necrotising enterocolitis (NEC), these findings come from small, pre-surfactant era trials and observational studies with potential confounding. The overall certainty of the evidence is very low, and no significant benefit has been demonstrated for mortality, bronchopulmonary dysplasia or neurodevelopment. In contrast, lower fluid volumes were associated with increased postnatal weight loss, and the risk of dehydration remains unclear. Given this uncertainty and the clinical importance of individualised care, we suggest either lower or higher parenteral fluid volumes may be used, guided by the infant’s condition, humidity settings, local protocols and biochemical results.

Subgroup considerations

No subgroup analysis was available by gestational age, birthweight or ethnicity. Individualised care based on gestation, clinical status, and fluid balance monitoring is particularly crucial for extremely preterm and extremely low birthweight infants, who may be more vulnerable to both fluid overload and dehydration

Implementation considerations

Given the lack of strong evidence favouring one approach over another, standardised yet flexible fluid protocols are recommended. These should include a clear framework for adjusting fluid volumes based on gestational age, postnatal age, clinical condition and weight trends while monitoring consistent fluid balance.

Monitoring and evaluation

Routine data collection on daily fluid intake/output, weight change, biochemistry, and clinical outcomes (e.g., NEC, PDA, BPD) should be part of standard care. If services make trial adjustments to fluid volumes (e.g., more restrictive approaches), then a local audit is essential to assess safety, tolerance, and short-term outcomes. Data should be disaggregated by ethnicity to ensure that equity is monitored and prioritised.

Research priorities

• Studies comparing lower versus higher parenteral fluid volumes in contemporary neonatal care settings, including the use of surfactant, antenatal steroids and early feeding strategies.
• Subgroup analyses based on gestational age, birthweight and ethnicity, particularly for Māori and Pacific populations.

Justification

Current RCT evidence does not show statistically significant benefits or harms on critical clinical outcomes (e.g. neurodevelopment, mortality, NEC); however, these studies were not conducted in Aotearoa New Zealand where the soil is iodine-deficient.

Subgroup considerations

No subgroup analyses were available by gestational age, birthweight, or ethnicity. In settings where parenteral nutrition or breastmilk is iodine-deficient, supplementation is likely more critical. Further research should stratify outcomes by gestation and type of nutrition provided.

Implementation considerations

Iodine intakes from parental and enteral nutrition sources should align with ESPGHAN 2018 and ESPGHAN 2022 recommendations (1–10 μg/kg/day parenterally, 11–55 μg/kg/day enterally). Ensure parenteral intake from standard solutions is on the higher side of the range (≥5 μg/Kg/day) due to the high risk of iodine deficiency in the Aotearoa New Zealand context.

Monitoring and evaluation

Audit of iodine intake via parenteral/enteral nutrition. Consider monitoring of thyroid function in high-risk infants or those on prolonged parenteral nutrition and monitor urinary iodine.

Research priorities

• RCTs comparing different iodine doses and routes of administration (enteral vs parenteral) conducted in Aotearoa New Zealand
• Long-term neurodevelopmental outcomes by iodine exposure level in the Aotearoa New Zealand population
• Exploration of iodine status in Aotearoa New Zealand's preterm population, especially for pēpi Māori, Pacific families and underrepresented groups

Subgroup Considerations

Extremely preterm infants may require lower initial rates and more gradual escalation due to immature glucose metabolism.

Implementation considerations

Protocols for standardised initiation and titration of glucose infusions should be in place; ensure glucose monitoring systems are available.

Monitoring and evaluation

Monitor glucose concentrations at least 4–6 hourly during the first 24–48 hours and audit incidence of hypoglycaemia.

Research priorities

• Trials comparing different glucose initiation strategies in the first 48 hours after birth

Justification

The evidence comparing insulin therapy with glucose reduction for hyperglycaemic preterm infants is of very low certainty and based on small, dated trials with limited reporting on critical outcomes. Insulin treatment may lead to improved weight gain and faster progression to full enteral feeds and was associated with fewer sepsis episodes in one small trial. However, there was no evidence of benefit for mortality, length, head circumference or bronchopulmonary dysplasia and both interventions had similar hypoglycaemia risks. Long-term neurodevelopmental outcomes were not reported. The limited evidence means no strong recommendation can be made. Insulin may benefit nutritional intake but requires careful monitoring due to its potential risks. Glucose reduction is simpler and safer in many settings but could compromise energy intake and growth in some infants. At least 4 to 8 mg/Kg/minute is needed to meet the minimum glucose needs to sustain the brain’s metabolic needs.

Subgroup considerations

All included studies involved extremely low birthweight infants; however, these are the infants at the highest risk of hyperglycaemia.

Implementation considerations

Insulin use requires protocols for titration, blood glucose monitoring and training to ensure safe administration. Glucose reduction requires careful nutritional monitoring to avoid energy and protein deficits. If glucose load is reduced, ensure the glucose infusion rate is at least 4 to 8 mg/Kg/minute to meet the minimum glucose needs to sustain the brain’s metabolic needs. If insulin is used, ensure the glucose infusion rate is in the recommended range of 6 to 11 mg/Kg/minute (9 to 16 g/Kg/day) NICE 2022.

Monitoring and evaluation

Blood glucose should be monitored closely regardless of the method used.

Research priorities

• Large, high-quality trials comparing insulin and glucose reduction in hyperglycaemic preterm infants, with long-term neurodevelopmental follow-up.
• Studies evaluating the safety and efficacy of insulin in different birthweight and gestational age groups
• Ethnicity-stratified analyses to identify differential impacts and support equitable treatment choices, especially whānau Māori, Pacific families and underrepresented groups experiences.

Subgroup considerations

Caution is needed in extremely preterm infants and those with pre-existing metabolic instability.

Implementation considerations

No implementation required

Monitoring and evaluation

If serum triglyceride concentrations are monitored, this should be no more than once a day.

Research priorities

• Whether routine monitoring of serum triglyceride concentrations and reduction of the lipid emulsion dose are beneficial and if so at what serum triglycerides concentration should the lipid emulsion dose be reduced.

Justification

Low certainty evidence suggests a large beneficial effect of fish oil-only lipids on the resolution of intestinal failure-associated liver disease, with an odds ratio of 6.1 (95% CI 2.3 to 16.6) and a substantial absolute increase in recovery (403 more per 1,000 infants treated). While the data are derived from non-randomised and retrospective studies, they consistently report improved biochemical liver function indices. No increase in adverse effects such as sepsis or mortality has been demonstrated; however, again, the evidence is weak. Fish oil only lipid emulsion is considerably more expensive (~$110/day vs. ~$35/day for SMOFLipid in 2025). However, when used selectively as rescue therapy, this cost may be justified given the potential to reduce complications and long-term liver damage.

Subgroup considerations

Infants with established intestinal failure-associated liver disease are most likely to benefit from fish oil-based lipid. There is no direct evidence specific to pēpi Māori but given higher rates of preterm birth and comorbidities, equitable access to rescue therapy is a key consideration.

Implementation considerations

Fish oil-only-based lipid is already used on a case-by-case basis. Local protocols should ensure clear criteria for initiation (e.g., rising conjugated bilirubin, failure to respond to standard management) and should include guidance on volumes of Omegaven required (as this is twice the usual volume of SMOF), amount of fat-soluble and water-soluble vitamins added and to beware of potentially harmful side effects e.g. vitamin E toxicity. Fish oil only based lipid lacks essential fatty acids such as omega 6, which may not meet the complete nutritional needs of growing preterm infants if used long-term or as a sole lipid source. Fish oil-only-based lipids should be used in conjunction with a multidisciplinary approach that includes a referral to the National Intestinal Failure and Rehabilitation Service.

Monitoring and evaluation

Monitor liver function tests (bilirubin, ALT, GGT) and serum fat-soluble vitamins regularly, paying particular attention to vitamin E. Document duration and outcomes of fish oil-only based lipid use to inform future evaluations.

Research priorities

• High-quality studies comparing fish oil only with multi-component lipids in infants with intestinal failure-associated liver disease
• Long-term neurodevelopment and growth outcomes following fish oil only lipid emulsion exposure
• Optimal dosing and duration of fish oil only lipid emulsion in rescue therapy
• Ethnic equity analyses to determine access for pēpi Māori, Pacific families and underrepresented groups

Subgroup Considerations

Extremely preterm infants at risk of retinopathy of prematurity may benefit more from LC-PUFAs.

Implementation

~1.5 g/kg/day SMOFLipid ensures an adequate intake of linolenic acid. Most NICUs already use SMOFLipid. Standardisation of parenteral lipid emulsions across units would be beneficial.

Monitoring and Evaluation

Track clinical use and any adverse effects related to LC-PUFA products.

Research priorities

• What is the optimal dose of parenteral and enteral LC-PUFA supplementation and do the benefits outweigh the risks.

Justification

The recommendation is based on expert consensus and current practice because the available evidence is of very low certainty and is largely indirect, with limited data specifically addressing very preterm infants. A before–and–after study reported a reduction in catheter-related sepsis and central line-associated bloodstream infections with longer aqueous solution hang times. However, small RCTs reported no significant effect on bloodstream infections when changing lipid sets less frequently. One observational study showed a halving of central line infections (0.4 vs. 0.8 per 1000-line days, p<0.01) with longer aqueous hang times. Cost analyses indicated substantial annual savings when switching from 24 to 48-hour hang times and reduced workload for nurses. Nonetheless, concerns about infection risk and nutrient loss persist due to the very low certainty of evidence and lack of data in extremely preterm populations.

Subgroup considerations

Standardised longer hang time protocols may improve the consistency and equity of parenteral nutrition practices nationally. Pēpi Māori are at increased risk of sepsis and suboptimal outcomes; standardised, efficient parenteral nutrition care may reduce these inequities.

Implementation considerations

Hospital policies may need revision to support hang time alignment across centres.

Monitoring and evaluation

Implement audit systems to monitor:
Rates of CLABSI and other bloodstream infections specifically include data on Māori and Pacific infants to monitor the impact on equity, cost tracking, nurse workload and satisfaction

Research priorities

• RCTs in very preterm infants comparing infection rates and mortality across hang time protocols
• Nutrient and stability and safety data for extended hang times in Aotearoa New Zealand hospital settings
• Health economic modelling based on local cost data, including costs associated with decreased infection rates

Justification

High-and moderate-certainty evidence shows uncertain effects on mortality, infection, growth and major morbidities, indicating no consistent advantage of ceasing parenteral nutrition at lower fluid volumes. Limited very low-certainty evidence suggests potential growth benefits at or above 120 ml/Kg/day, but this requires further confirmation.

Subgroup considerations

There is currently no birthweight, GA or ethnicity-specific data available. Pēpi Māori may be at increased risk of delayed enteral feeding due to systemic barriers. Clear guidance to delay stopping IVN until ≥120 ml/Kg/day may help standardise care and improve equity.

Implementation considerations

Develop unit protocols that define a minimum enteral volume threshold (e.g. ≥120 ml/Kg/day) for stopping parenteral nutrition. Provide staff training on risks associated with early parenteral nutrition cessation. This is the minimum enteral volume and parenteral nutrition can continue until full enteral feeds are reached. Ensure adequate lactation and feeding support to promote enteral feed advancement.

Monitoring and evaluation

Track adherence to defined parenteral nutrition cessation thresholds by unit and patient characteristics. Monitor rates of late-onset sepsis and postnatal faltering growth across gestational age groups, birthweights and ethnicities. Evaluate changes in practice and outcomes following the implementation of new guideline.

Research priorities

• Studies comparing different cessation thresholds for stopping parenteral nutrition, including outcomes beyond discharge
• Research on the impact of parenteral nutrition cessation volume on Māori, Pacific and under-represented groups
• Cost-effectiveness analyses considering sepsis-related costs and long-term growth/neurodevelopment

Subgroup Considerations

Protocols for energy targets should be standardised.

Implementation

Ensure standard parenteral nutrition solutions that meet these recommendations are available.

Monitoring and Evaluation

Audit energy and macronutrient delivery vs. recommendations

Research Priorities

• Early vs. delayed energy escalation
• Determine optimal energy and macronutrient intake for survival, growth and long-term development

Justification

Neurodevelopmental outcomes: Evidence is very low to moderate certainty and mostly shows no clear benefit or harm, with some isolated signals of adverse effects (e.g. lower Bayley III Language scores). Evidence at 2 years of potential for harm with higher parenteral protein intakes, but preliminary evidence at 6 years suggests no effect. For growth, moderate-certainty evidence suggests that higher parenteral protein reduces the risk of postnatal faltering growth and modestly improves head circumference growth to discharge. Higher protein intake may increase the risk of refeeding syndrome. No consistent signal for NEC, sepsis or bronchopulmonary dysplasia. Standardising parenteral protein intake and ensuring all infants receive adequate protein supports equity, especially for whānau Māori and Pacific families who experience higher rates of preterm birth and complications.

Subgroup considerations

Most evidence applies to extremely preterm or very low birthweight infants (<32 weeks or <1500 g) as this is the group where the intervention is most applicable.

Implementation considerations

Standardise protocols nationally to start parenteral protein at ≥2 g/kg/day on the day of birth, increasing to a maximum of 3.5–4.0 g/kg/day to align with NICE guidance. Monitor clinical parameters of refeeding syndrome, particularly during the first 5-7 days, especially in infants with birthweight <1,000 g or <10th centile. Provide guidance on refeeding syndrome testing and phosphate supplementation. Ensure concurrent non-protein energy supply (at least 65–80 kcal/kg/day) to support protein anabolism and reduce the risk of protein catabolism or acidosis.

Monitoring and evaluation

Track actual protein intakes, not just prescribed. Regularly audit growth and feeding information e.g. days of parenteral nutrition and protein-to-energy ratio.

Research priorities

• Long-term neurodevelopmental outcomes of different parenteral protein regimens and different amino acids formulations with outcomes by ethnicity.

Justification

There is evidence of better outcomes with Vitamin A (intramuscular) and D supplementation without harm at low overall cost. However, intramuscular vitamin A injections may not be feasible or acceptable in the Aotearoa New Zealand setting. There is moderate to very low certainty of evidence overall. A few outcomes such as reduction in retinopathy of prematurity (Vitamin A) and bronchopulmonary dysplasia (Vitamin C and D)—suggest potential benefits with higher vitamin intake. However, most outcomes show no statistically significant difference between higher and lower doses, and the precision is low across many estimates, with wide confidence intervals and small sample sizes. Despite the limited evidence, the anticipated desirable effects on important outcomes and the very low risk of harm (with no significant increase in adverse outcomes) support a cautious, conditional recommendation in favour of higher vitamin intake. Cost is trivial, and feasibility and acceptability are high, particularly given this practice is already part of standard care in Aotearoa New Zealand.

Subgroup considerations

Supplementation may offer increased benefit to populations at high risk of Vitamin D deficiency.

Implementation considerations

ESPGHAN dosing of parenteral vitamins is recommended. Close monitoring of individual vitamin intakes is warranted due to wide recommended ranges and the potential risk of providing excessive supplementation e.g. vitamin E, which is provided in very high doses in SMOFLipid emulsion. Family education and culturally safe communication should be prioritised, especially when engaging with whānau Māori, Pacific families and underrepresented groups.

Monitoring and evaluation

Monitor for signs of both deficiency and toxicity, with appropriate lab tests where indicated. Consider including vitamin supplementation data in national neonatal audits to inform future quality improvement and research.

Research priorities

• High-quality RCTs of individual and combined vitamins in preterm infants (stratified by gestational age), with appropriate pēpi Māori, Pacific families and underrepresented groups representation.
• Pharmacokinetic studies to determine vitamin stability and optimal dosing in diverse preterm populations.
• Research to clarify the role of maternal vitamin status, particularly vitamin D, on neonatal outcomes and maternal supplementation as an option to increase vitamin intakes from breastmilk if feasible.
• Economic evaluations of vitamin supplementation strategies. Evaluation of long-term outcomes, including school-age development.
   

Subgroup Considerations

Adjustments may be needed in cases of cholestasis or renal impairment.

Implementation

Routine standard solutions and monitoring protocols can facilitate standardisation.
Provide guidance on refeeding syndrome testing and phosphate supplementation.

Monitoring and Evaluation

Routine monitoring of biochemical markers and growth parameters is necessary.
Monitor clinical parameters of refeeding syndrome, particularly during the first 5-7 days, especially in infants with birthweight <1,500 g or <10th centile.

Research Priorities

• RCTs to determine optimal dosing for micronutrients in extremely preterm and growth-restricted infants.