Breast milk in a sticky eye: what the evidence says

Key takeaways

No published study has tested breast milk as a treatment for an eye that is already infected. The researchers who ran the laboratory study people quote for it wrote that its antibacterial activity has not been adequately studied.
What has been studied is prevention in babies who were not yet infected, and a blocked tear duct. Those are small studies, and the reviews collecting them describe their own evidence base as limited and mixed.
A baby under 30 days old with a red, swollen or discharging eye needs to be seen the same day, whatever anyone in the house suggests trying first. The risk is the delay, not the drop.
Eye infection in the first month is a different disease from pink eye in a toddler. It is caught during birth, and untreated gonococcal infection can perforate a cornea within 24 to 48 hours of symptoms starting.
For cleaning, boil water and let it cool, then wipe with a clean cotton wool pad, one fresh piece for each eye. Nothing needs to go into the eye to clean the outside of it.

In a baby's first month, the advice about the eyes usually arrives through somebody else. A grandmother, an aunt, the confinement lady running the house that month. And it is nearly always the same advice, offered with real kindness. Express a little susu ibu, put a drop in the eye, it will be clear by morning.

I am not going to laugh at that. Ridicule changes nobody's mind, and there is more behind this piece of advice than most doctors admit. So here is what the research shows, what it does not, and the one fact about a newborn's eye that decides the question.

Does putting breast milk in the eye actually work?

Not for an eye that is already infected. Nothing published tests breast milk as a treatment for a newborn eye infection. The researchers who ran the laboratory study people quote for it wrote, in the opening of their own paper, that the antibacterial activity of topically applied breast milk has not been adequately studied. They wrote that in 2013 and it has not changed.

What has been studied is different: whether milk put into the eyes of newborns who are not yet infected means fewer become infected later, and whether it helps a watery eye that is blocked rather than infected. Both are useful questions. Neither is the one a mother is asking at two in the morning with a lid stuck shut.

Why does everyone say it, then?

Because it is one of the oldest pieces of medical advice we still repeat. Ocular use of human milk is documented in ancient Egyptian, Indian, Greek, Roman and Byzantine texts, and Galen recommended it specifically for conjunctivitis. The same paper notes it is still in circulation today and still appears in widely distributed parenting guidebooks.

So the grandmother in your kitchen is not making something up. She is passing on advice that was passed to her, and that a book somewhere probably printed. The reasoning is not silly either. Breast milk carries antibodies, and that is why the idea survives. It is not evidence that a drop of it clears an infection out of an eye.

What do the studies actually show?

Three small things, and none of them is treatment. The first is laboratory work. Milk donated by 23 women was tested against nine bacterial species on culture plates. The published review summarising that work records growth of three organisms being significantly inhibited, one of them the organism behind the newborn eye infection we most fear. A plate is not a baby's eye, and slowing growth in a dish is not curing a disease.

The second is prevention. In one trial, 269 preterm newborns whose eye swabs were negative to begin with were split three ways: two drops of colostrum in each eye, an antibiotic ointment, or nothing. Conjunctivitis was commonest in the group given nothing, less common with colostrum, least common with the antibiotic. An older case-control study of 565 newborns found conjunctivitis in 9.1 per cent of babies given milk topically against 25.6 per cent of controls.

The risk is not the milk. It is the days spent on it.

The third looked at the watery eye rather than the infected one. A single small study of babies with a blocked tear duct, 45 in one group and 20 in the other, found the watering settled sooner with milk than with antibiotic drops.

Read those honestly and they say something narrow. Both prevention studies were done in babies who were not yet infected, and one was not randomised. The reviews that gathered this literature together say plainly that their own evidence base is limited and mixed, and one states it did not perform a risk of bias assessment on the individual studies at all. If someone tells you breast milk is proven for eye infections, that is not what this research says.

Is my newborn's sticky eye an infection or a blocked tear duct?

Look at the white of the eye and at the lid. A blocked tear duct leaves the white white and calm. Tears pool, the lashes crust, you wipe and a few hours later it is back. More than 5 per cent of babies have it in one eye or both, and about 90 per cent clear in the first year with nothing done. What a blocked tear duct looks like and the massage that helps is a separate article, because it is a separate problem.

An infection is different. The white turns red or angry, the lids swell, the discharge is heavier and keeps coming back. Pink eye in an older child follows fairly predictable rules and often settles on its own. In the first month of life, though, that distinction is not one you can safely make at your kitchen table, and not one I would make over the phone either.

When is a gunky eye in a newborn an emergency?

In the first 30 days of life, a red, swollen or discharging eye is a same-day appointment. Not tomorrow, not after one more night of trying something. Today, at a klinik kesihatan or a hospital.

An eye infection in the first month is a different disease from pink eye in a toddler, and it has its own name: ophthalmia neonatorum, conjunctivitis of the newborn. It is caught during birth, by contact with a birth canal carrying infection, and it can be bacterial, chlamydial or viral. Timing hints at the cause. Chemical irritation shows up within the first 24 hours. Gonococcal infection appears between days two and five, chlamydial between days five and 14, herpes simplex after day seven.

Gonococcal infection is the one that frightens ophthalmologists. It can involve the cornea, the clear window at the front of the eye, and without prompt antibiotic treatment it can progress to frank corneal perforation within 24 to 48 hours of symptoms starting. A newborn is poorly equipped to fight it. The immune system is not finished, and tear flow is sluggish, so tears fail to dilute and wash organisms away the way an older child's do.

Which is why the professional guidance is not measured. An optometrist who sees a case is advised to refer every one immediately to the on-call ophthalmologist, with swabs taken and treatment started without delay. Treatment here is not a bottle of drops from a counter. It is given by mouth, by injection or into a vein as well as into the eye. The NHS puts it in one line: a baby less than 30 days old with red, sticky eyes needs an urgent appointment.

So what is the real risk of trying the milk first?

The risk is the delay, not the drop. No study I have found measured what expressed milk does inside a baby's eye, so I will not tell you it causes an infection and I will not tell you it is harmless. What I can measure is time. Three days of trying is three days of an untreated infection in an eye belonging to someone two weeks old. In a gonococcal infection, three days is the whole illness.

One practical point. Milk is not sterile in the sense a doctor means the word, which is why the standard instructions for expressing it start with washing your hands and a sterilised container. That is milk on its way to a baby's mouth, not onto the surface of an eye.

I have never been cross with a parent for trying it. What worries me is how long they kept trying before they came.

How do I clean the eye properly, then?

Boil the water, let it cool, and wipe with a clean cotton wool pad, one fresh piece for each eye. Air masak yang disejukkan, nothing else. Wipe from the inner corner outward, and take a new pad each time you go back. Nothing needs to go into the eye to clean the outside of it.

Cleaning is not treatment, and the difference matters. A blocked duct stays blocked until the membrane inside it opens, usually on its own. An infection that needs antibiotics will not be wiped away, however carefully you wipe. Cleaning keeps the lashes clear so you can see what the eye underneath is doing, which is the thing I want you watching. If the lid itself starts to swell rather than crust, a swollen eyelid has its own short list of urgent signs.

Everything above is one sentence in a longer coat. A baby in the first month of life with a red, swollen or discharging eye is seen the same day, whatever anybody in the house has suggested trying. Bring the Buku Rekod Kesihatan, the child health record book, if you have it. If the eye turns out to be a blocked duct you will have lost a morning, and the rest of the first year's eye checks carry on as normal. That is a trade I will take every time.

See a doctor the same day if…

Your baby is under a month old and the eye is red, sticky or discharging. At that age it is seen today, not watched.
The eyelids are swollen, or the white of the eye is red and angry rather than calm and clear.
The discharge is heavy, keeps coming back within hours, or the lids are gluing shut through the day and not just on waking.
Your baby seems unwell, feverish or unusually sleepy alongside the eye, or there are blisters or a rash on the eyelid or around the eye.
There is a painful red swelling on the skin beside the nose with yellow-green discharge. Treat that as an emergency.

Common questions

Does squirting breast milk in the eye actually work, or is it an old wives' tale?

It is neither, quite. It is old advice with a small amount of real research attached to it, none of which tested breast milk on an eye that was already infected. The studies that exist looked at preventing infection in newborns who did not have one yet, and at a blocked tear duct. So it is not folklore invented from nothing, and it is also not a treatment.

My mother-in-law says breast milk will clear it, is it safe to try?

Under a month old, my answer is no, and the reason is time rather than the milk itself. Nobody has measured what expressed milk does inside a baby's eye, so I cannot promise you it is harmless, but we know exactly what an untreated newborn eye infection does and how fast it moves. In an older baby with a watery eye that is otherwise calm, I would still put nothing in the eye, because nothing needs to go in to keep it clean.

My newborn's eye is gunky, is it an infection or a blocked tear duct?

The white of the eye usually tells you. A blocked duct leaves the white calm and clear, with watering and crusting as the main features, and it comes back a few hours after every wipe. An infection reddens the white and swells the lids. Under 30 days old, though, that call is made in a clinic and not at home.

When is a gunky eye an emergency?

In the first 30 days of life, any red, swollen or discharging eye is an urgent same-day appointment. At any age, a baby who seems unwell, an eye that is very red, or blisters on the lid should be seen the same day. Newborn eye infection is caught during birth and can involve the cornea, which is why nobody waits on it.

How do I clean the eye properly, cooled boiled water or saline?

Boiled water that has been left to cool is the standard advice and it is easy to get right at home. Use a clean cotton wool pad, one fresh piece for each eye, and wipe from the inner corner outward. Take a new pad every time you go back. Cleaning keeps the lashes clear so you can see what the eye is doing; it is not a treatment for what is underneath.

The eye is red and the lid is puffy and she is only two weeks old, what now?

Go today, to a klinik kesihatan or a hospital. Do not wait for a morning appointment slot to be convenient, and do not try anything in the eye first. Bring the Buku Rekod Kesihatan, the child health record book, if you have it, and say how old she is and when the discharge started, because the timing helps narrow the cause.

If milk is not the answer, will she need antibiotic eye drops?

That depends entirely on the cause, and it is not always drops. Newborn eye infection is often treated with antibiotics given by mouth, by injection or into a vein as well as into the eye, because the infection is not only on the surface. Swabs are usually taken first so the treatment matches the organism. Nothing about that is a shop-counter decision.

Does it matter that both eyes are affected rather than one?

Not as much as parents expect. Newborn conjunctivitis usually affects both eyes, though one may look worse than the other, so two sticky eyes do not make it less serious. A blocked tear duct can also happen in one eye or in both, because each eye drains through its own channel. Age and the state of the white of the eye matter far more than the count.

References

NHS · Conjunctivitis · www.nhs.uk
College of Optometrists · Clinical Management Guideline, Ophthalmia neonatorum · www.college-optometrists.org
StatPearls (NCBI Bookshelf) · Ophthalmia Neonatorum (Neonatal Conjunctivitis) · www.ncbi.nlm.nih.gov
Baynham et al. · Antibacterial effect of human milk for common causes of paediatric conjunctivitis, British Journal of Ophthalmology 2013 · bjo.bmj.com
Witkowska-Zimny & Kaminska-El-Hassan · Milk Therapy: Unexpected Uses for Human Breast Milk, Nutrients 2019 · pmc.ncbi.nlm.nih.gov
Ozdemir & Ozdemir (eds) · The Use of Human Milk for Therapeutic Purposes Other Than Nutrition, Turkish Archives of Pediatrics 2022 · pmc.ncbi.nlm.nih.gov
AAPOS · Nasolacrimal duct obstruction (blocked tear duct) · aapos.org
NHS · Expressing breast milk · www.nhs.uk

By Dr Chan Li Yen. General information only. It does not replace a consultation. If you are worried about your child's eyes or vision, please see an eye doctor.