Palliative cough: find the cause, then skip the codeine

Clinical Rounds, the GPAtlas podcast. Palliative Care. 18 minutes. Published 8 October 2026. Free to listen.

From our Ocean topic Palliative care - cough, reviewed September 2025.

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 it's cough in palliative care, and the news that codeine's come off the cough ladder. But the bigger point is what comes before any suppressant at all.

Sarah: Ray's seventy-four. He's got lung cancer with metastases, he's at home, and his wife Jean rings on Monday morning. He's been coughing all night, every night, and neither of them has slept for a week. It's dry, she says. Hacking. He's wrecked. Could he have more of the codeine linctus? And honestly, most of us would sign that repeat between patients, or go one stronger, and never lay eyes on him.

Ben: Three things today. Why codeine's off the cough ladder, and what replaces it. Why you look for the cause before you suppress anything. And how the same wet cough gets treated in opposite directions, depending on where the patient is in their illness.

Sarah: So, the news, because it's exactly what Jean was asking for. Codeine is no longer advised for cough. And my honest first reaction was, really? We've reached for codeine linctus for years. It always felt like the gentle, safe option. The one you could sign without thinking.

Ben: Two reasons. Codeine has to be converted to morphine in the body, and that conversion is variable and unpredictable. So you never quite know what the patient's getting. And the trials show it doesn't reliably suppress cough anyway. Unpredictable, and unreliable. That's not much of a case for it.

Sarah: But let's be careful what we take from it. It's advice about one drug. It doesn't tell me why Ray's coughing. It doesn't mean every cough in advanced illness now gets morphine. And it says nothing about a wet cough.

Ben: Agreed. The news itself is narrow. For a distressing dry cough, go from simple linctus to low-dose morphine, and leave codeine out. For Ray, that means reviewing his codeine linctus. It's been prescription-only since twenty twenty-four, so it'll be sitting on his repeats. If he still needs a suppressant once you've looked for a cause, it's low-dose morphine. And that word, if, is doing a lot of work.

Sarah: Codeine's off the cough ladder. Simple linctus first, then low-dose morphine, if it's still needed.

Ben: It matters, too. Around a quarter of people in their last year of life find their cough very distressing. And the usual mistake isn't the drug. The Scottish palliative care guidelines name it outright. Treating an exhausting cough in advanced illness as untreatable, and reaching straight for a suppressant. The cough's often driven by something specific and reversible.

Sarah: Which is exactly where Ray caught me out. I nearly did it over the phone. Instead I went round, sat on the end of the bed, and went through his tablets. He's on an angiotensin- converting enzyme inhibitor, an ACE inhibitor, for his blood pressure. Nobody had given it a thought.

Ben: That's a common, reversible cause. Stop it. Then the other usual suspects. Decompensated heart failure. A chest infection. Reflux, so gastro-oesophageal reflux disease. Bronchospasm. Each one has its own remedy, and each beats a blanket cough suppressant. So the suppressant comes after the detective work, not instead of it.

Sarah: And a cough's rarely on its own. Behind all that, look for the underlying cause. For Ray, it's the cancer and the metastases. But think about a malignant pleural effusion, heart failure or end-stage respiratory disease. And the treatment itself. Chemotherapy, radiotherapy-induced pneumonitis and aspiration can all do it.

Ben: Examine the chest and review the drug chart before you suppress anything.

Sarah: So what did I actually ask Ray? First, what kind of cough is it? Jean had said dry, but I asked him to cough for me anyway. And it was. Tight, tickly, and it went on and on.

Ben: Good, because the character of the cough, and whether he can still clear secretions, decide the whole plan. Dry, productive, barking, or that long, low bovine cough. Then triggers. Eating, talking, cold air, lying flat. Then what comes with it. Coughing up blood, so haemoptysis. Sputum, breathlessness, a runny nose or post-nasal drip.

Sarah: His trigger was lying flat. Every time he lay down, off it went. Then I asked what it was costing him. Sore ribs. He'd been sick twice. And very quietly, while Jean was making tea, he told me he'd been leaking urine when he coughed.

Ben: That's exactly why you ask. Muscle or rib pain, vomiting, fainting, urinary incontinence, broken sleep. Then the background. Asthma, chronic obstructive pulmonary disease, heart failure, reflux, smoking, previous chemotherapy or radiotherapy, and his swallow. And the Association for Palliative Medicine suggests one more question. What does he most fear about the cough?

Sarah: So I asked. Choking, he said. He's convinced he'll choke in his sleep. Then Jean said hers was different. She thinks the cough means the cancer's getting worse. Two different fears in the same room, and neither of them had said it out loud.

Ben: Ask what kind of cough, what it's costing them, and what they're frightened of.

Ben: And a fear like that is a reason to examine properly. Start with the voice. A hoarse voice with a prolonged, low, bovine cough suggests a left recurrent laryngeal nerve palsy, from hilar or mediastinal disease. Then the chest. Crackles and fever for infection. Dullness and reduced breath sounds for an effusion. Raised jugular venous pressure, peripheral oedema and crackles at both bases for heart failure.

Sarah: And the face and neck. Facial or conjunctival oedema, distended veins in the neck and across the chest wall, Pemberton's sign. Headache too. That's superior vena cava obstruction, and it needs urgent specialist advice. Ray had none of it. Normal voice, clear chest, no swelling. So I could tell him, honestly, that I hadn't found anything dangerous.

Ben: The other red flags. Massive or persistent haemoptysis could be a major airway or tumour bleed: urgent review, and consider crisis medication. Stridor or rapidly worsening breathlessness, think upper airway obstruction. Sudden pleuritic chest pain with breathlessness, think pulmonary embolism. And anything new that doesn't fit the known illness, reassess and get specialist palliative care advice. All judged against the agreed goals of care.

Sarah: Listen to the voice, look at the neck. Massive haemoptysis, stridor or a swollen face is urgent.

Sarah: Then tests. My registrar was with me and wanted a chest X-ray, bloods, the lot. And I get the instinct. You want to feel you've done something. But what would any of it have changed for Ray?

Ben: That's the right question. Clinical assessment and a review of the cause come first. Then every test is tailored to prognosis, likely benefit and what the patient wants. A chest X-ray only if the result would change management, so a suspected effusion, consolidation or progression. Sputum culture only if it's purulent, you suspect infection, and you intend to treat.

Sarah: Bloods, so a full blood count, urea and electrolytes and C-reactive protein, only if infection or a reversible metabolic cause would alter treatment. Computed tomography or bronchoscopy are specialist-initiated, and rarely appropriate in advanced disease, unless someone's chasing a specific reversible cause. Ray had a clear chest and an obvious suspect. So we did none of it.

Ben: Only test if the answer would change what you do, and the patient wants it.

Ben: So, treatment, and the first rung isn't a drug. Sit him upright. Lying flat on his back makes coughing less effective and raises the risk of aspiration. For a man frightened of choking, that's worth saying to him directly.

Sarah: We moved things round so he sleeps propped up on pillows. Jean's doing honey in warm water, and steam inhalations. None of it took long. But I'll admit part of me thought, that's granny's remedies, not medicine.

Ben: It's medicine. Moist inhalations and honey in warm water soothe the irritated throat. Then nebulised normal saline, two point five to five millilitres, up to four times a day. It helps either kind of cough, because it reduces airway irritation and loosens secretions. And if there's any wheeze, treat the bronchospasm with nebulised salbutamol.

Sarah: And the ACE inhibitor came off. So that's four changes before anyone's touched a suppressant. Then I went back a week later and asked what was bothering him most. Because that's what you treat, not the noise alone. For Ray, it was the sleep.

Ben: Sit them up, soothe the throat, nebulise saline, stop the ACE inhibitor. Then treat what bothers them most.

Sarah: And at that review, Ray's better. Less coughing in the day. But he's still awake half the night. Still dry, still distressing. So now he genuinely needs a suppressant. Take me up the ladder.

Ben: The first drug is simple linctus. It's a demulcent, five to ten millilitres three to four times a day, and there are sugar-free versions. Next, low-dose morphine. Morphine sulfate oral solution, the ten milligrams in five millilitres strength. Two point five to five milligrams every four hours, as required, for a distressing dry cough.

Sarah: Ray's frail, and his kidney function's been drifting down, so I started at the bottom of that range. I stopped the codeine linctus. And Jean went very quiet when I said morphine. She heard it as the end. So that conversation took a lot longer than the prescription.

Ben: Right, and that's the rule. Lower doses in frail, elderly or renally impaired patients, because the active metabolites accumulate. Always co-prescribe a stimulant laxative. And if someone's already on a regular opioid, say for pain, increase the background dose. Don't add a second opioid for the cough.

Sarah: And if it's still severe and uncontrolled after all that, I pick up the phone to specialist palliative care. Things like nebulised local anaesthetic, lidocaine for example, or a nebulised opioid, are specialist-initiated. That's not something I'd start on a home visit.

Ben: Simple linctus, then low-dose morphine. Lower in frailty or renal impairment, and always a laxative.

Sarah: Then, about a month later, it changed. Jean rang again. Now it's wet. He's bringing up thick green sputum and he's got a temperature. And this is where I nearly got it wrong. I'd spent weeks suppressing this cough, and my reflex was to give him more.

Ben: Two things first. New fever, purulent sputum or focal chest signs point to a chest infection that may be treatable, even in advanced illness. Antibiotics can be justified for comfort. Then the cough. If he can still cough effectively, the aim flips completely. You're not suppressing it any more. You're helping him clear it.

Sarah: Which feels completely backwards. I'd spent a month telling Jean we were calming it down. So the suppressant stops? Even though he's exhausted again at night?

Ben: Avoid routine antitussives, yes, because of the risk of sputum retention. A single night- time dose can be used, but only if the cough's exhausting and stopping him sleeping. Then help him clear it. Nebulised saline. Physiotherapy, to teach effective coughing and huffing. And carbocisteine if the sputum's thick and tenacious.

Sarah: He had antibiotics for the infection. Then carbocisteine, seven hundred and fifty milligrams three times a day, reducing once it's working. I checked his history first, because it's contraindicated in active peptic ulceration, and you stop it if there's any gastrointestinal bleeding. And I kept a single night-time dose of morphine, because the cough was still stopping him sleeping.

Ben: If they can still cough it up, help them clear it. Don't suppress it routinely.

Sarah: And then the last few days. Ray's too weak to cough now. There's a rattle with every breath, and Jean's sitting beside him listening to it. She asked me if he was drowning. And that sound is hard to sit with, even for us.

Ben: And this is the same wet chest going the opposite way. When someone's dying and too weak to cough, you stop trying to clear it. You dry the secretions and put comfort first. Reposition him, then an antimuscarinic. Hyoscine butylbromide is commonly first-line. Twenty milligrams subcutaneously as required, or sixty to one hundred and twenty milligrams over twenty-four hours by continuous subcutaneous infusion, so a syringe driver.

Sarah: I like that one, because it doesn't cross the blood-brain barrier. So little sedation, little confusion. That mattered here. Jean wanted him to still know she was there, holding his hand. What are the alternatives?

Ben: Glycopyrronium, two hundred micrograms subcutaneously, then nought point six to one point two milligrams over twenty-four hours by syringe driver. NICE also lists hyoscine hydrobromide or atropine, but those do cross the blood-brain barrier, so they can cause delirium or sedation. And start early. Secretions that are already there respond poorly.

Sarah: He couldn't swallow by then either, so his cough morphine went subcutaneous. One point two five to two point five milligrams as required, about half the oral dose. Regular mouth care, because both drugs dry the mouth. And I sat with Jean and explained the rattle. I think that helped her more than anything I prescribed.

Sarah: Dying and too weak to cough: reposition, hyoscine butylbromide, and explain the rattle to the family.

Ben: Right, exam corner. Pat is seventy-eight, frail, with lung cancer and reduced kidney function. She has a dry, exhausting cough that's wrecking her sleep. Her chest is clear and there's no wheeze. You've stopped her ACE inhibitor, she sleeps upright, and simple linctus hasn't been enough. What's the best next step? A, codeine linctus. B, low-dose oral morphine at the lower end, with a stimulant laxative. C, carbocisteine three times a day. D, nebulised salbutamol.

Ben: B. She has a distressing dry cough, the reversible causes are dealt with, and simple linctus isn't enough, so the next step is low-dose morphine. Lower dose because she's frail with renal impairment, and always a laxative. A is the tempting one, because it's what many of us learned. But codeine's no longer recommended: its conversion to morphine is unpredictable, and it doesn't reliably suppress cough. C is for thick sputum. D is for wheeze.

Sarah: So. Three things for Monday. One. Before you suppress a cough in advanced illness, examine the chest and review the drug chart, starting with any ACE inhibitor. Two. Codeine's off the ladder. Simple linctus, then low-dose morphine, with a laxative every time. And three. Ask whether they can still cough it up. If they can, help them clear it. If they're dying and too weak, dry the secretions and explain the rattle.

Sarah: One thing to reflect on, if you're logging this. The last time you saw someone with advanced illness and a cough, did you look for a reversible cause before you reached for a suppressant? The full transcript and the sources are on the episode page. Ben and I are natural AI voices, and what we've said comes from our reviewed Ocean topic. Check your local guidance before you act on it. See you next week.

More from Clinical Rounds

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