Acute otitis media: the bulge, not the blush

Clinical Rounds, the GPAtlas podcast. ENT. 7 minutes. Published 25 September 2026. Free to listen.

Clinical source: our own Ocean article "Otitis Media (OM) - Acute", last reviewed 13 September 2025. Prompted by this week's paper, "Doctors frequently misdiagnose acute otitis media".

Transcript

Sarah: Hello, and welcome to GPAtlas Clinical Rounds. I am Sarah, a GP, and today Ben and I are going to be picking apart the clinical reality of diagnosing and managing acute otitis media.

Ben: Thanks, Sarah. Let us put a specific child in the room. I had a nineteen-month-old boy, Leo, in yesterday. His mum was clearly exhausted, and she was worried because he had been miserable for forty-eight hours. She described him as waking up crying every two hours, pulling at his left ear, and he had been off his food since a cold started earlier in the week. His temperature had peaked at thirty-eight point two that morning. When I went to look, he was a typical toddler: wriggling, crying, and definitely not interested in having an otoscope near his head. Mum was convinced he needed antibiotics because 'the doctor gave them last time,' but I needed to be sure what I was looking at. The temptation, especially when a child is screaming, is to just catch a quick glimpse of a red eardrum and sign the prescription for amoxicillin, but I spent a moment getting him settled on his mum’s lap to try and get a proper look at that tympanic membrane. It really hits home why this diagnosis is so frequently missed, or rather, over-diagnosed.

Sarah: That is the crux of it, isn't it? If we stop at 'the drum is red,' we are going to end up over-prescribing. You are looking for that specific evidence of a middle-ear effusion combined with acute inflammation. As we know, an upper respiratory tract infection can easily make the tympanic membrane appear hyperaemic because of the reflected light or the child’s distress, but that isn't acute otitis media. When you managed to get that look, Ben, what were you looking for specifically? You want that bulging, right? That is the hallmark. It’s that loss of the normal concave appearance, where the drum looks pushed out towards you, perhaps looking a bit cloudy or opaque. If you can’t see the landmarks, like the handle of the malleus, that is your clue. And you’re quite right about the exam, it is about ruling out those mimics. You have to check the tragus and the pinna; if you touch them and the child jumps, you are looking at otitis externa, not the middle ear. I often ask about systemic symptoms too, not just 'does he have a fever,' but 'is he drinking,' 'is he lethargic.' If I see a child who is toxic, that changes my entire threshold for investigation and management. You mentioned the international data showing that primary care doctors get this wrong nearly forty percent of the time. I suspect that is because we are often relying on the redness alone. We have to be disciplined. If the drum isn't bulging, or if it looks retracted, we need to consider if we are dealing with glue ear, otitis media with effusion, which is basically a vacuum in the middle ear post-infection, and that is a completely different kettle of fish. It’s painless, usually, and doesn't require antibiotics. You have to have that mental checklist: is it bulging? Is it opaque? Is the light reflex lost? If the answer is no, it is likely a viral ear or something else entirely.

Ben: The biggest shift for me on Monday is being much more rigid about that diagnostic threshold. I have stopped using redness as a proxy for infection. If I can't confirm a bulging drum, I am not diagnosing acute otitis media. That means for someone like Leo, if the view is too poor because he is crying, I don't guess. I talk to the parent. I explain that the ear might look red simply because he is upset, and that we need to be careful with antibiotics. This is the new advice: prioritize analgesia, use paracetamol or ibuprofen at the correct doses, so for a nineteen-month-old, that is weight-appropriate paracetamol, which we ensure is given regularly, not just 'as needed.' The evidence is clear that antibiotics make almost no difference to the pain at twenty-four hours, yet the potential for side effects remains. So, the management plan changes from 'just in case' prescribing to a delayed, or back-up, prescription. I write the script, but I explain that if he is no better in three days, or if he gets suddenly worse, then they have the safety net. But I also have to be mindful of the high-risk groups. If Leo had been under two and had it in both ears, or if he had discharge, otorrhoea, then I would consider an immediate antibiotic because those are the kids who actually benefit. But for a straightforward, unilateral case in an otherwise healthy toddler, we are shifting the focus back to the parent-as-carer. We explain that the 'earache' is a symptom, not a diagnosis, and that it is self-limiting. The shift is empowering the parent to manage the pain while we monitor the condition. I think we have spent too long worrying about the 'rare' complication, like mastoiditis, to the point where we are over-treating the ninety-nine percent of children who will recover on their own in three days. I am also being much clearer about the warning signs: I explicitly tell parents to look for that post-auricular swelling or a change in the child's neurological state. It turns the consultation into a teaching moment rather than a pharmacy-dispensing moment.

Sarah: The absolute classic pitfall is the 'red drum' trap, exactly as you said. I see it time and time again in the notes: 'tympanic membrane erythematous, diagnosis AOM.' It is such a common error because children with a URTI, or children who are crying, will almost always have a pink or red eardrum. It is a physiological response, not a pathology. If you treat that as AOM, you are just feeding the statistics for unnecessary antibiotic use. Another pitfall, I think, is neglecting the 'analgesia first' rule. We are so quick to reach for the prescription pad that we forget to emphasize that regular, max-dose paracetamol and ibuprofen are actually more effective for the pain in the first forty-eight hours than the amoxicillin would be. We need to be very specific: 'give it four times a day,' or 'keep up the schedule.' If parents are just giving one dose of paracetamol, the pain is going to come right back. And then, there is the risk of missing a complication because we haven't given the safety net properly. If you are going to suggest a delayed prescription, you must have a clear plan for what constitutes 'no improvement.' I usually give it three days. If they are not over the worst of it by then, or if they are getting worse, and I clarify what 'getting worse' looks like, then they use the script. A really common mistake is not warning about the perforation. If the child has been in agony and then suddenly there is a gush of discharge, the parents panic. I make a point of saying, 'If you see some yellow discharge, that is actually the eardrum letting the pressure out, and the pain will likely stop immediately. Don't rush to A and E, just bring them back for a review in six to eight weeks to ensure it has healed.' Knowing that this isn't a disaster, but rather a way for the body to resolve the pressure, saves so much anxiety for the parents.

Sarah: So, to recap: one, do not diagnose AOM on redness alone, look for the bulge. Two, analgesia is the main event, not the antibiotic. Three, reserve antibiotics for the high- risk groups or those who are truly failing to improve after three days.

Ben: Exactly. Sarah, thanks for talking this through. That is all for this week’s GPAtlas Clinical Rounds. We will see you next time.

More from Clinical Rounds

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