Proceedings, 2023, ABREAST Conference 2023

Sodium (Na) and potassium (K) concentrations in breastmilk are often used as biomarkers to define mastitis in lactating women and can be measured with small portable point-of-care ion-selective electrodes (ISEs). The aim of this study was to test the ISEs at the point of care for accuracy and acceptability in women with mastitis. Up to 5 mL of expressed breastmilk from the affected breast of 43 women with mastitis was collected at three timepoints (day 1, 3, and 10). Immediate Na and K ISE testing was later compared to the laboratory measure of inductively coupled plasma–op-tical emission spectrometry (ICP-OES). The results revealed a statistically significant difference in Na and K concentrations between the point-of-care and laboratory testing (both p = 0.001, Wilcoxon signed-rank test); however, the difference was not statistically significant when compared for Na:K ratio (p = 0.49, Wilcoxon signed-rank test). The Bland–Altman limits of agreement were acceptable, with the majority of measurements lying within two standard deviations of the mean (Na: 94%; K: 95%; and Na:K: 96%). The testing techniques were significantly correlated for Na (R² = 0.79, p = 0.001) and Na:K (R² = 0.99, p = 0.001). Overall, participants rated the ISE point-of-care testing as very acceptable. In conclusion, immediate ISE point-of-care testing for breastmilk Na:K ratio in women with mastitis is clinically accurate and acceptable. Full article

Mastitis is an inflammatory condition of the breast, with or without accompanying infection. The recent release of the revised Academy of Breastfeeding Medicine protocol on mastitis (Clinical Protocol #36) has caused controversy within the field. The updated protocol positions multiple typical human milk commensal bacteria as causative agents of mastitis. However, data to support these relationships are lacking. Here, we critically review human and animal evidence for the role of the milk microbiota in mastitis. Only three metataxonomic/metagenomic studies have characterized the human milk microbiome in cases of mastitis. These studies consistently report reduced alpha diversity and elevated levels of Staphylococcus aureus in mastitic milk samples; however, the remaining findings presented are conflicting and inconsistent. Collectively, the three studies included 45 sub-acute mastitis cases, 24 acute mastitis cases, and 52 healthy controls, and are thus limited by low participant numbers. In addition, the studies vary in their definition of sub-acute/acute mastitis, their methodologies, and antibiotic exposure in the mastitic groups. Further, these studies provide data on the state of the microbiome during mastitis, with no data currently available on the milk microbiome preceding the onset of mastitis. These kind of longitudinal data are critical to identify candidates for disease causation. Emerging evidence from animal models is suggestive of the involvement of the gut microbiota. Studies have reported that fecal microbiota transplantation from mastitic cows to germ-free mice results in mastitis symptoms. Future studies should therefore consider the maternal microbiome more broadly when assessing the etiology of mastitis. While S. aureus is frequently recognized as a mastitis-related pathogen, data from culture-based and culture-independent studies demonstrate that this species is present in healthy women and cannot be detected in a significant portion of cases. This suggests heterogenous causes for bacterial mastitis, necessitating broader screening. Overall, data in this field are sparse, and current clinical guidelines lack high-quality evidence to support them. There is therefore a pressing need for further research in this area to better characterize the causes of mastitis and thereby underpin future therapeutics. Full article

Staphylococcus aureus is a bacterium found in the milk of up to 38% of healthy lactating mothers; however, S. aureus is isolated with increased frequency from colostrum and mastitis milk. Both of these milk types have lower lactose concentrations compared to mature milk from healthy lactating mothers, which may indicate that lactose has a role in determining whether S. aureus can survive in human milk. The aim of this study was (1) to investigate whether the presence of S. aureus in human milk is associated with the milk’s lactose concentration, and (2) to determine whether different lactose concentrations can affect the ability of S. aureus isolates to grow in vitro. Human milk samples were collected at 10 weeks postpartum from mothers participating in the Drakenstein Child Health Study (Cape Town, South Africa) and underwent NMR spectroscopy to determine their metabolome. A subset of these samples (n = 117) was cultured to isolate S. aureus. Milk samples with lactose concentrations of less than 166 mM were more likely to have S. aureus present, compared to samples with lactose concentrations of over 166 mM (p < 0.001). In vitro, the growth of S. aureus was negatively correlated with the lactose concentration of axenic culture. Lactose concentrations associated with human milk appear to have an inhibitory effect on the growth of S. aureus human milk isolates. Therefore, low-lactose human milk could potentially be a risk factor for increased S. aureus growth and the development of S. aureus-associated mastitis. Full article

Animal models show a more rapid mammary gland response and more milk with subsequent lactations, as well as impairment of lactation performance by obesity. Whilst maternal obesity is linked to reduced breastfeeding initiation, breastfeeding confidence, and duration as well as early introduction of formula, maternal adiposity, breast anatomy and milk production (MP) have not been assessed in this population. Thirty-four lactating mothers 1–6 months postpartum and with BMI range of 17–35 kg/m² participated in this study. We conducted ultrasound examination imaging to assess breast anatomy. The amount of glandular tissue (glandular tissue representation (GTR)) was classified as low, moderate, or high. Number and diameters of milk ducts as well as mammary blood flow (the resistive index) were measured. Maternal bra cup volume was calculated from current bra size. Maternal body composition was measured with bioimpedance spectroscopy. Mothers completed a questionnaire regarding their medical, obstetric and lactation history, and conducted a 24 h MP study to enable calculation of total volume, average and maximum feed volumes and breast storage capacity (24 h MP parameters). For statistical analysis, we used the correlation networks method (directions of multiple significant correlations are reported). Correlation networks show that pathways culminating in either high or low MP start as early as puberty. In this study, later menarche correlates with the absence of breast growth during both puberty and pregnancy, which further correlate with lower numbers of ducts and smaller diameters. Higher maternal adiposity correlates with larger bra cup volume (both correlate with absence of breast growth during pregnancy and low GTR) and lower 24 h MP parameters. Larger numbers of ducts and duct diameters correlate with higher parity and longer durations of previous lactations, and higher 24 h MP parameters. Mammary blood flow shows no correlations. Findings from this cross-sectional study corroborate animal studies showing that a number of modifiable and non-modifiable maternal factors may impact breast development and MP. Further research may inform interventions, such as maintaining healthy adiposity not only during pre-conception, pregnancy, and lactation, but as early as childhood and potentially infancy. Moreover, the results provide rationale for antenatal lactation assessment of women and intervention in high-risk mothers to ensure they reach their full lactation potential. Full article

Infants born sick or preterm and admitted to the neonatal intensive care unit (NICU) face challenges to their sleep and the establishment of breastfeeding. The reported sleep patterns of NICU graduates are conflicting, and there is evidence of bi-directional relationships between infant feeding and sleep challenges and poor maternal mental health. A prospective observational longitudinal cohort study was conducted on mothers of sick/preterm infants with ≥ 5 days admission to the NICU. Participants were recruited at discharge, with follow up at 2 and 6 weeks, and 3, 6, and 9 months. Infant sleep, settle and cry patterns, maternal bother with infant behaviours, and maternal confidence were measured using the Sleep and Settle Questionnaire (SSQ). Feeding information was obtained, and breastfeeding confidence was measured using the Breastfeeding Self-Efficacy Scale–Short Form (BSES-SF). Our sample consisted of 94 mothers of 101 sick/preterm infants. The range of infant birth gestations was 23–42/40 weeks, with n = 37 (36.6%) born < 33/40 weeks, n = 26 (25.7%) born at 33 to 36⁺⁶/40 weeks, and n = 38 (37.6%) born ≥ 37 weeks. The NICU length of stay was 5–173 days. Linear mixed modelling was used to examine associations between maternal bother and infant feeding and sleep variables. Night waking frequency (p < 0.001), durations of crying (day, evening, night; all p ≤ 0.001), and durations of settling (evening, p = 0.010) were positively associated with increased maternal bother scores. Maternal confidence was negatively associated with maternal bother (p < 0.001). The median BSES-SF score at discharge was 59/70, with 49.5% of mothers no longer breastfeeding by 6 months. Maternal bother did not differ according to feeding method (p = 0.44), or birth gestation subgroup (i.e., born < 33 weeks, 33 to 36⁺⁶ weeks, or term sick) (p = 0.44). Whether NICU infants were term sick or preterm was not associated with maternal concern with infant sleep and settling patterns in the 9 months after hospital discharge. Mothers of sick/preterm infants reported greater bother with increased night waking and longer durations of crying and evening settling. The findings from this study can inform individualised anticipatory guidance and support for this vulnerable population. Full article

Human milk (HM) components are highly variable, and infants consume different volumes of milk. There has been little focus on evaluating the effectiveness of protocols in estimating infant intake of milk components before studying their impact on infant outcomes. Our goal was to compare 24 h measured intake with estimated intakes from different sampling protocols in order to determine the most accurate method for estimating infant intakes of milk leptin, adiponectin, insulin, glucose and total lipids. Mothers of term infants (n = 20) collected pre- and post-feed samples and measured their infant milk intake during each feed over a 24 h period using the test weighing method at 3–6 months postpartum. Infant true intakes of HM leptin, adiponectin, insulin, glucose and total lipids were calculated by averaging the measured pre- and post-feed concentrations and multiplying by the milk intake for the corresponding feed. Intakes were then summed to provide total intake of each component over 24 h. The estimated intakes were calculated with concentrations determined using five different sampling protocols, designed to be representative of sampling protocols used in previous HM component studies: (a) morning pre-feed sampling, (b) morning post-feed sampling, (c) average of morning pre- and post-feed sampling, (d) average of three pre-feed samples from the morning (06:00–09:00), afternoon (13:00–16:00) and evening (19:00–22:00) and (e) average of six pre- and post-feed samples from the morning (06:00–09:00), afternoon (13:00–16:00) and evening (19:00–22:00). The concentration from each protocol was further multiplied by true measured intake, a constant average intake of 800 mL/24 h and a global average milk intake of 766 mL/24 h to obtain the estimated intakes (15 protocols). The average intake of HM was 791 ± 212 mL. Comparison revealed that using the average measured concentration from three sets of pre- and post-feed samples, taken in the morning (06:00–09:00), afternoon (13:00–16:00) and evening (19:00–22:00), multiplied by either true infant 24 h measured intake, a constant estimate of milk intake (800 mL) or global average of milk intake (766 mL) provided the most accurate estimation of the infant’s intake for all components (p > 0.05). To obtain accurate estimates of HM leptin, adiponectin, insulin, glucose and total lipid intake in the absence of 24 h sampling, it is recommended to use a sampling protocol that involves taking samples before and after at least three breastfeeding sessions in a 24 h period. Full article

Breastfeeding is particularly important for vulnerable preterm infants as it provides protection from infections and reduces newborn mortality. However, preterm infants are often too immature to breastfeed after birth and may have medical conditions that require admission to the neonatal nursery. The published literature on the development of preterm feeding skills has focused mostly on bottle feeding. In order to better support breastfeeding after preterm birth, there is a need for evidence on the development of breastfeeding skills in preterm infants. The aim of this study was to examine breastfeeding skill development in a group of infants born at 25–33 weeks’ gestation. Infants were assessed during weekly monitored breastfeeds from 33 weeks corrected gestational age (CGA) using the Preterm Infant Breastfeeding Behaviour Scale (PIBBS), and milk transfer was measured. Mothers rated PIBBS items—rooting, areolar grasp, latch to the breast, sucking, longest sucking burst and swallowing—and clinical staff performed test weights. Pearson correlation was used to assess changes in PIBBS scores items over time and associations between total PIBBS score and milk transfer volume. Total PIBBS scores at 33, 34 and 35 weeks’ CGA were compared between groups of infants born at <30/40 and 30–33/40 weeks using Student’s t-test. Our cohort consisted of 60 preterm mother–infant dyads recruited from the neonatal nurseries at King Edward Memorial Hospital between February 2015 and February 2016. A positive trend was found between increasing CGA and higher ratings for six PIBBS items: rooting (R² = 0.08, F (1, 164) = 13.9, p < 0.001), areolar grasp (R² = 0.11, F (1, 164) = 21.0, p < 0.001), latching (R² = 0.14, F (1, 164) = 27.5, p < 0.001), sucking (R² = 0.14, F (1, 164) = 27.1, p < 0.001), longest sucking burst (R² = 0.17, F (1, 164) = 32.3, p < 0.001) and swallowing (R² = 0.14, F (1, 163) = 26.1, p < 0.001). A higher total PIBBS score was associated with a higher milk transfer volume (mL) (R² = 0.214, F (1, 164) = 44.8, p < 0.001). When compared to infants born at 30–33 weeks’ gestation, infants born at 25–29⁺⁶ weeks’ gestation had similar PIBBS scores at 33 weeks’ CGA (9.2 ± 3.6 vs. 9.5 ± 4.1, p = 0.83) and lower scores at 34 weeks’ CGA (9.2 ± 3.4 vs. 11.7 ± 4.3, p = 0.036) and 35 weeks’ CGA (12.3 ± 3.1 vs. 14.9 ± 3.5, p = 0.031). The development of preterm breastfeeding skills advances from 33 weeks CGA with wide inter-individual variation and slower progression observed in those born < 30 weeks’ gestation. Therefore, an individualised approach to anticipatory guidance regarding breastfeeding progression during the neonatal nursery stay is needed. Findings from this study can contribute to the formation of breastfeeding information resources for clinical staff and parents of preterm infants. Full article

Numerous factors not consistently identified in pregnancy are linked with decreased breastfeeding exclusivity and durations. These factors may be considered in three domains: the anatomical, metabolic, and psychosocial domains. As fundamental research into lactation has increased, it is now often possible to identify or speculate the mechanisms by which these factors potentially reduce milk production. The first domain describes the anatomical characteristics of the breast, including intrinsic factors such as hypoplasia (underdevelopment), which may have a genetic component and can be masked by breast augmentation surgery. Hypoplasia has long been associated with the inability to make a full milk production that satisfyies the infant’s needs, although it is not possible to predict a woman’s 24-h milk production so that appropriate complementary feeds can be advised. Extrinsic causes such as breast reduction surgery impact the volume of glandular tissue, thereby reducing the synthetic capacity of the breast. Whereas nipple piercings may damage milk ducts, obstructing milk flow from the breast and thereby reducing milk supply via the autocrine pathway. Various maternal metabolic disorders (intrinsic) comprise the second domain, which includes conditions such as gestational diabetes mellitus, type 1 and 2 diabetes, polycystic ovarian syndrome (often undiagnosed), and hypothyroidism. The aberrant levels of hormones associated with these disorders, such as insulin, are also implicated in breast development, raising the possibility of reduced mammary growth in pregnancy and, consequently, milk production. Much more research is needed in this area, not only to understand mechanisms by which lactation is impacted but also to identify the women at risk of reduced lactation capacity. The third and final domain includes psychosocial issues such as short intended breastfeeding durations, a lack of breastfeeding support, and maternal anxiety and depression. With respect to anxiety and depression, their association with reduced breastfeeding is likely multifaceted, encompassing mood and the potential biochemical changes associated with these states, such as lower levels of circulating oxytocin and higher cortisol levels. Possessing a knowledge of the negative impacts of the intrinsic and extrinsic factors within the maternal anatomical, metabolic, and psychosocial domains provides the impetus for antenatal lactation screening. The antenatal identification of risk factors enables anticipatory guidance and education during pregnancy, as well as early postpartum intervention should breastfeeding issues occur. Full article

Human milk contains macronutrients possibly affecting infant and early childhood growth. Most studies suggest a sex-specific difference in macronutrient levels in favor of males, with a higher energy content from fat, lactose, and protein. Further, macronutrient levels may differ according to mixed or exclusive breastfeeding; however, the literature is inconsistent. Our aim was to investigate sex-specific differences in macronutrient content in exclusively breastfed infants in a Danish child cohort, and whether macronutrient levels differed between exclusive or mixed (breast and formula) breastfeeding. Participants were part of the prospective birth cohort Odense Child Cohort. Baseline characteristics were obtained from medical records. Weekly SMS questions were sent to the mothers until the cessation of breastfeeding, asking whether they were breastfeeding and/or formula feeding. Mothers delivered a milk sample at the planned 3–4-month examination of the infant. Macronutrient analyses were performed on 182 samples using mid-infrared transmission spectroscopy (Miris Human milk Analyzer). We included 150 mother–infant dyads with both macronutrient analysis, and SMS data on breastfeeding. Baseline characteristics did not differ according to sex. The median interquartile range (IQR) infant age at the time of sampling was 4.1 (3.7–4.5) months. A total of 39 males and 38 females were exclusively breastfed at the time of milk sampling, while 36 males and 37 females were mixed-fed. We found no significant sex-specific differences in macronutrients among exclusively breastfed infants. The median (IQR) levels for males and females, respectively, were; protein, 0.85 g/100 mL (0.77, 0.90), and 0.82 g/100 mL (0.80, 0.90), p = 0.91; lactose, 7.83 g/100 mL (7.70, 7.95), and 7.73 g/100 mL (7.53, 7.90), p = 0.17; fat, 3.23 g/100 mL (2.07, 4.37), and 3.07 g/100 mL (2.10, 3.60), p = 0.34; energy, 65.5 kcal/100 mL (54.17, 77.00), and 63 kcal/100 mL (56.00, 69.33), p = 0.13. Further, we found no significant differences in macronutrient content in human milk samples from exclusively versus mixed-feeding mothers either prior and after adjusting for confounders, p > 0.36. This study does not confirm the previous findings of sex-specific differences in macronutrients in human milk. It is still unknown if sex-specific formula products tailored to meet possible sex-specific requirements can optimize child growth. Further research on this topic is needed. Full article

Domperidone is a commonly prescribed galactagogue used off-label for lactation insufficiency. Prescriber unfamiliarity or safety concerns can lead to therapeutic delay and potential early breastfeeding discontinuation. To facilitate access, the study site pharmacy department developed a structured administration and supply arrangement (SASA) for International Board Certified Lactation Consultants to screen and initiate domperidone using a checklist. The study aimed to validate a domperidone screening tool via an analysis of its use and compliance. Records were extracted from the RedCAP^® database for the first 50 women with a documented domperidone supply and reviewed against medical records. A staff survey was distributed assessing compliance and attitudes towards the SASA. Records of supply from the RedCAP^® database revealed 34% (17/50) of patients were referred to a physician, revealing a discrepancy between RedCAP^® reporting and checklists as no referrals were documented. Overall staff satisfaction with the SASA was rated 4.6/5. In total, 77.7% (7/9) felt confident counselling and supplying domperidone with the SASA in place, and 88.9% (8/9) felt confident using the checklist to identify the appropriateness of therapy and referral to a physician. Only 55.6% (5/9) indicated the checklist was used with each screening. The SASA education package is being updated to clarify the requirements for checklist completion and standardise frameworks to document follow-up. Full article

Caesarean section birth (CS) accounts for 38% of Australian births and is known to negatively impact breastfeeding exclusivity and duration. Comparison of breastfeeding outcomes between elective (ELUSCS) and non-elective lower uterine segment caesarean section (NELUSCS) births is needed to inform clinical care. Secondary analysis was performed on data from Australian women who had birthed by CS within the previous 12 months and completed an anonymous online questionnaire. Women responded to items relating to CS birth type and feeding methods immediately post birth, during the hospital stay and in the first 2 weeks at home. Self-reported pain at those time points was rated using a scale of 0–10 with 0 indicating no pain and 10 indicating severe pain. Associations between CS birth type, pain scores and feeding methods were determined, and breastfeeding prevalence at the time of survey completion examined. Our sample consisted of 851 women at 5 ± 3.5 months postpartum, of which 435 (51.1%) were primiparous and 362 (42%) had a NELUSCS birth. Infants born by NELUSCS were more likely to receive formula (37.1% vs. 28.8% ELUSCS, p = 0.024) and less likely to be breastfed (78.6% vs. 85.9% ELUSCS, p = 0.022) during the hospital stay. During the 2 weeks after discharge, women who birthed by NELUSCS had higher mean pain scores (6.5 vs. 4.6 ELUSCS, p < 0.001). Few women expressed and fed their milk in the weeks after discharge, with higher rates seen after NELUSCS (7.7% vs. 3.5%, p = 0.017). Breastfeeding status at study completion was not associated with CS birth type (19.6% ELUSCS vs. 19% NELUSCS, p = 0.99). The study findings indicate that Australian women who give birth by NELUSCS are more likely to experience breastfeeding challenges, with higher rates of infant formula supplementation and a more painful recovery in the days and weeks after birth. In light of the higher postpartum pain scores and lower rates of exclusive breastfeeding, women who birth by NELUSCS need additional postpartum support. Full article

Diet is a key factor that shapes the gut microbiome. Maternal diet has been proposed as a potential modulator of the human milk microbiome. However, the effect of diet during lactation on the maternal microbiota remains unclear. This study, therefore, set out to determine the effect of a two-week reduced fat and sugar and increased fibre maternal dietary intervention on the maternal faecal and milk microbiota. Faecal swabs and human milk samples were collected from mothers (n = 11) immediately pre-intervention, immediately post-intervention, and 4- and 8-weeks post-intervention, and were analysed using full-length 16S rRNA gene sequencing. The maternal macronutrient intake was assessed across one week prior to the intervention using 24 h dietary recall and during the intervention using FoodWorks 10 Software. The maternal fat and sugar intake significantly decreased from pre-intervention (fat: 120.9 ± 39.4 g; sugar: 114.1 ± 40.9 g) to the first (fat: 52.9 ± 4.3 g, p < 0.001; sugar: 83.2 ± 5.1 g, p = 0.005) and second week of the intervention (fat: 52.3 ± 6.2 g, p < 0.001; sugar: 82.7 ± 6.5 g, p = 0.005). The dietary fibre intake significantly increased from pre-intervention (28.8 ± 8.3 g) to the first week of the intervention (34.6 ± 2.8 g, p = 0.012) but was not different in the second week of the intervention compared to pre-intervention. Significant changes in the bacterial composition of maternal faeces were detected after the dietary intervention, with decreases in the relative abundance of Bacteroides caccae and increases in the relative abundance of Faecalibacillus intestinalis. In human milk, a significant increase in Cutibacterium acnes and a decrease in Haemophilus parainfluenzae were detected. Significant differences in maternal faecal and human milk bacterial composition were maintained 4 to 8 weeks after the intervention. This pilot study demonstrates that short-term changes in maternal diet during lactation can alter the maternal faecal and human milk microbiota. Full article