Pilot Study of the Feasibility of Feeding Preterm Infants With Fresh Versus Frozen Mother's Own Breast Milk in the NICU

Background Necrotizing enterocolitis (NEC) is a severe inflammatory disorder of the intestine that primarily affects very low birth weight (<1,500 g)/very preterm infants (≤32 weeks' gestation); it is also the leading cause of death in the neonatal intensive care unit (NICU). The incidence of NEC (≥stage 2) in very preterm infants across Canada is 5% and increases as gestational age at birth decreases, with an incidence of 13% in infants born at <25 weeks' gestation and 10% in infants born at <29 weeks' gestation. Overall mortality from the disease has been reported to range from 15 to 30%, with more severe cases of NEC having a higher mortality rate.For infants who survive there is a significant risk of long-term complications and morbidity including neurodevelopmental issues.

Currently there is no known effective treatment for NEC. Infants are initially managed medically through bowel rest (cessation of oral feeds), abdominal decompression, administration of broad spectrum antibiotics, and provision of supportive care including parental nutrition, ventilator support, and blood transfusions as necessary. Should bowel perforation occur then surgery is required to remove damaged tissue and gas build up through laparotomy and/or primary peritoneal drainage.

Perhaps the best form of treatment for NEC is prevention. Mother's breast milk is best for preventing NEC. Breast milk contains both nutritional components (proteins, amino acids, fats, carbohydrates, vitamins, and minerals) and bioactive components (macrophages, T cells, cytokines, hormones, and growth factors) that have antimicrobial and anti-inflammatory properties. More recently, Hassiotou4 showed that breast milk is also a rich source of pleuripotent stem cells that when ingested by the infant, enter the blood stream and are incorporated into major organ systems in the infant. Although their function is unknown, it is possible that they benefit the infant through growth or regeneration. The evidence strongly suggests that fresh breast milk may have a protective effect against both infection and NEC. When mother's own breast milk is not available the options are donor breast milk or formula. The use of formula in low birth weight/preterm infants has been reported to increase the risk of developing NEC when compared with donor breast milk.

However, current practices related to handling and use of human milk raise questions about whether we are deriving maximum benefit from their use. Currently, in the standard neonatal intensive care unit (NICU) breast milk expressed by mothers of preterm infants is frozen and stored in the milk bank. When the infant requires oral feeds an order is placed with the milk bank, the oldest batch of milk defrosted, fortifier added as needed, and the milk then sent to the NICU for use. The act of freezing breast milk,either at -20°C or -80°C, along with the passage of time decreases the energy content and reduces the levels of several of the most useful components in the milk, including fat, carbohydrate,secretory immunoglobulin A,lactoperoxidase,lysozyme, antibacterial factors,and antioxidants. In addition to the decrease in these components, any stem cells in the milk have a half-life of 4 hours, so the defrosted milk does not contain any stem cells.

The current NICU breast milk feeding procedure exists as a means of ensuring that infants have consistent access to their mother's breast milk even if the mother is not able to spend time in the NICU. The process also allows for stricter quality and infection control, as well as computerised inventory and monitoring via electronic health records.However, the process deprives infants of the benefits of the cellular content of breast milk, including the stem cells.

Objective The primary objective of this study isto evaluate the feasibility and safety ofproviding very preterm infants (born at <30 weeks' gestation) with fresh milk (within 4 hours of expression).While we acknowledge that as a pilot the study will not be powered to detect a statistically significant difference, our secondary objective is to identify if this approach has the potential to improve infant outcomes, particularly with regards to the occurrence of NEC.

Our hypothesis is that it is feasible for many mothers to provide at least 1 feed of fresh breast milk (<4 hours post expression per day, and is not frozen, chilled or pasteurized) and that this may decrease the prevalence of NEC.

Design Study design The study will be a prospective cohort analytic design. Mothers who agree to be enrolled in the study will be assigned to the intervention group. In the intervention group, mothers will be asked to provide at least 1 feed of fresh milk (i.e. within 4 hours of milk expression) per day. Mothers who do not agree to provide fresh milk but are willing to allow their information to be collected and analyzed will be assigned to the control group. In the control group, the current standards of breast milk handling and feeding apply.

Study population The study population will include infants and their mothers admitted to participating NICUs (Zhengzhou Hospital, Nanjing Maternity and Child Health Hospital, Nanjing Children's Hospital, Zhejiang Maternity Hospital, Xiangya Second Hospital, Shanghai First Maternity Hospital, Nanjing Maternal and Child Health Hospital, Children's Hospital of Fudan University) which admits, on average, 500 infants per year who would be eligible for inclusion in the study. The goal is to enroll 100 infants in each of the intervention and control groups in the study. Eligible infants will include those born at <30 weeks gestation,and who have never received infant formula. For inclusion in the intervention group, the mother must bewilling to commit to providingat least 1 feed of fresh breast milk a day, 7 days a week. Infants with major congenital anomalies, receiving palliative care, or where illness of the mother or infant prevents the administration of breast milk feeds in the first week of the infant's life will be excluded from the trial.Infants whose mothers are unable to commit to providing at least 1 feed of fresh breast milk a day but who are willing to have their data collected may be included in the control group.

Enrollment Eligible mothers will be approached at their first visit to the NICU following their infant's admission. Ideally enrollment should occur before the infant's first oral feed. All mothers enrolled in the study will follow the standard protocol in their NICU for feeding their infant breast milk.

Outcomes Outcomes will be measured from birth until discharge or transfer to another neonatal unit.

The primary outcomeof the study will be:

1. Feasibility as defined by the percentage of mothers who can provide at least 1 feed of fresh milk a day from study enrollment until the infants are 32 weeks CGA, including:
2. Rates of enrollment in the intervention and control groups
3. Rate of retention in the study
4. Compliance with the study protocol
5. Number of days with at least 1 feeds of fresh milk (within 4 hours of expression)
6. Number of feeds per day with fresh milk
7. Number of feeds using frozen mother's own milk or donor milk
8. Number and type of critical incident reports including feeding errors 2. The secondary outcomes of the study will be (definitions as per Canadian Neonatal Network): (1)NEC stage 2 or above (2)NEC stage 1 (3)Mortality (4)Need for surgery for NEC (5)Late-onset sepsis (6)Growth as measured by weight, length, head circumference (7)Retinopathy of prematurity (8)Bronchopulmonary dysplasia Study Procedures Feeding protocol Standard procedures - feeding and milk fortification: Infants born at <30 weeks' gestation are initially provided with nutrition intravenously (total parenteral nutrition, TPN). If the infant is not suffering from any condition that affects gastrointestinal system functioning, oral feeds are introduced as early as possible. Milk is introduced into the stomach through a nasogastric (NG) tube according to a schedule that is dependent on the infant's weight at birth. The amount of milk is very small at first and then gradually increased if the infant tolerates the milk. The frequency of feeding varies but is most often every 2 hours or every 3 hours depending on the weight of the infant. As the amount of milk being fed to the infant increases, the amount of TPN provided is decreased until the infant is receiving all its nutrition from milk (see typical Feeding Guidelines appended to this protocol).

Once the infant is receiving all its nutrition from milk ('full enteral feeds'), fortifiers and other nutritional supplements are added to each feed. The first supplement added is human milk fortifier, four days later extra protein is also added , and then a further 3 days later vitamins and iron are added to each feed (see Feeding Guideline).

Mother's own expressed breast milk is the food of choice, following by donor breast milk, and then formula. The supplementation or replacement of mother's own milk by donor milk or formula is dependent on the mother's ability to produce milk in sufficient quantity for all feeds. Often in the first few days of life very preterm infants are fed donor milk as the mother may not have started producing breast milk at that point or may still be recovering from the birth. Mothers may also produce less milk than is needed, either initially or throughout the breast feeding period, so some feeds may be supplemented with donor milk or formula.

In every case where expressed breast milk is used currently, the milk is collected from the mother and frozen in the milk bank. Each evening the feeds for the following day are ordered and the oldest batch of breast milk defrosted and prepared. Infants move from enteral feeds to breast feeding when they are clinically judged to be able to do so.

Intervention protocol - overview:The exact same standard procedures for feeding (timing of initiation, amount, frequency, and supplementation with donor milk as needed) and fortification will be followed for study participants. Only infants who have not previously received infant formula are eligible. The only difference in feeding will be that for at least 1 feed per day the breast milk fed to the infant will be fresh, not previously frozen. 'Fresh' breast milk is defined as milk that was expressed less than 4 hours prior to the time it is fed to the infant, and is not frozen, chilled or pasteurized. The reason for this time window is that the half-life of stem cellsbreast milk is only about 4 hours.

Intervention protocol - details: Mothers will be asked to be present in the NICU and pump breast milk for a 'fresh' feedat least 1 time per day, 7 days a week using the Medela Symphony pump applied to both breasts.As this is a "real world" study, the actual volume of fresh milk feeds is not as important as the number of feeds because feed volumes are determined by a standard protocol in the NICU, and volumes of feeds administered must conform to the ability of the infant to tolerate the feeds, and cannot be dosed according to a pre-set amount. Breast milk will be expressed into a sterile plastic container as per standard NICU protocol. The container will be labelled with the mother's bar code and immediately given to a nurse dedicated to the implementation of this study (1.0FTE, see study budget). The nurse will take the milk to the milk preparation area where the details of the milk and feeds will be entered into a database and then she will prepare and bring the milk back to the NICU to be fed to the infant. The milk will not be heated.

Any additional milk expressed by the mother in the NICU or at homethat exceeds the 4 hour window will be transferred to a sterile container, frozen, and stored in the milk bank for use when the mother is not available to provide fresh milk. Milk collection and handling for feeds of previously frozen milk will be as per standard NICU protocol.

Provision of breast milk in this manner will commence when the infant is able to tolerate enteral feeds and continue until the infant is discharged from hospital.Mothers will also be encouraged to express breast milk when they are away from the unit and bring it into the unit as per standard NICU protocols. In cases where a mother is not able to produce enough breast milk for all her infant's feeds, the remaining feeds that are not being provided as fresh or frozen mother's own breast milk may be supplemented using donor milk.Infants enrolled in the study who develop NEC will cease the study protocol and receive treatment as per the NICU standard protocol.

Mothers who provide less than 1 feed of fresh milk every day will remain in the study. This will enable us to determine the feasibility of expecting mothers to provide at least 1 feed of fresh milk a day, and the percentage of mothers who can actually do so.

Safety For an individual patient, the feeding protocol will be stopped if there is an occurrence of NEC or any other medical condition that necessitates that feeds be stopped as per standard practice.

All cases of mortality, NEC, sepsis, and critical incidence reports (including mix-up of milk destined for individual infants, infection that is related to feeds, missed feeds arising from delays in milk preparation, and any other incident that raises concerns of the clinicians) that occur in infants enrolled in the study will be critically reviewed by a Data Safety and Monitoring Committee and the study stopped if concerns regarding the safety of the study are raised.Stopping criteria and decisions will be decided by the Data Safety and Monitoring Committee.

Preparation (2 months): REB approval will be obtained. Education for staff regarding the study protocol will be developed and rolled out. Education for staff will include in-service sessions that will be held before the study period begins. The necessary unit policies will be developed and implemented.

Enrollment period (9 months): Enrollment into the study will continue until 100 infants and their mothers have completed the intervention protocoland 100 infants and their mothers have completed the control protocol.

Analysis (1 month): Data collection will be analysed and results reported.

Data collection Infant outcomes: Outcomes data will be collected on paper and entered into a database for subsequent analysis. Study-specific variables will include compliance with intervention protocol, the number and volume of stem cell rich fresh milk feeds provided to study infants on each day, and the response of the infant following the feed (see attached data forms).

Infant and maternal characteristics: In addition to infant outcomes, data on maternal and infant characteristics, as well as predisposing factors will be collected.

Statistical analysis Justification for sample size:The mean number of infants born at <30 weeks' gestation and admitted to participating NICUs in 2014 was over 500 per year. We expect that 25% of mothers will consent to be in the intervention group. If the admission rateremains similar during the study period, there should be ample feasibility to complete the study.

In 2013 the mean annual rate of NEC was 9.3% in infants born at <29 weeks' gestation, so we would expect to see 9 cases of NEC in the control group. Although the study will not be powered to detect a statistically significant difference, the occurrence of 5 or less cases of NEC in the intervention group may be an indication that further study is warranted.

Analyses: The primary outcomes of feasibility (rate of consent, rate of enrolment, rate of retention, and compliance with the intervention protocol) and safety (number of critical incidence reports) will be reported.The incidence of NEC and other secondary outcomes will be compared between the two groups using descriptive methods.