Investigation of the use and outcomes of feeding very preterm infants with fortified donated and mother’s own milk

Background: Preterm birth is a major health problem, affecting nearly 10% of all live births globally. It is associated with increased health care costs and is an additional burden on parents. Although survival of preterm infants has increased with advances in medical technologies, surviving neonates are at greater risk of a range of morbidities, including necrotising enterocolitis, feeding difficulties, growth restriction, bronchopulmonary dysplasia, infections, intraventricular haemorrhage, cerebral palsy and visual and hearing problems. Feeding strategies are important predictors of prematurity outcomes. Mother’s own milk (MOM) is the preferred feeding option for all newborns, especially those born prematurely, for whom it protects against necrotising enterocolitis and improves neurodevelopmental outcomes. Unfortunately, MOM is not always available in sufficient volumes, and, in such cases, donated human milk (DHM) is often used as an alternative. Due to the increased nutritional requirements of preterm infants, expressed breast milk (EBM)—both MOM and DHM—is often supplemented with a multi-nutrient fortifier to meet their nutritional needs. The optimal time to initiate fortification of EBM is not known, and, as a result, the practice varies. There is also insufficient research to allow an effective comparison of fortified DHM with formula in situations where MOM is unavailable. This is important because fortification of EBM (both MOM and DHM) has become standard practice in contemporary neonatal care. Furthermore, the use of DHM in Scotland has not previously been described. Aims: The aim of this thesis was to investigate the outcomes of fortified EBM feeding in modern neonatal practice in very low birth-weight infants. The specific objectives were 1) to describe the timing of the introduction of human milk fortifier and to explore any pattern of change in practice over time in different regions; 2) to determine, from the literature, the optimal time to start fortification of EBM that results in improved health outcomes; 3) to audit the use of DHM in Scottish neonatal units and explore the impact of feeding with DHM on maternal milk availability; and 4) to compare the outcomes of feeding different proportions of milk, including DHM, MOM and formula, on in-hospital outcomes in infants born at less than 32 weeks gestational age admitted to Scottish neonatal units. Methods: Two stages were involved in determining the optimal time at which to start fortification. The first stage involved a scoping review of the literature to describe the timing in current EBM fortification practices and explored whether fortification is initiated early or late, depending on the degree of prematurity, fortification origin and publication date of the study (chapter 2). The second stage involved a systematic review that aimed to compare the outcomes of early and delayed fortification (chapter 3). The use and outcomes of DHM were studied in two stages. In the first stage, DHM use in Scotland was described and the impact of using DHM as first milk on subsequent MOM availability was explored. In the second stage, the in-hospital outcomes of using DHM in preterm infants were investigated. DHM use was explored through an audit against the local guidelines for DHM use. The sample was of infants admitted to the Royal Hospital for Children, Glasgow, from 2014 to 2016 who had received DHM. The impact of the introduction of DHM as the first enteral feed on subsequent MOM availability was explored (chapter 4). To investigate the in-hospital outcomes of early feeding with various combinations of fortified EBM (both MOM and DHM) and formula, all preterm infants born at less than 32 weeks gestation and admitted to Scottish neonatal units were studied. Data were sourced from the National Neonatal Research Database (NNRD) by requesting it from the Neonatal Data Analysis Unit in London. Research Ethics Committee and Health Research Authority ethical approvals were obtained. The database was explored, and relevant data were carefully cleaned, extracted and tabulated to create the final curated datasheet (chapter 5). Chi -square and Kruskal–Wallis tests were used to analyse the cohort and multivariate regression analysis was used for the study outcomes (chapter 6). Results: First, regarding the optimal timing of fortification, the scoping review showed that the timing ranged from 33 to 165 ml/kg/day. In general, recent studies (published after 2000) introduced fortifier before full feeding (150 ml/kg/day) was achieved, in contrast with older studies, which introduced fortifier after achieving full feeding (chapter 2). The systematic review identified only two randomised clinical trials, which did not show any significant difference between early and delayed fortification (chapter 3). Second, regarding DHM feeding, the audit of DHM included 165 infants and showed that the most common indication for use (in 69% of recipients) was supplementation of MOM in preterm infants. Using DHM on its own or with other milk to initiate feeding did not have a significant negative impact on subsequent MOM availability in preterm infants born at less than 32 weeks of gestation (chapter 4). Identifying and applying for the required ethical approvals and requesting and managing the data took a very long time (approximately three quarters of the PhD). The quality of the data obtained from the NNRD was very good, but daily total milk intake was poorly recorded (49% of care days was missing). The large database analysis included 1,450 preterm infants from 13 neonatal Scottish units, representing more than 70% of the population. The feeding groups were defined based on the feeding exposure (days of feeding with any amount of milk prescribed) within the first 14 days of life. These groups were exclusively/predominantly MOM fed (PMOM group, n=655), MOM supplemented with DHM (MDHM group, n=302) and MOM supplemented with formula (MF group, n= 315). Analysis of outcomes was stratified by the level of prematurity [extremely preterm (EPT) infants and very preterm (VPT) infants]. Results for EPT infants were obtained for PMOM and MDHM groups only, because there was small number of infants in MF group. At the age of one month, weight gain was not significantly different between the different feeding groups for both EPT infants [median (interquartile range ): PMOM=10 [8–13], MDHM=10 [7–13] g/kg/day , p=0.545) and VPT infants (PMOM=11 [8–14], MDHM=11 [8–14], MF=11 [8–14] g/kg/day, p=0.341), neither were any significant differences found for head growth (EPT infants: PMOM=0.6 [0.3–0.7], MDHM=0.5 [0.2–0.7] cm/week , p=0.670; VPT infants: PMOM=0.6 [0.3–0.7], MDHM=0.5 [0.2–0.7], MF=0.5 [0.2–0.7] cm/week, p=0.670). In EPT infants, the secondary outcomes, including morbidities, time to discharge home, days on parenteral nutrition and postmenstrual age at discharge, did not significantly differ between the PMOM and MDHM groups. In VPT infants, there was no significant difference in any secondary outcome between the PMOM and MDHM groups. However, infants in the MF group had significantly fewer days of parenteral nutrition than those in the MDHM group (6 [3–8] vs. 9 [6–14] days, p<0.001) and significantly lower postmenstrual age at discharge (35 [34–37] vs. 37 [35–39] weeks, p=0.011). In the multivariate regression model analysing the time to discharge home, receiving the fortifier later was associated with a significantly lower probability of early discharge home (in EPT infants, p=0.011; in VPT infants, p