Gestational diabetes and late milk: what antenatal expressing does and does not do

If you have gestational diabetes, or another type of diabetes in pregnancy, someone may have suggested you hand express colostrum in the last weeks before your baby is born. The reasoning sounds sensible. Get a little milk moving early, have some of your own milk on hand for the first feeds, and maybe your milk will come in sooner.
Does the research back that up? Partly. A large randomized trial tested it, and the answer to "sooner" is no. This post walks through what the trial did, what it found, what benefit it did show, and what it still cannot tell you.
First, what antenatal expressing is
Hand expressing means collecting colostrum by hand rather than with a pump. Colostrum is the thick, golden first milk your breasts make in pregnancy. In late pregnancy some people hand express a few drops at a time, freeze them, and bring them to the hospital. In the trial below, women were advised to hand express twice a day, for 10 minutes in total, from 36 weeks until birth.
Why the timing of milk coming in matters
After birth, milk production is low for the first day or so. It then rises substantially, usually somewhere between 30 and 40 hours after birth. That shift is what people mean by their milk "coming in". When it happens after 72 hours, it is called delayed onset of lactation.
Diabetes in pregnancy is one of the things linked with a later start. A meta-analysis of eight studies, cited in the 2024 paper below, estimated that about 35% of women with gestational diabetes had delayed onset, with higher odds than women without diabetes (odds ratio 1.84, 95% confidence interval 1.34 to 2.52).
That matters because of what can happen in the meantime. Babies of mothers with diabetes are at higher risk of low blood sugar after birth. If milk is delayed, that can lead to formula supplementation, and formula supplementation is linked with shorter breastfeeding. So if expressing before birth really did bring milk in sooner, it would be a meaningful benefit for exactly the families who are told to try it.
What the trial did
The DAME trial (Diabetes and Antenatal Milk Expressing) ran from 2011 to 2015 at six sites in Victoria, Australia. It included 635 women with diabetes in pregnancy, gestational diabetes and pre-existing diabetes among them. Of those, 319 were advised to hand express twice a day, for 10 minutes in total, from 36 weeks until birth. The other 316 had standard care.
All the women were low risk, expecting one baby, and planning to breastfeed. Women with a previous classical cesarean scar, more than one cesarean, any bleeding, or a suspected big baby were not included.
The 2024 paper below asked a different question from the original trial report. It asked whether milk came in later for women with diabetes, and whether being advised to express changed that. To compare, the researchers also followed 192 women without diabetes who gave birth at one Melbourne hospital in 2014.
What they found
Delayed onset, meaning milk coming in at 72 hours or later:
Standard care, diabetes in pregnancy: 58.6% (143 of 244)
Antenatal expressing, diabetes in pregnancy: 55.8% (148 of 265)
No diabetes in pregnancy: 46.9% (90 of 192)
After adjusting for parity, mode of birth, BMI and gestation, milk came in later for women with diabetes than for women without it. For the standard care group that difference was clear. For the expressing group it was smaller and could be chance. Between the two diabetes groups the difference was 4.0 percentage points (95% confidence interval minus 4.5 to 12.5).
The authors put it plainly: their findings do not provide evidence that antenatal expressing hastens the onset of lactation in women with diabetes in pregnancy.
So what was the benefit?
There was one. It was not about milk arriving sooner.
In the trial's own reported results, summarized in the 2024 paper, 69% of babies in the expressing group received only breast milk in their first 24 hours, compared with 60% in the standard care group (adjusted relative risk 1.15, 95% confidence interval 1.02 to 1.28). The authors describe this as moderate evidence of an association.
The same trial found no difference between the groups in how many babies were admitted to a neonatal intensive or special care unit.
So the benefit the trial actually showed is a better chance of your baby getting only your milk through the first day, not your milk coming in earlier. The study cannot tell us why that difference appeared. One reasonable explanation is not that milk came sooner, but that families had expressed colostrum on hand to feed in those first hours. The trial does not test that explanation, so it stays an explanation rather than a finding.
Does earlier milk matter at all?
Yes, early feeding matters. A 2024 systematic review of 19 studies, with 175,392 babies, looked at whether feeding babies early, usually within the first hour, prevents low blood sugar. It found that early feeding may be associated with less neonatal low blood sugar (four cohort studies, 744 infants, odds ratio 0.19, 95% confidence interval 0.10 to 0.35). The authors are careful with that result. They describe the evidence as very uncertain, and they say early feeding should keep being recommended for its many other benefits.
Put the two together and you get a clear, limited picture. Getting milk into a baby early is worth something, especially for a baby at risk of low blood sugar. But the trial did not find that expressing before birth gets milk in sooner, so earlier milk is not the reason to express.
What this research cannot tell you
The women in the trial were low risk, expecting one baby, and planning to breastfeed. Women with more complicated pregnancies were not included.
"Milk coming in" was based on each mother's own report of when her breasts felt noticeably fuller, not a measurement. That measure is well validated, but it is not exact, with sensitivity of 71% and specificity of 79% against weighing feeds.
The comparison group without diabetes was smaller, recruited separately, and came from one hospital.
The trial ran between 2011 and 2015. The women were mostly partnered, degree educated, born in Australia and English speaking, so the results may not describe everyone.
Earlier claims that expressing speeds up full lactation came from small observational studies and two earlier trials that measured when milk came in poorly, which is part of why this trial was run.
What this means for you
Late milk is more common when you have diabetes in pregnancy. It is common, it is not your fault, and it is something your care team can plan around.
If you have gestational or other diabetes in pregnancy, whether to express before birth is a conversation for your obstetric or midwifery team. They know your pregnancy, your history and your baby. Questions worth asking: what is our plan for the first 24 hours, is expressing advised for me, and who will help me with feeding if my milk is slow to come in?
A lactation consultant can help you plan for those first days: how to hand express, how to feed what you collect, and what to expect. You can read about our lactation consultations at milkwelllactation.com/lactation-consultations.
Sources
Moorhead AM, Forster DA, Donath S, De Bortoli J, Amir LH. Does Antenatal Expressing Affect Onset of Lactogenesis for Women With Diabetes? Results From a Randomised Controlled Trial and Cohort Study. Australian and New Zealand Journal of Obstetrics and Gynaecology. 2024;65(4):441 to 448. https://pubmed.ncbi.nlm.nih.gov/39740238/ (free full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC12668880/)
Roberts LF, Harding JE, Crowther CA, Watson E, Wang Z, Lin L. Early Feeding for the Prevention of Neonatal Hypoglycaemia: A Systematic Review and Meta-Analysis. Neonatology. 2024;121(2):141 to 156. https://pubmed.ncbi.nlm.nih.gov/38194933/
This article is general education, not individualized medical advice. Talk with your own clinician or lactation consultant about your situation.

