DIAMOND - DIfferent Approaches to MOderate & late preterm Nutrition: Determinants of feed tolerance, body composition and development

Once consented participants will be randomised to one of eight conditions involving different feeding strategies. Babies will be randomized to one of eight conditions to receive either an intravenous nutrition amino acid solution or dextrose solution, supplemental milk whilst waiting for expressed breastmilk feeds or exclusively breastmilk, and taste / smell given prior to gastric tube feeds or no taste / smell prior to gastric tube feeds. The participants will receive the allocated intervention until breastmilk feeding is fully established and the intervention is no longer required. The eight conditions are below: Condition 1 Parenteral nutrition, breastmilk + supplemental milk to meet enteral fluid volumes, taste/smell with tube feeds Condition 2 Parenteral nutrition, exclusively breastmilk, taste/smell with tube feeds Condition 3 Parenteral nutrition, breastmilk + supplemental milk to meet enteral fluid volumes, no taste and smell with tube feeds Condition 4 Parenteral nutrition, exclusively breastmilk, no taste and smell with tube feeds Condition 5 Intravenous dextrose solution, breastmilk + supplemental milk to meet enteral fluid volumes, taste/smell with tube feeds Condition 6 Intravenous dextrose solution, exclusively breastmilk, taste/smell with tube feeds Condition 7 Intravenous dextrose solution, breastmilk + supplemental milk to meet enteral fluid volumes, no taste and smell with tube feeds Condition 8 Intravenous dextrose solution, exclusively breastmilk, no taste and smell with tube feeds Interventions and comparators (i) parenteral nutrition vs dextrose intravenously; (ii) supplemental milk (donor breastmilk if available, else infant formula) vs only mother’s own milk as available; (iii) infants exposed to smell and taste of milk prior to every tube feed vs no exposure (milk administered only via gastric feeding tube). All babies will receive nutrition according to individual neonatal intensive care unit practices. The volume of fluids, parenteral fluid adjustments, amount of enteral feed and frequency of increases will as per clinician's discretion and no guidelines will be given around this. In addition, babies will be randomised to one of eight conditions. It is important to note that the first two interventions only apply until the baby is established on full enteral feeds with mothers’ own milk, which remains the primary nutritional goal. If randomised to receive smell and taste prior to tube feeds, this intervention will continue until the baby is no longer receiving any gastric tube feeds. Parenteral nutrition: if randomised to receive parenteral the baby will receive an amino acid solution P100 (according to local hospital practice) intravenously, either by peripheral or central line as deemed appropriate. Administration of intravenous lipid is at the discretion of the clinical team, as is administration of any supplementary fluids, such as dextrose solution. Babies not randomised to parenteral nutrition will receive intravenous dextrose solution only (according to local hospital practice). Milk supplement: if randomised to receive milk supplement, the baby will receive donor breastmilk or infant formula (according to local practice) whilst waiting for mother's breastmilk to meet prescribed fluid amounts, any breastmilk the mother provides will be given to the baby in preference of formula and formula will only be used to make up any deficit as per medical team fluid prescription. Babies with a birthweight < 2 Kg will receive preterm infant formula; babies with a birthweight > 2 Kg will receive standard infant formula. Babies not randomised to receive milk supplement will only receive mother’s breastmilk as available which is standard practice given it takes a couple of days for mother's breastmilk to come in. If the baby's birth weight is < 2000g breastmilk will be fortified once receiving 5 mL every 2 hours as is standard practice in the units. As it is a pragmatic trial we anticipate that after a certain number of days some clinician's will not be happy to wait any longer for breastmilk it will then be up to their discretion on whether to start formula or intravenous nutrition depending on their "normal"practice". We do not propose any recommendations on number of days to wait as currently there is no evidence to guide practice therefore we cannot make any assumptions. Commercially available preterm infant and standard formula will be used depending on the site. Taste and smell: If randomised to receive taste and smell, the baby will be exposed to the taste and smell of the milk feed prior to every enteral feed. If the baby is receiving both breastmilk and supplementary formula, the smell and taste will be of breastmilk if available, but if there is insufficient breastmilk, then smell and taste can be of formula. However, if the baby is randomised to not receive supplementary infant formula then smell and taste can only be provided with breastmilk and taste should be given in preference to smell. To administer smell place 0.1 – 0.5 mL of milk onto a piece of gauze or cotton swab and place by the baby’s nose to remain in place until completion of the enteral feed. To administer taste for preterm babies give 0.2 mL of milk in a syringe on to the tip of babies tongue. Both smell and taste should be given immediately prior to administering the tube feed. The goal for all babies enrolled in the study is to transition to full feeds of expressed breastmilk as soon as possible. If at any time the responsible clinician feels that any of the randomised interventions is no longer appropriate, they may withdraw the baby from the relevant intervention for clinical reasons. They will be encouraged to discuss this with the trial Lead Investigator, Principal Investigator or another member of Steering Group before making the decision. The baby will remain in the allocated condition group for the purposes of analysis (intention-to-treat principle).