This is another big topic in Emergency that covers a host of different diseases and patient groups. When a patient can’t breathe, everyone else’s pulse jumps up too. My hope is that we can navigate some of the potholes in ED that show up time and time again and give you a structure to approach most common presentations.
Let’s start with the basics!
Pneumonia can be deceptively tricky to nail down sometimes. Patients rarely present with a classic lobar pneumonia on their chest x-ray and comorbidities often muddy the waters. And if they do the findings on xray are still subtle.
A fairly typical pneumonia x-ray - not obvious to the untrained eye but still evident.
Knipe H, Right lower lobe consolidation - pneumonia. Case study, Radiopaedia.org (Accessed on 13 Jun 2026) https://doi.org/10.53347/rID-31352
We should always consider pneumonia in a breathless patient given how common this is, and we can ask about fever, a productive cough and how their breathing has changed. Some risk factors like immunosuppression are also helpful to think about, especially for atypical pneumonia patterns. You may hear crackles on their chest though you may not. It can also be difficult to distinguish in someone with existing lung disease.
Ordering Inflammatory markers such as a white cell count and CRP are usually helpful, but some ED bosses don’t like taking CRPs and view them as a waste of resources. A chest x-ray is often definitive but some cases can be tough. Take the AP radiographs below. Both are from men in their 30s. One of them has pneumonia and one of them does not.
A normal CXR.
Hacking C, Normal CXR and lateral. Case study, Radiopaedia.org (Accessed on 13 Jun 2026) https://doi.org/10.53347/rID-40794
Left lower lobe pneumonia - quite subtle.
Sim K, Left lower lobe consolidation. Case study, Radiopaedia.org (Accessed on 13 Jun 2026) https://doi.org/10.53347/rID-33258
I find it really hard to see a pneumonia on this PA view. Without a convincing history I would have absolutely missed the pneumonia on a frontal x-ray.
When we look at the lateral the change becomes more obvious.
A normal lateral CXR.
Hacking C, Normal CXR and lateral. Case study, Radiopaedia.org (Accessed on 13 Jun 2026) https://doi.org/10.53347/rID-40794
Lower lobe pneumonia now clearer.
Sim K, Left lower lobe consolidation. Case study, Radiopaedia.org (Accessed on 13 Jun 2026) https://doi.org/10.53347/rID-33258
Lateral x-rays should become blacker (decreased opacification) as you move down the image. In the second film we can see an opacification at the base that was previously hidden behind the cardiac silhouette. Request and use lateral x-rays when you can!
To get better at looking at x-rays and CTs in general, look at every scan yourself to see what you can see and then look at your report. If you miss something go back to the scan and look again. This back-and-forth checking can help improve your image interpretation enormously.
When we’ve nailed a pneumonia, we can use a few scores to help us. CURB-65 can estimate pneumonia severity and SMART-COP can estimate the need for ICU admission for vasopressors or respiratory support. These scores, in turn, can help us pick which antibiotics are best, though in my experience most clinicians are happy to start with the broadest spectrum the local guidelines allow and narrow it down as patients improve.
Sometimes patients are obviously unwell. If you suspect your patient has pneumonia and their vital signs are deranged, I want you to ask if they have sepsis. Sepsis has a very clear set of criteria. The shortcut in my head is “are they febrile? Are they tachycardic? If both yes, then sepsis”. In my centre, patients who have suspected sepsis have a little bomb icon that appears next to their name on our patient tracker software Firstnet. Our aim is to administer IV antibiotics within one hour of this person’s arrival to Emergency. Let your boss know, take blood cultures if the nurses haven’t already and chart antibiotics. Ask for help if you aren’t sure how to do that. eTG is always quite helpful. After antibiotics are in, you can complete the rest of your history and examination.
Nursing home patients or those who are elderly with dementia or delirium are worth talking about separately. These patients will often have a “septic screen” to explain their worsening confusion or vague deterioration. They can’t give us a good history. Aspiration pneumonia is particularly common in this group. I find it exceptionally useful to use a CRP in this situation because even if you don’t find anything on your other tests, a CRP of over 200 gives you great information and will often help guide management.
Usually, anyone who requires supplemental oxygen needs admission to hospital, but this may not be the case for your nursing home patients. Most nursing homes will be able to care for patients who need only nasal-prong oxygen. Some services can treat pneumonia inside a nursing home with daily IV antibiotics (typically ceftriaxone). I have worked at several hospitals with a “Geriatric Flying Squad”, nursing specialists who can visit patients in their nursing home, administer their IV antibiotics and assess their progress. This disposition pathway is very helpful and can save a bed in hospital. If this is available where you work, it’s worth thinking about for every patient from a nursing home.
Wheeze can be a subtle auscultation finding or it can be pretty obvious in a very scared and sick resus patient. Ask for help early. If you find yourself with a very wheezy breathless patient, giving bronchodilators is rarely a bad idea. You are not going to hurt them by giving a trial of puffers – 12 puffs of salbutamol +/- 8 puffs of ipratropium. If they need oxygen at the same time, you can give them nebulised. If this improves their symptoms, we can give a full “burst” therapy - salbutamol 12 puffs every 20 minutes for a total of 3 doses. Sometimes a patient’s lungs are so tight that you can hardly hear anything at all and after giving salbutamol then you hear a wheeze. This means you’re on the right track.
If salbutamol does not help, then it’s time to pivot. Sometimes patients have a wheeze, but this is not the reason they are breathless. Sometimes patients have a wheeze with a different pathology like pulmonary oedema instead of crackles as you would expect (yes, I promise, this has happened). Call for your senior and try a different plan of attack.
The main diagnoses you are treating with bronchodilators are COPD and asthma. They do present differently and asthma is usually the scarier one when it’s severe.
COPD patients often have exacerbations and may need steroids as part of their treatment. With fevers and increased sputum production, they will also often get antibiotics.
If your patient requires more respiratory support than supplemental oxygen then they may require Non-Invasive Ventilation (NIV) started by your senior.
A common question a nurse will ask you in ED is can this patient tolerate saturations between 88-92%? What’s the easiest way to answer this question?
You could of course look at their documentation to see what’s previously been done but if this isn’t available then a blood gas is quite helpful. Respiratory physicians love an ABG but for us a simple VBG is usually just fine and will answer our question. Look at the pH, CO2 and HCO3. Is the CO2 high? Is the pH normal? The patient is retaining CO2 and is compensating. Is the HCO3 elevated? This is a chronic compensated process. If the CO2 is high with an abnormal pH or a normal HCO3 then this is acute and you should think twice before allowing them to have low sats.
Sometimes shortness of breath is due to pulmonary oedema, a cardiac issue, instead of a lung pathology. You know what to look for – crackles on auscultation, worsening peripheral oedema, orthopnoea. The problem is that it’s rarely so clear cut in practice.
If pulmonary oedema is obvious then get a weight for the patient and keep daily weights going while the patient is in hospital. Start on some frusemide IV – a good rule of thumb is to take their usual oral dose and convert it to IV. This is roughly twice as potent as an oral regime. If they aren’t on any frusemide, then 40mg IV a day is not an unreasonable place to start. If they’re very frail and comorbid you might chart 20mg IV instead.
Patients with severe pulmonary oedema can become acutely hypoxic, show severe work of breathing and develop hypertension with diffuse rales on auscultation. This is a medical emergency requiring resus and your senior at the bedside and may need escalation with NIV and glyceryl trinitrate in an IV infusion. You can learn more here out of interest.
Below is a typical example of pulmonary oedema on a chest xray.
You might be able to see a "fluffy" appearance of the lungs with increased streaky opacifications in the upper lobes, little white lines projecting outwards. There is also a hazy white appearance around the hilar regions.
Gaillard F, Acute pulmonary oedema. Case study, Radiopaedia.org (Accessed on 13 Jun 2026) https://doi.org/10.53347/rID-15434
Please see my blog on chest pain.
Frustratingly, many cases fall somewhere between asthma, COPD, pulmonary oedema and a viral or bacterial pneumonia. Some people have a mix of all of these things. It’s not always clear from a good examination, bloods and a chest x-ray what people have. One last effort to help distinguish pulmonary oedema from other pathologies is to do a bedside lung ultrasound with the help of an emergency consultant or registrar to look for indicators of fluid overload.
In practice, you’re going to have to make multiple phone calls. Your senior can help coach you and direct you to the team they feel is most appropriate. If you’re getting lots of push back my approach is to ask the teams to discuss with each other and then let me know who will take the patient. If things are taking too long just let your boss know so that they can move things forward.