The purpose of this study is to determine whether early start of either breastfeeding or breast milk expression may positively affect milk production later in lactation.

120 breastfeeding mothers will be recruited and randomized into 4 groups (2 groups with preterm deliveries) and 2 groups with term deliveries. One group will be given a breast pump to pump early and regularly. the second group will be encouraged to breastfeed in response to baby's cues. Breast milk samples will be collected at the end of a week.

The initiation of lactation in women depends on the growth of the glandular tissue of the breast (mammogenesis) and the differentiation of the mammary secretory epithelial cells (lactocytes) during pregnancy (lactogenesis I). After birth, the continuous nourishment of the fetus from the mother via the umbilical cord is replaced by the equally important but intermittent nourishment from the mother's breast. The onset of copious milk production (lactogenesis II) is closely coupled to the factors controlling the birth process. In mothers who have delivered pre-term, mammogenesis and lactogenesis I may be truncated and close consideration must be given to all physiological aspects of this complex process when assisting preterm mothers to establish successful lactation. In Singapore, the problem is not restricted to mothers of preterm babies. In a nation-wide survey by the Health Promotion Board in 2001, inability to produce enough breast milk was stated as the reason for not breastfeeding by 53% of mothers two months after birth.

The hypothesis of this study is that early, regular and complete emptying of the breasts can hasten and improve the establishment of lactogenesis II after preterm and term deliveries.

It aims to compare the effect of early and regular breast milk expression on the success and timing of establishment of lactogenesis II compared with routine breastfeeding practice in mothers after preterm and term deliveries.

The second aim is to study the effect of early and regular breast milk expression on blood prolactin and progesterone levels in women after preterm and term deliveries.

The final aim is to study the volume of breast milk produced and fed to babies after preterm and term deliveries.

80 women with preterm deliveries (28 weeks to 36 weeks gestation) and 80 women with term deliveries (after 37 weeks gestation) will be recruited and randomised with minimization by the stratification factors of previous live births and breastfeeding experience.

Groups A or C mothers will be given a breast pump to start expressing within one hour after delivery and thereafter every two to three hours besides breastfeeding their babies directly.

Groups B or D mothers will breastfeed according to routine hospital protocol. Standard postnatal nursing care will be followed. 7 days of breast milk samples (of 1ml in the morning and one ml in the evening )will be collected at the end of the 8th day. The details of the delivery, birth weight, and neonatal condition will be collected prospectively at birth. Mothers will be given a neonatal feeding diary to record the frequency and amount of breast milk expressed, the frequency and duration of direct breast feeds, and any other fluids fed to their babies in the first week. They will be asked to record their own perception of when their breast milk flow was established. The babies will be weighed on Day 7.

The primary outcomes are:

1. Successful initiation of lactation as determined by lactogenesis II markers
2. Successful initiation of lactation as determined by maternal perception
3. timing of successful establishment of lactogenesis II.

All statistical analyses will be conducted by SPSS 11.0. The differences in successful initiation of lactation as determined by lactogenesis II markers and the successful initiation of lactation as determined by maternal perception between the 2 groups will be assessed by Chi-Square or Fisher Exact tests. A logistic regression analysis will be performed to adjust for relevant covariates. 2 sample t-tests will be performed to determine the differences between the 2 groups in the timing of successful establishment of lactogenesis II, volumes of colostrums and milk expressed and fed and weight change in baby 7 days after birth when normality assumptions are satisfied. Otherwise the Mann Whitney U test will be applied. A multiple regression analysis will be carried out to take into account relevant covariates. Statistical significance is set at p <0.05.

Outcomes and advantages of the project:

1. If early and regular expression of breast milk can speed up or improve the chance of successful initiation of lactogenesis II, it can be a useful aid to women planning to breastfeed their babies, especially if they deliver prematurely.
2. This study will help us to determine the normal timing of establishment of lactogenesis II in Singaporean women after term and preterm delivery. Validated and precise breast milk markers of lactogenesis II will be used and compared with simple maternal perception. This knowledge will help dispel the many myths regarding Asian women and breastfeeding.
3. Currently, the actual physiological nutritional requirements of a breastfed newborn is not known. In our study, the volume of breast milk produced daily and fed to the baby in the first week will be measured in after both term and preterm delivery. This can be correlated to the weight change of the newborns to determine how much breast milk a neonate normally requires.

This will be the first of a series of collaborative projects on the various aspects of the science of human lactation between the Lactation Research Laboratory at the University of Western Australia and the Early Childhood Programming Group of the National University of Singapore

Inclusion Criteria:

• 60 women with preterm pregnancies who want to breastfeed their babies will be recruited and 60 women with term pregnancies who want to breastfeed their babies will be recruited and randomized.

Exclusion Criteria:

• Those with maternal diabetes and hyperprolactinaemia, as well as fetal congenital anomalies and multiple pregnancies will be excluded.