Breastfeeding problems: the instinct that makes it worse

Clinical Rounds, the GPAtlas podcast. Pregnancy & Antenatal Care. 18 minutes. Published 31 August 2026. Free to listen.

Clinical source: our own Ocean article "Breastfeeding Problems", last reviewed 4 August 2026.

Transcript

Sarah: Hello, and welcome to GPAtlas Clinical Rounds. I'm Sarah, and this is the podcast where we take one thing you'll genuinely see this week and work out what to do with it.

Ben: And I'm Ben. Today, breastfeeding problems, and an instinct that almost every one of us has had, which turns out to be backwards.

Sarah: So. A twenty-nine year old first-time mother brings her three-week-old. Her nipples are cracked and painful, she's feeding roughly two-hourly, and she says she has no milk. She's tearful. She says she feels like a failure, and she's thinking of switching to formula. And when I check, the baby has regained birth weight and is having six wet nappies a day.

Ben: Three things. The single most useful thing you can do in that appointment, which isn't a prescription. The instinct to rest the breast, and why it makes everything worse. And the three or four prescribing details that have changed, including one licensing change most people haven't caught.

Sarah: So the first thing, and our article is unequivocal about it: the single most useful step is to observe a feed. Not take a history about the feed. Watch one. And I know exactly how that lands in a ten-minute appointment, because a feed is not a ten-minute event, and there's a waiting room.

Ben: But the reason it's first is what it explains. Our article says most nipple pain and most low supply concerns trace back to poor attachment, which can be seen and corrected at the bedside before anything is prescribed or stopped. So the two things this mother has come in with, pain and no milk, are both most likely one problem, and it's a problem you can only see.

Sarah: And there's a second thing watching a feed does, which isn't clinical. She has just told me she feels like a failure. If I take a history and write a prescription, I've responded to her words. If I sit and watch her feed her baby and then show her one adjustment that stops it hurting, I've responded to the actual problem, and I've done it without either of us having to have a conversation about whether she's failing.

Ben: Watch a feed. Both of her problems are probably the same problem.

Sarah: Now the instinct, which is the heart of this episode. Everything hurts, so the natural advice, and I have given it, is to take the pressure off. Give it a rest. Skip a feed. Let it heal. And our article says that is almost always the wrong move, in those words: stopping or spacing out feeds usually makes things worse, not better.

Ben: Because the mechanism runs the other way. Most mastitis is milk stasis rather than infection, so the treatment is effective milk removal plus analgesia. Resting the breast increases the stasis, which is the thing causing the problem. It's one of those situations where the compassionate-sounding advice is the harmful one.

Sarah: So for engorgement and for mastitis, the instruction is: do not stop feeding, and hand express if it's too painful to latch. Keep feeding or expressing from the affected side to drain the breast. Which is a hard thing to ask of somebody in pain, and needs saying with some warmth, because you are asking her to do more of the thing that hurts.

Ben: And two simple physical measures that our article gives, which are easy to get the wrong way round. Warmth before feeds, because it aids flow. Cold after feeds, because it reduces swelling. And reverse-pressure softening if the areola is hard. Warm before, cold after.

Sarah: Do not rest the breast. It is stasis, and rest makes stasis worse.

Ben: So when does mastitis actually need an antibiotic? Conservative measures first, effective milk removal plus analgesia. Then flucloxacillin five hundred milligrams four times a day for ten to fourteen days, and there are three triggers: if she's systemically unwell, if there's an infected nipple fissure, or if there's no improvement after about twenty-four hours of effective milk removal.

Sarah: That last one is the useful one, because it gives you a clean review point. It isn't watch and wait indefinitely, and it isn't antibiotics on day one for everybody. It's twenty-four hours of doing the milk removal properly, and if that hasn't shifted it, you treat. And clarithromycin or erythromycin if she's penicillin allergic.

Ben: With two escalation points. Hospital assessment for severe systemic sepsis, or no response to antibiotics after forty-eight hours. And a suspected abscess, so a fluctuant, red, painful lump, goes to surgery or the breast clinic.

Sarah: Twenty-four hours of proper drainage. Then flucloxacillin.

Sarah: Thrush, where the single most important point is that it isn't one patient. Our article says to treat mother and baby together to prevent ping-pong reinfection. Treat one, and the other reinfects them, and you go round again while she loses confidence with every cycle.

Ben: Mother gets miconazole two per cent cream to the nipples after each feed. And for the baby there's a licensing change that a lot of people have not caught: miconazole oral gel four times a day is now unlicensed under four months, because of choking risk. If you use it, apply only a small amount to the affected areas, away from the back of the throat. Nystatin suspension is the licensed alternative in young infants.

Sarah: And for a three-week-old, which is our patient, that means nystatin is the straightforward choice. This is exactly the kind of change that doesn't reach everybody, because it's a familiar drug in a familiar situation, and the prescription writes itself from memory.

Ben: Treat them both. And under four months, miconazole gel is unlicensed.

Sarah: Analgesia, which she needs and which people are too cautious about. Paracetamol and ibuprofen are both compatible with breastfeeding and are first-line for nipple pain or mastitis. So she can have proper pain relief, and being frightened to give her any is its own harm, because pain is part of why feeding fails.

Ben: What to avoid is codeine, because of variable metabolism and the risk of neonatal opioid toxicity, which is MHRA advice, and aspirin. And the general rule our article gives is a good habit for everything else: check every drug against a breastfeeding-specific source before prescribing.

Sarah: Paracetamol and ibuprofen, yes. Codeine, no.

Ben: Now the baby, because so far this has all been about the mother, and there are red flags that shift this from a feeding problem to a paediatric one. Faltering growth, so a fall across centile spaces, or weight loss of ten per cent or more of birth weight in the first week. Dehydration, so a sunken fontanelle, dry mucous membranes, reduced urine output. And prolonged jaundice, beyond fourteen days at term or twenty-one days if preterm.

Sarah: And the practical version of that in the room is what our article calls verifying milk transfer: check the wet and dirty nappies, and check the weight. Which is exactly why our mother's numbers matter so much. She has regained birth weight and she's having six wet nappies a day. Objectively, the feeding is working. Her experience of it is that she has no milk.

Ben: And that gap between the objective picture and her experience is the actual clinical problem. Reassurance that ignores her experience won't land, and agreeing that she has no milk isn't true. The honest position is that the milk is clearly going in, and it is hurting, and the pain is the thing we're going to fix.

Sarah: Weight back, six wet nappies. The milk is going in.

Sarah: And if the pain persists despite a good latch, our article says reconsider three things: tongue-tie, thrush, or nipple vasospasm, which is Raynaud's of the nipple, where the nipple blanches and then throbs after a feed. That last one is worth knowing because it's easily missed and the description is distinctive once you've heard it.

Ben: And there are referral routes rather than dead ends. A breastfeeding specialist or lactation consultant for persistent positioning or attachment problems. A tongue-tie service where ankyloglossia is interfering with feeding. And peer support, the National Breastfeeding Helpline, La Leche League, or local groups.

Sarah: And the referral criterion that people don't think of as one: maternal mental health, so significant distress or depression linked to feeding, is listed as a reason to escalate in its own right. She is tearful and calling herself a failure at three weeks postpartum. That is a finding, not a mood.

Ben: Our article gives a question for it, which I like because it does three jobs at once. How are these challenges affecting your mood, your confidence, and your bond with your baby? That asks about depression without making it an interrogation, and it treats the feeding and the mood as one conversation, which is how she is experiencing it.

Sarah: Tearful at three weeks is a finding. Ask the mood question.

Sarah: And the last thing, which is what to do if she wants to stop. Because she's said she's thinking of formula, and there's a real risk of getting this wrong in either direction. Our article's framing is careful and I'd read it almost as written: address the fixable problem, almost certainly a shallow latch, and offer practical help, while genuinely respecting her autonomy if she chooses to stop, and without missing early postnatal depression behind the feeding distress.

Ben: And it spells out what good looks like: observe a feed, validate her experience, treat the latch and the pain, screen her mood sensitively, safety-net the baby's weight and output, and signpost peer support, without pressuring her in either direction. That last clause is the one to hold. Not pressuring her to continue, and equally not agreeing too quickly that she should stop because it's easier for everyone.

Sarah: And the sentence our article offers her is the one I'd want to end on, because it separates the pain from the failure. Pain usually means the latch needs fixing, not that you have to stop. That reframes the whole thing. It isn't her body failing. It's a mechanical problem with a fix, and if she still chooses to stop after that, she's choosing rather than surrendering.

Ben: Fix the latch. Then it is her choice, not her defeat.

Sarah: And I want to say something about the appointment itself, because everything we've described assumes a kind of consultation that a ten-minute slot doesn't naturally allow, and pretending otherwise is unhelpful.

Sarah: You cannot watch a full feed in ten minutes. But you can watch a latch, which is the first thirty seconds, and that's where the information is. You're looking at how wide the mouth opens, how much areola goes in, whether the chin is in contact, whether she's in pain from the moment it starts. That's a minute of watching and it answers most of the question.

Ben: And knowing who else can do it matters as much as doing it yourself. Our article's referral routes include a breastfeeding specialist or lactation consultant for persistent positioning and attachment problems, and a tongue-tie service where ankyloglossia is interfering with feeding. Those exist precisely because this needs more time than we have.

Sarah: What we shouldn't do is let the shortage of time turn into a prescription. The instinct when you can't watch a feed is to treat something, and the two things most likely to get treated are thrush that isn't there, and pain with an analgesic that doesn't address the cause. Both feel like action and neither fixes the latch.

Ben: You cannot watch a whole feed. You can watch a latch.

Sarah: And one last thought about the mood question, because I've learned to ask it earlier than feels natural. At three weeks postpartum, with cracked nipples and two-hourly feeding, almost anybody would be tearful, and that's exactly why it's easy to attribute all of it to exhaustion and move on.

Sarah: But our article lists significant distress or depression linked to feeding as a reason to escalate in its own right, and it gives the question that opens it: how are these challenges affecting your mood, your confidence, and your bond with your baby. Asking that while you're still on the feeding problem, rather than as a separate agenda item at the end, gets a much more honest answer.

Ben: Ask the mood question inside the feeding conversation, not after it.

Ben: Exam corner. A twenty-nine year old woman is three weeks postpartum with a painful, red, wedge-shaped area in one breast. She is apyrexial and systemically well, and there is no nipple fissure. She has been avoiding feeding on that side because it hurts. What is the most appropriate initial management? A, advise resting the affected breast and review in forty-eight hours. B, effective milk removal from the affected side plus analgesia, and review at twenty-four hours. C, flucloxacillin five hundred milligrams four times daily immediately. Or D, refer urgently to the breast clinic.

Ben: It's B. Most mastitis is milk stasis rather than infection, so conservative management is effective milk removal plus analgesia first. A is the central error of this episode, since resting the breast worsens the stasis that is causing it, and she is already doing it. C would be right if she were systemically unwell, had an infected nipple fissure, or had failed about twenty-four hours of effective milk removal, and she has none of those yet. D is for a suspected abscess, a fluctuant lump, or failure to respond to antibiotics at forty-eight hours.

Sarah: So. Three things for Monday. One. Watch a feed, because most nipple pain and most low- supply worry is attachment, and you cannot see it from a history. Two. Never advise resting the breast, because the problem is stasis and rest makes it worse. And three. Under four months, miconazole oral gel is unlicensed, so nystatin is the one to reach for.

Sarah: One thing to reflect on, if you're logging this. Think about whether your appointments make it possible to watch a feed at all, and if they don't, who in your practice can. The full transcript and the references are on the episode page. Ben and I are synthetic voices. The medicine isn't. See you next week.

More from Clinical Rounds

Every episode of Clinical Rounds, free to listen with a full transcript.