Description of the condition
Numerous health benefits to both the mother and baby can be ascribed to breastfeeding, in addition to the substantial cost savings it affords to families and health services (Renfrew 2012). The World Health Organization (WHO) recommends that infants be exclusively breastfed, whether directly from the breast or as expressed breast milk, for the first six months, that is, without any supplements, artificial formula or solid food (WHO 2011).
Infants born to mothers who have diabetes in pregnancy (gestational or pre-existing) are at increased risk of neonatal hypoglycaemia (low blood sugar) compared to other infants (Hanson 1993). This can be explained by their exposure to higher glucose levels in utero than usual, with subsequent increased insulin secretion. These infants may then need to adjust their insulin secretion to deal with postnatal glucose intake levels. It is for this reason that, in the first few days of life, many of these infants become hypoglycaemic and will require additional glucose, provided by donor human milk, artificial formula or via an intravenous infusion, as well as the intake from breastfeeding or breast milk expressed after birth. Because euglycaemia (normal levels of glucose in the blood) appears to be an important influence on the onset of lactogenesis II (the copious flow of milk 30 to 40 hours after giving birth), women with diabetes in pregnancy with hypoglycaemia or hyperglycaemia may be at increased risk of delaying this progression (Arthur 1994; Neubauer 1993). Thus, the infant who is already at increased risk of morbidity related to his/her mother’s diabetes, may also be exposed to artificial formula and separation from the mother if transferred to a nursery facility for intravenous fluid administration and glucose monitoring.
Further, avoidance of dietary exposure to some proteins found in cow's milk and the potential for a stronger immune system in exclusively breastfed infants may decrease the likelihood of these children subsequently developing B-cell autoimmunity and Type 1 diabetes (Ip 2007; Newburg 2005; Silverman 1995).
Description of the intervention
Antenatal breast milk expression has historically been proposed as a means of breast preparation (Chapman 2012a), although its popularity declined when the evidence emerged demonstrating no benefits in doing this (for example, Brown 1975). However, the practice has since been utilised as a means of building up a store of colostrum antenatally. The advantage of doing this is that, following birth, should additional nutrition be required, maternal colostrum can be given instead of artificial formula (Cox 2006). A survey of lactation consultants in Australia reported a growing awareness of antenatal breast milk expression, even when the practice was not promoted by the individual lactation consultants who responded to the survey (Chapman 2012b).
How the intervention might work
The storage of expressed colostrum to be given (if required) in addition to breast milk obtained directly from the breast or expressed after birth, may avert the need for artificial formula or intravenous fluid administration if correction of hypoglycaemia is required. Some clinical guidelines (e.g.NICE 2008) recommend close monitoring of the baby's blood sugar level in the postnatal period, with the mother and baby remaining together for care. Should the infant become hypoglycaemic (often defined as a true blood glucose (TBG) of < 2.6 mmol/L), a prescribed series of escalating interventions is followed, which may include separation of the baby from the mother through admission to a special or intensive care nursery if an additional feed of breast milk or formula does not result in euglycaemia within an hour, or by the time of the next feed (NETS 2009). Some hospitals mandate the infant's automatic admission to the special or intensive care nursery following birth, rather than mother and baby being cared for together, for example, for the infant of a woman with Type 1 diabetes, or an infant of a woman with gestational diabetes who required in excess of a specified number of units of insulin daily (e.g. Southern Health 2011). Moreover, there are substantial economic and social costs attributable to such admissions and to separation of the mother and her baby (Argus 2009; Figueiredo 2009). The limited expenses involved in educating women to express and the provision of sterile containers and freezer storage would be likely to be considerably less than the costs of specialised nursery admission and treatment.
Potential concerns arising from breast/nipple stimulation
Uterine contractions may result from the release of the hormone, oxytocin, that accompanies nipple stimulation (Christensson 1989). Therefore, the potential for this intervention to cause harm by bringing on labour early raises concern. Specifically, breast stimulation may be utilised as a means of inducing labour, as reported in a systematic review of six trials (719 women) comparing breast stimulation with no intervention in women from 37 weeks of gestation (Kavanagh 2005). The review reported a significant reduction in the proportion of women not in labour within 72 hours (62.7% versus 93.6%, risk ratio (RR) 0.67, 95% confidence interval (CI) 0.60 to 0.74). Although these findings were only significant in women who entered the study with a favourable cervix (that is, ready for labour), other randomised trials have demonstrated an improvement in the Bishop's score, which gauges cervical preparedness for labour (Damania 1992; Di Lieto 1989; Salmon 1986). To address this concern, Soltani 2012 reported a retrospective cohort study of 94 diabetic women. Infants of mothers who had expressed antenatally were more likely to be born a week earlier than infants whose mothers had not undertaken antenatal breast milk expression.
Further concerns include the potential for earlier birth to contribute to neonatal nursery admission and/or for hypoglycaemia to develop or persist despite being given the colostrum. The study by Soltani 2012 reported that more babies were admitted to the special care nursery in the group that expressed milk antenatally. Forster 2011 enrolled 43 women with diabetes in a prospective non-randomised study of antenatal breast milk expression twice a day for 10 minutes from 36 weeks' gestation. Outcomes for this group were compared with those from a retrospective audit of 89 women with diabetes who had not expressed during pregnancy. The study reported that five women experienced uterine tightening or Braxton Hicks contractions after expressing and did not continue this activity. Forty per cent of infants of women who had expressed milk received artificial formula within 24 hours of birth compared with 56% of the comparison group (RR 0.72, 95% CI 0.48 to 1.09). The finding of potentially increased rates of admission to the special care nursery in the expressing group were of concern even though they did not reach statistical significance (RR 1.79, 95% CI 0.94 to 3.33). The wide confidence interval suggests that more participants would be required to confirm or refute this concern.
Why it is important to do this review
Despite the concerns for the potential of earlier birth or neonatal nursery admissions for interventions to correct hypoglycaemia, antenatal breast milk expression and storage is emerging within clinical practice on the basis of its theoretical benefits to infants of women with diabetes in pregnancy (for example, Cox 2010; Ramsay Health Care 2011). The observational evidence that suggests the potential for an increased risk to the mother of premature labour and to the baby of premature birth and nursery admission following such practice (Forster 2011; Soltani 2012), needs to be followed through with a systematic review of randomised controlled trial evidence to determine the benefits and harms of antenatal breast milk expression, to then inform clinical practice. When it is determined that this practice is, or is not, beneficial to infants, there will be implications for promoting successful breastfeeding in the mother to reduce her risk of diabetes later in life and for the child's potential for developing diabetes.