Colostrum expression

Although the value of colostrum is undisputed, in some traditional cultures the unusual colour, texture and smell of colostrum compared to mature milk led people to think that it was not safe or appropriate for newborns to consume (Wickes 1953). Thus, colostrum expressing was initially performed in order to discard the colostrum, to avoid newborn infants consuming it for the first few days of life.

Anthropologists and ethnographers have documented colostrum taboos across different cultures^{(Reference Hinde6)}, and unfortunately, the practice of colostrum avoidance still exists at time of writing. For example, one in five breastfeeding mothers were found to discard colostrum in Ethiopia^{(Reference Biset, Dagnaw and Abebaw7)}. The most influential individuals for colostrum avoidance in this study were grandmothers and traditional birth attendants, indicating the effect of traditional taboos passed down from previous generations. It was believed that colostrum could not be digested properly. As a result, butter, cow’s milk or water was given instead of human colostrum for the first few days post-birth^{(Reference Biset, Dagnaw and Abebaw7)}. Similar findings of colostrum avoidance have been observed in other countries including Nigeria^{(Reference Joseph and Earland8)}, Guatemala^{(Reference Atyeo, Frank and Vail9)}, Egypt^{(Reference Abdelmenam, Youness and Arief10)} and India^{(Reference Subbiah and Jeganathan11)}. Instead of consuming colostrum, various liquids including sugar water, coconut water, honey, alcohol, castor oil and cow’s milk are among examples that have been previously documented in Asia, Africa or Latin America in order to either to ‘purify and clear’ or to ‘prepare the baby for adult life’^{(Reference Lefèber and Voorhoeve12)}.

These taboos persist despite evidence to the contrary, even evidence within the same cultures. For example, in Ethiopia, newborns deprived of colostrum were more likely to have stunted growth compared to those who received colostrum^{(Reference Teshome, Kogi-Makau and Getahun13)}. Given the agreement in the literature that colostrum is a highly valuable and nutritious milk to support optional infant health, it is hoped that the practice of expressing and discarding colostrum will cease in the near future.

Expressing of colostrum antenatally

Antenatal colostrum expression (ACE) was originally performed as a means of preparing the breasts for breastfeeding after birth. The first person to popularise ACE is noted as the British doctor Harold Waller in the 1940s, who aimed to investigate and address causes of early breastfeeding difficulties^{(Reference Waller14)}. Waller believed that women who were taught and practiced the skill of hand expression during pregnancy could use it post-birth to help manage breast engorgement. Antenatal massage and expression of colostrum also had the intention of strengthening the nipples and avoiding obstruction of the ducts.

In the British Hospital for Mothers and Babies, Woolwich, Waller conducted a study of 200 nulliparous women alternately allocated to either an ACE intervention for the last three months of pregnancy (around 28 weeks gestation) or a control group^{(Reference Waller14)}. Glass nipple shields were also used for women in either group during pregnancy to help draw out inverted nipples. After birth, women in the intervention group were encouraged to manually express out any ‘leftover’ milk in the breast after feeding their babies. The intervention itself was referred to by later researchers as the ‘Woolwich method’ or regime^{(Reference Blaikley, Clarke and Mackeith15,Reference Ingelman-Sundberg16)} .

The project was initially planned to include 200 women in each group, double the number actually included. However, due to war-related bombing, women in late pregnancy were more likely to be in the provinces, and in hospital mothers had to be regularly moved to the basement for safety, which interfered with data collection^{(Reference Waller14)}. The study therefore may have been underpowered.

Despite this, Waller’s initial results seemed promising. Women in the intervention group reported substantially lower rates of engorgement with obstruction (5 mothers compared to 19), lower rates of nipple injury (12 v. 24) and higher rates of wholly breastfeeding at six months (83 v. 42)^{(Reference Waller14)}. Although this early study seemed encouraging, other researchers have noted some important limitations to Waller’s research^{(Reference Alexander, Alexander, Levy and Roch17)}. For women with engorgement problems, hand expression of breast milk after feeds was routinely used and some mothers were given stilboestrol, a synthetic oestrogen, to inhibit lactation. Stilboestrol use was subsequently discontinued due to concerns around efficacy and safety and was banned by the United States Food and Drug Administration in 2000^{(Reference Zamora-León18)}. The published manuscript did not record how many women from each group received these treatments. In addition, the intervention group received additional support from a midwife both antenatally and postnatally, which may have contributed to improved breastfeeding rates in this group^{(Reference Alexander, Alexander, Levy and Roch17)}.

A replication of Waller’s study was conducted by John Blaikley and colleagues in Guy’s Hospital, London and published in 1953^{(Reference Blaikley, Clarke and Mackeith15)}. The intervention was slightly different, in that massage of the breasts and expression of colostrum was only started from around 32–36 weeks gestation (as compared to around 28 weeks). Similar results were found in Waller’s study – women in the intervention group who performed ACE were significantly more likely to be breastfeeding their infants at 6 months of age compared to the control group (53 v. 26%, respectively)^{(Reference Blaikley, Clarke and Mackeith15)}. However, this difference in long-term breastfeeding was only for women with hospital births, not for women who birthed at home, and the researchers concluded that post-birth support provided was more important than the support provided during the antenatal period^{(Reference Blaikley, Clarke and Mackeith15)}. The value of ACE was determined to be learning the skill of hand expressing to apply post-birth if required, rather than the physicality of expressing colostrum in pregnancy.

Further doubt was cast over the Woolwich method in the late 1950s. Research in Stockholm by Axel Ingelman-Sundberg investigated antenatal breast massage and colostrum expression from the 20th week of pregnancy, in conjunction with manual expression of residual milk following breastfeeding after the baby was born^{(Reference Ingelman-Sundberg16)}. There were 313 women in the intervention group and 343 in the control group. Rates of full breastfeeding upon discharge from hospital (one-week post-birth) were similar between intervention and control women (94 v. 93%, respectively). They do not seem to have been followed up past this point. These breastfeeding rates were higher than observed in Britain at around the same time, with Waller quoting 80% in a 1944 Ministry of Health report as background to his work^{(Reference Waller14)}. Ingelman-Sundberg also investigated mastitis in the Stockholm study, noting a concerning trend: ‘The frequency of mastitis during the stay in hospital was 0.88 ± 0.51 per cent among the controls and 2.88 ± 0.95 among the antenatally treated patients, the P-value of the difference being less than 0.06’. (Note there is very little detail given about this variable in the original article - it is described as a ‘frequency of mastitis’ and given in units of ‘per cent’. We are not certain if this means the percentage of women affected or something to do with the number of times during the week. We have used a direct quote so readers can make up their own mind.) Ingelman-Sundberg concluded that ACE was of little value as a routine treatment, although noted that it may be helpful for women with inverted nipples.

It is interesting to note that ACE was not practiced in isolation with these studies, but expression post-breastfeeding was also performed by the mothers. Given the supply-demand principle of lactation, it is likely that hand-expressing additional milk after feeds led to a greater milk supply than was required. This may have contributed to higher rates of engorgement and mastitis, potentially explaining the results observed by Ingelman-Sundberg^{(Reference Ingelman-Sundberg16)}.

A move away from ACE

During the 50 years that followed, few studies were conducted on ACE – those that were conducted generally had low participant numbers (<100), often with inadequate explanation of how sample size was determined. Other substantial limitations included a lack of reliable, validated tools, a lack of ethics approval documentation and ambiguous randomisation procedures^{(Reference Chapman, Pincombe and Harris19)}. Studies on ACE and other forms of breast preparation during pregnancy did not show much benefit for breastfeeding, and by the early 1990s breast preparation during pregnancy was no longer recommended^{(Reference Chapman, Pincombe and Harris19)}.

Issues with safety were also raised, particularly in regard to ACE leading to early labour^{(Reference Chapman, Pincombe and Harris19)}. There were concerns about breast stimulation leading to the release of oxytocin, a hormone that helps milk in the breasts to flow and fill the ducts for breastfeeding. The same hormone also stimulates uterine contractions in labour during childbirth (and after birth). There were concerns that ACE could trigger premature labour^{(Reference Chapman, Pincombe and Harris19)}.

In a Cochrane review of the literature^{(Reference Kavanagh, Kelly and Thomas20)}, the relationship between breast stimulation (not ACE per se) and birth outcomes including inducement of labour was reported from over 600 participants across five trials. Analysis of trials comparing breast stimulation with no intervention in term women (from 37 to 42 weeks gestation) found that women performing breast stimulation were more likely to be in labour after 72 hours – the proportion not in labour was 62.7% for the stimulation group versus 93.6% for the control group. This was considered a beneficial result due to the women being considered full-term and being able to avoid induction of labour. A reduction in postpartum haemorrhage rates was also observed for the breast stimulation group, with no significant differences in the caesarean section rate. The authors recommended more research to further investigate these findings^{(Reference Kavanagh, Kelly and Thomas20)}.

In addition, breastfeeding (and therefore resulting breast stimulation) during pregnancy appears to be unrelated to adverse outcomes in low-risk pregnancies, although more research is required^{(Reference Stalimerou, Dagla and Vivilaki21)}. Likewise, it has been shown that less oxytocin is released in response to breast stimulation during pregnancy than when a woman is not pregnant^{(Reference Amico and Finley22)}. Taken as a whole, the literature suggests that it is unlikely that oxytocin released during pregnancy from breast stimulation will lead to adverse outcomes.

Renewed interest in ACE: antenatally expressed colostrum as an alternative to formula

After birth, if a baby requires additional sustenance beyond breastfeeding, for instance, if they are hypoglycaemic, formula or glucose gel is normally given. But with the advent of freezers and economically priced syringes and other collection containers, it is now feasible to collect and store any colostrum that has been expressed during pregnancy. Mothers can collect colostrum expressed in pregnancy and store it in their home freezer. If medically indicated, this colostrum can then be defrosted and given to the baby after birth, in place of formula. As a result, interest in ACE has increased in recent times. Having a supply of antenatal colostrum available is particularly relevant for newborns who are at higher risk of hypoglycaemia, including babies of mothers with diabetes mellitus (including gestational diabetes), babies with oral issues (cleft lip/palate), preterm babies and small or large for gestational age babies. Traditionally, these babies are given formula to help increase blood glucose levels quickly. However, if mothers have collected colostrum during pregnancy, this can be used in place of formula if required.

ACE shown to be safe

To date, the foremost research in this area is the Australian Diabetes and Antenatal Milk Expressing (DAME) trial^{(Reference Forster, Moorhead and Jacobs23)}. Investigating ACE is of particular interest when considering pregnant women with diabetes, as their babies are more likely to become hypoglycaemic after birth and require treatment such as formula. This is due to a relative increase in infant insulin secretion following being in a hyperglycaemic intrauterine environment^{(Reference Kole, Ayala and Clark24)}. The DAME trial randomised 635 women with low-risk diabetes into either standard care or an ACE intervention with women shown how to express colostrum twice a day from 36 weeks of pregnancy. The trial demonstrated that antenatal expressing was safe: there was no difference observed between ACE or control groups for mean gestational age at birth (38.6 ± 1.03 weeks for babies in the antenatal expressing group v. 38.7 ± 0.98 weeks for babies in the control group, adjusted mean difference –0.05, 95% CI –0.21, 0.10) or admissions to the Neonatal Intensive Care Unit (46/317, 15% assigned to antenatal expressing v. 44/315, 14% assigned to standard care; adjusted relative risk 1.06, 95% CI 0.66, 1.46, adjusted for diabetes type (gestational or not), parity (first baby or not), education (degree or not), and age). The authors concluded that there is no harm in advising women with diabetes in pregnancy, at low risk of complications, to express colostrum for the last few weeks of gestation.

Since the DAME trial, studies investigating ACE have found it to be a safe practice, with no differences in weeks gestation at birth compared to controls^{(Reference Chen, Washio and Liu25–Reference Moorhead, Amir and Crawford32)}. In general, safety of ACE from ∼36 weeks gestation has been established^{(Reference Demirci, Glasser and Himes28,Reference Estafanous, Lee and Thomas29,Reference Demirci, Glasser and Bogen33,Reference Johnsen, Klingenberg and Brand34)} . This is notably later than some previous studies, which included hand expression as early as 20 weeks gestation^{(Reference Ingelman-Sundberg16)}.

Although the DAME trial provided good evidence of safety in low-risk women with diabetes, researchers suggest that further studies into the introduction of ACE into high-risk groups, and earlier initiation of ACE, are necessary before it can be advised in these situations^{(Reference Glavey and Fallon30,Reference Demirci, Glasser and Fichner35,Reference Moorhead, Amir and Forster36)} .

Learning about and practicing ACE

ACE education seems to be acceptable within a range of cohorts^{(Reference Chen, Washio and Liu25,Reference Glavey and Fallon30,Reference Demirci, Glasser and Fichner35,Reference O’Sullivan, Cooke and McCafferty37)} . Most studies investigating ACE use a midwife or lactation consultant to teach pregnant women how to perform ACE. Due to staffing and funding restrictions, this is often difficult to do on a large scale. ACE instruction via a pre-recorded expert video has been previously trialled by our research team^{(Reference O’Sullivan, Cooke and McCafferty37)}. Our team has demonstrated that the use of an educational ACE video is both acceptable and efficacious in providing instruction to pregnant women, which they could then use to practice on their own and rewatch as required^{(Reference O’Sullivan, Cooke and McCafferty37)}.

Barriers to performing ACE include illness, burden of existing appointments for mothers with diabetes, and lack of time^{(Reference Davis, Glasser and Clemens27,Reference Demirci, Glasser and Fichner35,Reference Moorhead, Amir and Forster36,Reference Foudil-Bey, Murphy and Dunn38)} . For example, mothers with diabetes during pregnancy already have the increased burden of attending extra appointments and completing blood glucose tests throughout the day, in addition to their usual activities at work or home^{(Reference Moorhead, Amir and Forster36)}.

Overall, the experience of pregnant women learning about and practicing ACE is largely positive^{(Reference Ablett, Hopper and Maslin39)}. Qualitative research has shown that the hand expression and subsequent frozen storage of colostrum in the weeks prior to birth helps build confidence with breastfeeding^{(Reference Brisbane and Giglia40)}. Increased breastfeeding self-efficacy is a common theme that has been reported in multiple studies of ACE^{(Reference Juntereal and Spatz41)}, for example, it can give women confidence in their milk supply for post-birth breastfeeding^{(Reference Demirci, Glasser and Fichner35)}. Feeling more comfortable and confident with expressing breastmilk has also been reported^{(Reference Davis, Glasser and Clemens27)}.

Although most research has been conducted in Australian and US cohorts, a small study of Scandinavian women also found that ACE was well accepted by all participants, and almost all (26/27, 96%) would recommend it to other pregnant women^{(Reference Johnsen, Klingenberg and Brand34)}. A mother’s attitude towards breastfeeding has been shown to be a strong modifiable factor to help increase breastfeeding duration^{(Reference Cox, Giglia and Binns42)}, suggesting ACE may help to better prepare women for breastfeeding by increasing their confidence.

However, performing ACE can be disempowering as well as empowering for women. In interviews with a subset of 10 women participating in the DAME trial, ACE was on one hand associated with feelings of achievement and control in a pregnancy that was ‘controlled by diabetes’^{(Reference Moorhead, Amir and Forster36)}. But on the other hand, some women who couldn’t collect colostrum, or only collected a small amount, felt disappointed and anxious, with a sense that their bodies had not worked as they had wanted^{(Reference Moorhead, Amir and Forster36)}.

Negative feelings of performing ACE have also been reported in other studies. These included maternal discomfort with expressing, embarrassment, and concern about not being able to express colostrum^{(Reference Demirci, Glasser and Fichner35)}. Anxiety around not being able to produce colostrum through hand expression is a common theme^{(Reference Chen, Washio and Liu25,Reference Moorhead, Amir and Forster36)} .

Issues with use of expressed colostrum in hospital

Ideally, colostrum collected during pregnancy would be kept safely frozen at home and only brought in (frozen) to be defrosted for use during the hospital stay if medically indicated, with parents supported through that process. However, there have been reports of parents feeling disheartened by the lack of support for the use of stored colostrum within the hospital setting^{(Reference Glavey and Fallon30,Reference Demirci, Glasser and Fichner35,Reference Moorhead, Amir and Forster36)} . In these cases, parents would sometimes leave the hospital feeling like their ACE efforts had been in vain^{(Reference Moorhead, Amir and Forster36)}. Improved hospital staff education is needed to help avoid these experiences that may negatively influence breastfeeding practices when the families return home^{(Reference Demirci, Glasser and Fichner35)}. Appropriate on-ward freezer and fridge storage, and the ability to keep colostrum frozen in transit, is also important to allow antenatally expressed colostrum to be safely stored and accessible.

Anecdotally, we hear from midwives that postnatal women sometimes feel inclined to use their supply of antenatal colostrum during their hospital stay, even if not medically indicated. For example, a mother may ask for the collected colostrum to be fed to her baby while she catches up on sleep. This presents a potential disruption to the normal regular feeding pattern in the early days, which could have flow on effects to milk supply. It is therefore important to encourage breastfeeding to still be established as normal in the early days after birth. Ideally, antenatal colostrum would be kept frozen at home until needed or kept frozen at hospitals, so parents could take it home if not used during the hospital stay. Breastmilk including colostrum keeps for 3 months in freezer section of refrigerator with separate door (–18°C), and 6–12 months in deep freeze/chest freezer that is not opened frequently (–20°C)⁽⁴³⁾.

Colostrum and breastmilk are also useful for other areas besides feeding. Topical application of colostrum/breastmilk has been shown to be an effective treatment for nipple damage, lesions, nappy rash and conjunctivitis^{(Reference Amiri-Farahani, Sharifi-Heris and Mojab44,Reference Farahani, Ghobadzadeh and Yousefi45)} . Interestingly, bovine colostrum has even been recommended for inclusion in therapeutic protocols of cancer patients^{(Reference Artym and Zimecki46)}. Knowing that colostrum has other useful applications may ease any pressure that mothers may feel in not wanting to waste their expressed colostrum. (The lead author has had expired frozen breastmilk dried and turned into jewellery!) Women are also eager to donate excess colostrum to milk banks, to help other babies that may need it^{(Reference Casey, Mogg and Banks47)} – this already exists in some hospitals and could be promoted more.

ACE and breastfeeding outcomes

ACE education has generally shown positive outcomes for mothers intending to breastfeed their infants in the short term. Increased initiation of breastfeeding during the hospital stay^{(Reference Chen, Washio and Liu25,Reference Connolly, Reinkowsky and Giglia26)} , and a decrease in formula use while in hospital^{(Reference Demirci, Glasser and Fichner35,Reference Juntereal and Spatz41,Reference Casey, Banks and Braniff48)} are consistently reported in studies investigating ACE. The Breastfeeding and Antepartum Breast Milk Expression (BABE) randomised control trial was a notable exception to this; this study used a breast pump for antenatal expressing rather than hand expressing and found no difference in formula use during initial hospital stay compared with no intervention^{(Reference Estafanous, Lee and Thomas29)}. One disadvantage of using a pump instead of hand expressing is that women do not practice the skill of hand expressing. This skill is useful when wanting to express a small amount of colostrum post-birth, for example, to help baby with attaching.

For longer-term breastfeeding outcomes, there is conflicting evidence as to the efficacy of using ACE. A systematic review reported an increase in breastfeeding rates for women with diabetes up to three months post-birth^{(Reference Ablett, Hopper and Maslin39)}; a separate review found studies often reported increases but these were not always statistically significant^{(Reference Glavey and Fallon30)}. In randomised controlled trials, there is also evidence that mothers receiving an ACE intervention had similar breastfeeding rates at three and six months when compared to mothers in control groups^{(Reference Demirci, Glasser and Himes28,Reference Estafanous, Lee and Thomas29,Reference Moorhead, Amir and Crawford32)} . At this stage, there is no good quality evidence to show that ACE significantly improves rates of long-term breastfeeding.

Does performing ACE affect postnatal colostrum?

If a pregnant mother is expressing colostrum in the weeks prior to birth, could this result in any detrimental changes to the colostrum once the baby is born? There is currently little known on whether extracting colostrum prior to birth results in a change to the composition of or duration of colostrum post-birth.

Waller and colleagues noted back in the 1940s that some women reported that colostrum flowed more freely after they had practised ACE for a few weeks compared to when they first started^{(Reference Waller, Aschaffenburg and Grant49)}. They wondered whether this could be due to a reduction in viscosity or a change in composition, and investigated both the viscosity and protein (via total nitrogen) composition of antenatal and early postnatal colostrum secretion. Samples were collected on several occasions during the last 8–10 weeks of pregnancy and daily for approximately the first-week post-birth. Samples were taken via either hand expression or pump, and sent for testing immediately or stored in a refrigerator prior to testing. Results, from 45 participants, showed a wide variation in viscosity and total nitrogen between individual mothers^{(Reference Waller, Aschaffenburg and Grant49)}. However, neither of these factors were systematically lower or higher as a result of performing ACE. A tendency was noted in many cases for all factors to reach their highest values at the time of birth, or shortly after. The peak in viscosity and protein content in the majority of cases was on the second or third day after birth, although sometimes as early as a few hours after delivery and sometimes as late as the fourth day. In addition, they did not find any evidence that the regular removal of antenatal colostrum shortens the period of time postnatally when colostrum is produced. However, the researchers were not able to recruit as many mothers as they were aiming for, due to a bomb destroying a large part of the hospital including their testing equipment during the war. This suggests the study could have potentially been underpowered.

Since the 1940s our knowledge of colostrum composition and the equipment available to test different aspects of composition has improved greatly. The Antenatal Colostrum Expression (ACE) Study (detailed in the following section) aims to evaluate this further.

Studies underway

The ACE Study is currently investigating the efficacy of ACE on breastfeeding outcomes in non-diabetic, low-risk, first-time mothers in Australia^{(Reference Cuffe, Giglia and Cooper50)}. It will analyse data from three groups of participants: those randomised to in-person ACE education or pre-recorded video ACE education, and a control group who received standard antenatal care and education. Our research team is also aiming to investigate differences in colostrum composition between mothers in the ACE and control groups. ACE research in the USA is also currently being undertaken by Jill Demirci and colleagues^{(Reference Demirci, Glasser and Himes28)} who are looking at a similar cohort of primiparas birthing parents but with a pre-pregnancy body mass index of at least 25. Recruitment for this randomised controlled trial is finished (n = 280) and the research team are currently finishing data collection.

Directions for future research

Many of the studies reviewed were conducted in relatively homogeneous groups of pregnant women with diabetes^{(Reference Moorhead, Amir and Crawford32,Reference Johnsen, Klingenberg and Brand34,Reference Moorhead, Amir and Forster36,Reference Casey, Mogg and Banks47,Reference Casey, Banks and Braniff48)} . The majority of studies have been on cohorts from Australia and the United States, so results may not be generalisable to other populations^{(Reference Forster, Moorhead and Jacobs23,Reference Juntereal and Spatz31,Reference Johnsen, Klingenberg and Brand34,Reference Demirci, Glasser and Fichner35,Reference Casey, Mogg and Banks47,Reference Casey, Banks and Braniff48)} . Further research in more diverse population groups such as minority ethnic groups, different socio-economic groups, those giving birth to babies with congenital defects, those at risk of preterm birth, and people from the LGBTQI+ community would be beneficial to discover whether ACE would be an appropriate intervention for women and other birthing people from these groups^{(Reference Chen, Washio and Liu25,Reference Davis, Glasser and Clemens27,Reference Glavey and Fallon30,Reference Moorhead, Amir and Crawford32,Reference Demirci, Glasser and Fichner35,Reference Moorhead, Amir and Forster36,Reference Foudil-Bey, Murphy and Dunn38,Reference Ablett, Hopper and Maslin39,Reference Juntereal and Spatz41)} .

Future research could also investigate health effects of using defrosted frozen antenatally expressed colostrum for the infant. Both the freezing and thawing process and the colostrum being expressed antenatally may potentially influence the quality of the colostrum. Further, infants are exposed to circadian cues through breastfeeding via hormones such as melatonin that transfer to milk via the mother’s plasma^{(Reference Caba-Flores, Ramos-Ligonio and Camacho-Morales51)}. If colostrum is frozen, how could that affect circadian rhythms in the newborn? In addition to hormones present, nutritional aspects like fat and amino acids can also differ in concentration during the daytime and at night^{(Reference Caba-Flores, Ramos-Ligonio and Camacho-Morales51)}. Comparison of antenatally expressed and fresh colostrum versus formula for newborn feeding would be interesting to investigate further, in terms of gut health, immune function, allergy and longer-term health.